Describe your change here

This commit is contained in:
Kovid Goyal 2009-01-06 12:40:23 -08:00
commit 269c330512
344 changed files with 44476 additions and 15224 deletions

View File

@ -5,9 +5,5 @@
<pydev_property name="org.python.pydev.PYTHON_PROJECT_VERSION">python 2.5</pydev_property>
<pydev_pathproperty name="org.python.pydev.PROJECT_SOURCE_PATH">
<path>/calibre/src</path>
<path>/calibre/devices</path>
<path>/calibre/libprs500.devices.prs500</path>
<path>/calibre/prs500</path>
<path>/calibre/gui2</path>
</pydev_pathproperty>
</pydev_project>

View File

@ -0,0 +1,244 @@
Changes from 4.0 to 4.0.1
1) Added support for Python 2.6. On Windows a manifest file is now required
because of the switch to using the new Microsoft C runtime.
2) Ensure that hooks are run for builtin modules.
Changes from 4.0b1 to 4.0
1) Added support for copying files to the target directory.
2) Added support for a hook that runs when a module is missing.
3) Added support for binary path includes as well as excludes; use sequences
rather than dictionaries as a more convenient API; exclude the standard
locations for 32-bit and 64-bit libaries in multi-architecture systems.
4) Added support for searching zip files (egg files) for modules.
5) Added support for handling system exit exceptions similarly to what Python
does itself as requested by Sylvain.
6) Added code to wait for threads to shut down like the normal Python
interpreter does. Thanks to Mariano Disanzo for discovering this
discrepancy.
7) Hooks added or modified based on feedback from many people.
8) Don't include the version name in the display name of the MSI.
9) Use the OS dependent path normalization routines rather than simply use the
lowercase value as on Unix case is important; thanks to Artie Eoff for
pointing this out.
10) Include a version attribute in the cx_Freeze package and display it in the
output for the --version option to the script.
11) Include build instructions as requested by Norbert Sebok.
12) Add support for copying files when modules are included which require data
files to operate properly; add support for copying the necessary files for
the Tkinter and matplotlib modules.
13) Handle deferred imports recursively as needed; ensure that from lists do
not automatically indicate that they are part of the module or the deferred
import processing doesn't actually work!
14) Handle the situation where a module imports everything from a package and
the __all__ variable has been defined but the package has not actually
imported everything in the __all__ variable during initialization.
15) Modified license text to more closely match the Python Software Foundation
license as was intended.
16) Added sample script for freezing an application using matplotlib.
17) Renamed freeze to cxfreeze to avoid conflict with another package that uses
that executable as requested by Siegfried Gevatter.
Changes from 3.0.3 to 4.0b1
1) Added support for placing modules in library.zip or in a separate zip file
for each executable that is produced.
2) Added support for copying binary dependent files (DLLs and shared
libraries)
3) Added support for including all submodules in a package
4) Added support for including icons in Windows executables
5) Added support for constants module which can be used for determining
certain build constants at runtime
6) Added support for relative imports available in Python 2.5 and up
7) Added support for building Windows installers (Python 2.5 and up) and
RPM packages
8) Added support for distutils configuration scripts
9) Added support for hooks which can force inclusion or exclusion of modules
when certain modules are included
10) Added documentation and samples
11) Added setup.py for building the cx_Freeze package instead of a script
used to build only the frozen bases
12) FreezePython renamed to a script called freeze in the Python distribution
13) On Linux and other platforms that support it set LD_RUN_PATH to include
the directory in which the executable is located
Changes from 3.0.2 to 3.0.3
1) In Common.c, used MAXPATHLEN defined in the Python OS independent include
file rather than the PATH_MAX define which is OS dependent and is not
available on IRIX as noted by Andrew Jones.
2) In the initscript ConsoleSetLibPath.py, added lines from initscript
Console.py that should have been there since the only difference between
that script and this one is the automatic re-execution of the executable.
3) Added an explicit "import encodings" to the initscripts in order to handle
Unicode encodings a little better. Thanks to Ralf Schmitt for pointing out
the problem and its solution.
4) Generated a meaningful name for the extension loader script so that it is
clear which particular extension module is being loaded when an exception
is being raised.
5) In MakeFrozenBases.py, use distutils to figure out a few more
platform-dependent linker flags as suggested by Ralf Schmitt.
Changes from 3.0.1 to 3.0.2
1) Add support for compressing the byte code in the zip files that are
produced.
2) Add better support for the win32com package as requested by Barry Scott.
3) Prevent deletion of target file if it happens to be identical to the
source file.
4) Include additional flags for local modifications to a Python build as
suggested by Benjamin Rutt.
5) Expanded instructions for building cx_Freeze from source based on a
suggestion from Gregg Lind.
6) Fix typo in help string.
Changes from 3.0 to 3.0.1
1) Added option --default-path which is used to specify the path used when
finding modules. This is particularly useful when performing cross
compilations (such as for building a frozen executable for Windows CE).
2) Added option --shared-lib-name which can be used to specify the name of
the shared library (DLL) implementing the Python runtime that is required
for the frozen executable to work. This option is also particularly useful
when cross compiling since the normal method for determining this
information cannot be used.
3) Added option --zip-include which allows for additional files to be added
to the zip file that contains the modules that implement the Python
script. Thanks to Barray Warsaw for providing the initial patch.
4) Added support for handling read-only files properly. Thanks to Peter
Grayson for pointing out the problem and providing a solution.
5) Added support for a frozen executable to be a symbolic link. Thanks to
Robert Kiendl for providing the initial patch.
6) Enhanced the support for running a frozen executable that uses an existing
Python installation to locate modules it requires. This is primarily of
use for embedding Python where the interface is C but the ability to run
from source is still desired.
7) Modified the documentation to indicate that building from source on
Windows currently requires the mingw compiler (http://www.mingw.org).
8) Workaround the problem in Python 2.3 (fixed in Python 2.4) which causes a
broken module to be left in sys.modules if an ImportError takes place
during the execution of the code in that module. Thanks to Roger Binns
for pointing this out.
Changes from 3.0 beta3 to 3.0
1) Ensure that ldd is only run on extension modules.
2) Allow for using a compiler other than gcc for building the frozen base
executables by setting the environment variable CC.
3) Ensure that the import lock is not held while executing the main script;
otherwise, attempts to import a module within a thread will hang that
thread as noted by Roger Binns.
4) Added support for replacing the paths in all frozen modules with something
else (so that for example the path of the machine on which the freezing
was done is not displayed in tracebacks)
Changes from 3.0 beta2 to 3.0 beta3
1) Explicitly include the warnings module so that at runtime warnings are
suppressed as when running Python normally.
2) Improve the extension loader so that an ImportError is raised when the
dynamic module is not located; otherwise an error about missing attributes
is raised instead.
3) Extension loaders are only created when copying dependencies since the
normal module should be loadable in the situation where a Python
installation is available.
4) Added support for Python 2.4.
5) Fixed the dependency checking for wxPython to be a little more
intelligent.
Changes from 3.0 beta1 to 3.0 beta2
1) Fix issues with locating the initscripts and bases relative to the
directory in which the executable was started.
2) Added new base executable ConsoleKeepPath which is used when an existing
Python installation is required (such as for FreezePython itself).
3) Forced the existence of a Python installation to be ignored when using the
standard Console base executable.
4) Remove the existing file when copying dependent files; otherwise, an error
is raised when attempting to overwrite read-only files.
5) Added option -O (or -OO) to FreezePython to set the optimization used when
generating bytecode.
Changes from 2.2 to 3.0 beta1
1) cx_Freeze now requires Python 2.3 or higher since it takes advantage of
the ability of Python 2.3 and higher to import modules from zip files.
This makes the freezing process considerably simpler and also allows for
the execution of multiple frozen packages (such as found in COM servers or
shared libraries) without requiring modification to the Python modules.
2) All external dependencies have been removed. cx_Freeze now only requires
a standard Python distribution to do its work.
3) Added the ability to define the initialization scripts that cx_Freeze uses
on startup of the frozen program. Previously, these scripts were written
in C and could not easily be changed; now they are written in Python and
can be found in the initscripts directory (and chosen with the
new --init-script option to FreezePython).
4) The base executable ConsoleSetLibPath has been removed and replaced with
the initscript ConsoleSetLibPath.
5) Removed base executables for Win32 services and Win32 COM servers. This
functionality will be restored in the future but it is not currently in a
state that is ready for release. If this functionality is required, please
use py2exe or contact me for my work in progress.
6) The attribute sys.frozen is now set so that more recent pywin32 modules
work as expected when frozen.
7) Added option --include-path to FreezePython to allow overriding of
sys.path without modifying the environment variable PYTHONPATH.
8) Added option --target-dir/--install-dir to specify the directory in which
the frozen executable and its dependencies will be placed.
9) Removed the option --shared-lib since it was used for building shared
libraries and can be managed with the initscript SharedLib.py.
10) MakeFrozenBases.py now checks the platform specific include directory as
requested by Michael Partridge.
Changes from 2.1 to 2.2
1) Add option (--ext-list-file) to FreezePython to write the list of
extensions copied to the installation directory to a file. This option is
useful in cases where multiple builds are performed into the same
installation directory.
2) Pass the arguments on the command line through to Win32 GUI applications.
Thanks to Michael Porter for pointing this out.
3) Link directly against the python DLL when building the frozen bases on
Windows, thus eliminating the need for building an import library.
4) Force sys.path to include the directory in which the script to be frozen
is found.
5) Make sure that the installation directory exists before attempting to
copy the target binary into it.
6) The Win32GUI base has been modified to display fatal errors in message
boxes, rather than printing errors to stderr, since on Windows the
standard file IO handles are all closed.
Changes from 2.0 to 2.1
1) Remove dependency on Python 2.2. Thanks to Paul Moore for not only
pointing it out but providing patches.
2) Set up the list of frozen modules in advance, rather than doing it after
Python is initialized so that implicit imports done by Python can be
satisfied. The bug in Python 2.3 that demonstrated this issue has been
fixed in the first release candidate. Thanks to Thomas Heller for pointing
out the obvious in this instance!
3) Added additional base executable (ConsoleSetLibPath) to support setting
the LD_LIBRARY_PATH variable on Unix platforms and restarting the
executable to put the new setting into effect. This is primarily of use
in distributing wxPython applications on Unix where the shared library
has an embedded RPATH value which can cause problems.
4) Small improvements of documentation based on feedback from several people.
5) Print information about the files written or copied during the freezing
process.
6) Do not copy extensions when freezing if the path is being overridden since
it is expected that a full Python installation is available to the target
users of the frozen binary.
7) Provide meaningful error message when the wxPython library cannot be
found during the freezing process.
Changes from 1.1 to 2.0
1) Added support for in process (DLL) COM servers using PythonCOM.
2) Ensured that the frozen flag is set prior to determining the full path for
the program in order to avoid warnings about Python not being found on
some platforms.
3) Added include file and resource file to the source tree to avoid the
dependency on the Wine message compiler for Win32 builds.
4) Dropped the option --copy-extensions; this now happens automatically since
the resulting binary is useless without them.
5) Added a sample for building a Win32 service.
6) Make use of improved modules from Python 2.3 (which function under 2.2)
Changes from 1.0 to 1.1
1) Fixed import error with C extensions in packages; thanks to Thomas Heller
for pointing out the solution to this problem.
2) Added options to FreezePython to allow for the inclusion of modules which
will not be found by the module finder (--include-modules) and the
exclusion of modules which will be found by the module finder but should
not be included (--exclude-modules).
3) Fixed typo in README.txt.

View File

@ -0,0 +1,53 @@
Copyright © 2007-2008, Colt Engineering, Edmonton, Alberta, Canada.
Copyright © 2001-2006, Computronix (Canada) Ltd., Edmonton, Alberta, Canada.
All rights reserved.
NOTE: this license is derived from the Python Software Foundation License
which can be found at http://www.python.org/psf/license
License for cx_Freeze 4.0.1
---------------------------
1. This LICENSE AGREEMENT is between the copyright holders and the Individual
or Organization ("Licensee") accessing and otherwise using cx_Freeze
software in source or binary form and its associated documentation.
2. Subject to the terms and conditions of this License Agreement, the
copyright holders hereby grant Licensee a nonexclusive, royalty-free,
world-wide license to reproduce, analyze, test, perform and/or display
publicly, prepare derivative works, distribute, and otherwise use cx_Freeze
alone or in any derivative version, provided, however, that this License
Agreement and this notice of copyright are retained in cx_Freeze alone or in
any derivative version prepared by Licensee.
3. In the event Licensee prepares a derivative work that is based on or
incorporates cx_Freeze or any part thereof, and wants to make the derivative
work available to others as provided herein, then Licensee hereby agrees to
include in any such work a brief summary of the changes made to cx_Freeze.
4. The copyright holders are making cx_Freeze available to Licensee on an
"AS IS" basis. THE COPYRIGHT HOLDERS MAKE NO REPRESENTATIONS OR WARRANTIES,
EXPRESS OR IMPLIED. BY WAY OF EXAMPLE, BUT NOT LIMITATION, THE COPYRIGHT
HOLDERS MAKE NO AND DISCLAIM ANY REPRESENTATION OR WARRANTY OF
MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF
CX_FREEZE WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.
5. THE COPYRIGHT HOLDERS SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF
CX_FREEZE FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS AS
A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING CX_FREEZE, OR ANY
DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.
6. This License Agreement will automatically terminate upon a material breach
of its terms and conditions.
7. Nothing in this License Agreement shall be deemed to create any relationship
of agency, partnership, or joint venture between the copyright holders and
Licensee. This License Agreement does not grant permission to use
copyright holder's trademarks or trade name in a trademark sense to endorse
or promote products or services of Licensee, or any third party.
8. By copying, installing or otherwise using cx_Freeze, Licensee agrees to be
bound by the terms and conditions of this License Agreement.
Computronix® is a registered trademark of Computronix (Canada) Ltd.

View File

@ -0,0 +1,6 @@
include MANIFEST.in
include *.txt
recursive-include doc *.html
recursive-include initscripts *.py
recursive-include samples *.py
recursive-include source *.c *.rc

View File

@ -0,0 +1,22 @@
Metadata-Version: 1.0
Name: cx_Freeze
Version: 4.0.1
Summary: create standalone executables from Python scripts
Home-page: http://cx-freeze.sourceforge.net
Author: Anthony Tuininga
Author-email: anthony.tuininga@gmail.com
License: Python Software Foundation License
Description: create standalone executables from Python scripts
Keywords: freeze
Platform: UNKNOWN
Classifier: Development Status :: 5 - Production/Stable
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Python Software Foundation License
Classifier: Natural Language :: English
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: C
Classifier: Programming Language :: Python
Classifier: Topic :: Software Development :: Build Tools
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: System :: Software Distribution
Classifier: Topic :: Utilities

View File

@ -0,0 +1,12 @@
Please see cx_Freeze.html for documentation on how to use cx_Freeze.
To build:
python setup.py build
python setup.py install
On Windows I have used the MinGW compiler (http://www.mingw.org)
python setup.py build --compiler=mingw32
python setup.py build --compiler=mingw32 install

View File

@ -0,0 +1,14 @@
version = "4.0.1"
import sys
from dist import *
if sys.platform == "win32" and sys.version_info[:2] >= (2, 5):
from windist import *
from finder import *
from freezer import *
from main import *
del dist
del finder
del freezer

View File

@ -0,0 +1,279 @@
import distutils.command.bdist_rpm
import distutils.command.build
import distutils.command.install
import distutils.core
import distutils.dir_util
import distutils.dist
import distutils.util
import distutils.version
import os
import sys
import cx_Freeze
__all__ = [ "bdist_rpm", "build", "build_exe", "install", "install_exe",
"setup" ]
class Distribution(distutils.dist.Distribution):
def __init__(self, attrs):
self.executables = []
distutils.dist.Distribution.__init__(self, attrs)
class bdist_rpm(distutils.command.bdist_rpm.bdist_rpm):
def finalize_options(self):
distutils.command.bdist_rpm.bdist_rpm.finalize_options(self)
self.use_rpm_opt_flags = 1
def _make_spec_file(self):
contents = distutils.command.bdist_rpm.bdist_rpm._make_spec_file(self)
return [c for c in contents if c != 'BuildArch: noarch']
class build(distutils.command.build.build):
user_options = distutils.command.build.build.user_options + [
('build-exe=', None, 'build directory for executables')
]
def get_sub_commands(self):
subCommands = distutils.command.build.build.get_sub_commands(self)
if self.distribution.executables:
subCommands.append("build_exe")
return subCommands
def initialize_options(self):
distutils.command.build.build.initialize_options(self)
self.build_exe = None
def finalize_options(self):
distutils.command.build.build.finalize_options(self)
if self.build_exe is None:
dirName = "exe.%s-%s" % \
(distutils.util.get_platform(), sys.version[0:3])
self.build_exe = os.path.join(self.build_base, dirName)
class build_exe(distutils.core.Command):
description = "build executables from Python scripts"
user_options = [
('build-exe=', 'b',
'directory for built executables'),
('optimize=', 'O',
'optimization level: -O1 for "python -O", '
'-O2 for "python -OO" and -O0 to disable [default: -O0]'),
('excludes=', 'e',
'comma-separated list of modules to exclude'),
('includes=', 'i',
'comma-separated list of modules to include'),
('packages=', 'p',
'comma-separated list of packages to include'),
('replace-paths=', None,
'comma-separated list of paths to replace in included modules'),
('path=', None,
'comma-separated list of paths to search'),
('init-script=', 'i',
'name of script to use during initialization'),
('base=', None,
'name of base executable to use'),
('compressed', 'c',
'create a compressed zipfile'),
('copy-dependent-files', None,
'copy all dependent files'),
('create-shared-zip', None,
'create a shared zip file containing shared modules'),
('append-script-to-exe', None,
'append the script module to the exe'),
('include-in-shared-zip', None,
'include the script module in the shared zip file'),
('icon', None,
'include the icon along with the frozen executable(s)'),
('constants=', None,
'comma-separated list of constants to include'),
('include-files=', 'f',
'list of tuples of additional files to include in distribution'),
('bin-includes', None,
'list of names of files to include when determining dependencies'),
('bin-excludes', None,
'list of names of files to exclude when determining dependencies')
]
boolean_options = ["compressed", "copy_dependent_files",
"create_shared_zip", "append_script_to_exe",
"include_in_shared_zip"]
def _normalize(self, attrName):
value = getattr(self, attrName)
if value is None:
normalizedValue = []
elif isinstance(value, basestring):
normalizedValue = value.split()
else:
normalizedValue = list(value)
setattr(self, attrName, normalizedValue)
def initialize_options(self):
self.optimize = 0
self.build_exe = None
self.excludes = []
self.includes = []
self.packages = []
self.replace_paths = []
self.compressed = None
self.copy_dependent_files = None
self.init_script = None
self.base = None
self.path = None
self.create_shared_zip = None
self.append_script_to_exe = None
self.include_in_shared_zip = None
self.icon = None
self.constants = []
self.include_files = []
self.bin_excludes = []
self.bin_includes = []
def finalize_options(self):
self.set_undefined_options('build', ('build_exe', 'build_exe'))
self.optimize = int(self.optimize)
self._normalize("excludes")
self._normalize("includes")
self._normalize("packages")
self._normalize("constants")
def run(self):
metadata = self.distribution.metadata
constantsModule = cx_Freeze.ConstantsModule(metadata.version)
for constant in self.constants:
parts = constant.split("=")
if len(parts) == 1:
name = constant
value = None
else:
name, stringValue = parts
value = eval(stringValue)
constantsModule.values[name] = value
freezer = cx_Freeze.Freezer(self.distribution.executables,
[constantsModule], self.includes, self.excludes, self.packages,
self.replace_paths, self.compressed, self.optimize,
self.copy_dependent_files, self.init_script, self.base,
self.path, self.create_shared_zip, self.append_script_to_exe,
self.include_in_shared_zip, self.build_exe, icon = self.icon,
includeFiles = self.include_files,
binIncludes = self.bin_includes,
binExcludes = self.bin_excludes)
freezer.Freeze()
class install(distutils.command.install.install):
user_options = distutils.command.install.install.user_options + [
('install-exe=', None,
'installation directory for executables')
]
def expand_dirs(self):
distutils.command.install.install.expand_dirs(self)
self._expand_attrs(['install_exe'])
def get_sub_commands(self):
subCommands = distutils.command.install.install.get_sub_commands(self)
if self.distribution.executables:
subCommands.append("install_exe")
return [s for s in subCommands if s != "install_egg_info"]
def initialize_options(self):
distutils.command.install.install.initialize_options(self)
self.install_exe = None
def finalize_options(self):
if self.prefix is None and sys.platform == "win32":
import _winreg
key = _winreg.OpenKey(_winreg.HKEY_LOCAL_MACHINE,
r"Software\Microsoft\Windows\CurrentVersion")
prefix = str(_winreg.QueryValueEx(key, "ProgramFilesDir")[0])
metadata = self.distribution.metadata
dirName = "%s-%s" % (metadata.name, metadata.version)
self.prefix = "%s/%s" % (prefix, dirName)
distutils.command.install.install.finalize_options(self)
self.convert_paths('exe')
if self.root is not None:
self.change_roots('exe')
def select_scheme(self, name):
distutils.command.install.install.select_scheme(self, name)
if self.install_exe is None:
if sys.platform == "win32":
self.install_exe = '$base'
else:
metadata = self.distribution.metadata
dirName = "%s-%s" % (metadata.name, metadata.version)
self.install_exe = '$base/lib/%s' % dirName
class install_exe(distutils.core.Command):
description = "install executables built from Python scripts"
user_options = [
('install-dir=', 'd', 'directory to install executables to'),
('build-dir=', 'b', 'build directory (where to install from)'),
('force', 'f', 'force installation (overwrite existing files)'),
('skip-build', None, 'skip the build steps')
]
def initialize_options(self):
self.install_dir = None
self.force = 0
self.build_dir = None
self.skip_build = None
def finalize_options(self):
self.set_undefined_options('build', ('build_exe', 'build_dir'))
self.set_undefined_options('install',
('install_exe', 'install_dir'),
('force', 'force'),
('skip_build', 'skip_build'))
def run(self):
if not self.skip_build:
self.run_command('build_exe')
self.outfiles = self.copy_tree(self.build_dir, self.install_dir)
if sys.platform != "win32":
baseDir = os.path.dirname(os.path.dirname(self.install_dir))
binDir = os.path.join(baseDir, "bin")
if not os.path.exists(binDir):
os.makedirs(binDir)
sourceDir = os.path.join("..", self.install_dir[len(baseDir) + 1:])
for executable in self.distribution.executables:
name = os.path.basename(executable.targetName)
source = os.path.join(sourceDir, name)
target = os.path.join(binDir, name)
if os.path.exists(target):
os.unlink(target)
os.symlink(source, target)
self.outfiles.append(target)
def get_inputs(self):
return self.distribution.executables or []
def get_outputs(self):
return self.outfiles or []
def _AddCommandClass(commandClasses, name, cls):
if name not in commandClasses:
commandClasses[name] = cls
def setup(**attrs):
attrs["distclass"] = Distribution
commandClasses = attrs.setdefault("cmdclass", {})
if sys.platform == "win32":
if sys.version_info[:2] >= (2, 5):
_AddCommandClass(commandClasses, "bdist_msi", cx_Freeze.bdist_msi)
else:
_AddCommandClass(commandClasses, "bdist_rpm", cx_Freeze.bdist_rpm)
_AddCommandClass(commandClasses, "build", build)
_AddCommandClass(commandClasses, "build_exe", build_exe)
_AddCommandClass(commandClasses, "install", install)
_AddCommandClass(commandClasses, "install_exe", install_exe)
distutils.core.setup(**attrs)

View File

@ -0,0 +1,455 @@
"""
Base class for finding modules.
"""
import dis
import imp
import marshal
import new
import opcode
import os
import sys
import zipfile
import cx_Freeze.hooks
BUILD_LIST = opcode.opmap["BUILD_LIST"]
INPLACE_ADD = opcode.opmap["INPLACE_ADD"]
LOAD_CONST = opcode.opmap["LOAD_CONST"]
IMPORT_NAME = opcode.opmap["IMPORT_NAME"]
IMPORT_FROM = opcode.opmap["IMPORT_FROM"]
STORE_NAME = opcode.opmap["STORE_NAME"]
STORE_GLOBAL = opcode.opmap["STORE_GLOBAL"]
STORE_OPS = (STORE_NAME, STORE_GLOBAL)
__all__ = [ "Module", "ModuleFinder" ]
class ModuleFinder(object):
def __init__(self, includeFiles, excludes, path, replacePaths):
self.includeFiles = includeFiles
self.excludes = dict.fromkeys(excludes)
self.replacePaths = replacePaths
self.path = path or sys.path
self.modules = []
self.aliases = {}
self._modules = dict.fromkeys(excludes)
self._builtinModules = dict.fromkeys(sys.builtin_module_names)
self._badModules = {}
self._zipFileEntries = {}
self._zipFiles = {}
cx_Freeze.hooks.initialize(self)
def _AddModule(self, name):
"""Add a module to the list of modules but if one is already found,
then return it instead; this is done so that packages can be
handled properly."""
module = self._modules.get(name)
if module is None:
module = self._modules[name] = Module(name)
self.modules.append(module)
if name in self._badModules:
del self._badModules[name]
return module
def _DetermineParent(self, caller):
"""Determine the parent to use when searching packages."""
if caller is not None:
if caller.path is not None:
return caller
return self._GetParentByName(caller.name)
def _EnsureFromList(self, caller, packageModule, fromList,
deferredImports):
"""Ensure that the from list is satisfied. This is only necessary for
package modules. If the caller is the package itself, actually
attempt to import right then since it must be a submodule; otherwise
defer until after all global names are defined in order to avoid
spurious complaints about missing modules."""
if caller is not packageModule:
deferredImports.append((packageModule, fromList))
else:
if fromList == ("*",):
fromList = packageModule.allNames
for name in fromList:
if name in packageModule.globalNames:
continue
subModuleName = "%s.%s" % (packageModule.name, name)
self._ImportModule(subModuleName, deferredImports, caller)
def _FindModule(self, name, path):
try:
return imp.find_module(name, path)
except ImportError:
if not path:
path = []
for location in path:
if name in self._zipFileEntries:
break
if location in self._zipFiles:
continue
if os.path.isdir(location) or not zipfile.is_zipfile(location):
self._zipFiles[location] = None
continue
zip = zipfile.ZipFile(location)
for archiveName in zip.namelist():
baseName, ext = os.path.splitext(archiveName)
if ext not in ('.pyc', '.pyo'):
continue
moduleName = ".".join(baseName.split("/"))
if moduleName in self._zipFileEntries:
continue
self._zipFileEntries[moduleName] = (zip, archiveName)
self._zipFiles[location] = None
info = self._zipFileEntries.get(name)
if info is not None:
zip, archiveName = info
fp = zip.read(archiveName)
info = (".pyc", "rb", imp.PY_COMPILED)
return fp, os.path.join(zip.filename, archiveName), info
raise
def _GetParentByName(self, name):
"""Return the parent module given the name of a module."""
pos = name.rfind(".")
if pos > 0:
parentName = name[:pos]
return self._modules[parentName]
def _ImportAllSubModules(self, module, deferredImports, recursive = True):
"""Import all sub modules to the given package."""
suffixes = dict.fromkeys([s[0] for s in imp.get_suffixes()])
for dir in module.path:
try:
fileNames = os.listdir(dir)
except os.error:
continue
for fileName in fileNames:
name, ext = os.path.splitext(fileName)
if ext not in suffixes:
continue
if name == "__init__":
continue
subModuleName = "%s.%s" % (module.name, name)
subModule, returnError = \
self._InternalImportModule(subModuleName,
deferredImports)
if returnError and subModule is None:
raise ImportError, "No module named %s" % subModuleName
module.globalNames[name] = None
if subModule.path and recursive:
self._ImportAllSubModules(subModule, deferredImports,
recursive)
def _ImportDeferredImports(self, deferredImports):
"""Import any sub modules that were deferred, if applicable."""
while deferredImports:
newDeferredImports = []
for packageModule, subModuleNames in deferredImports:
self._EnsureFromList(packageModule, packageModule,
subModuleNames, newDeferredImports)
deferredImports = newDeferredImports
def _ImportModule(self, name, deferredImports, caller = None,
relativeImportIndex = 0):
"""Attempt to find the named module and return it or None if no module
by that name could be found."""
# absolute import (available in Python 2.5 and up)
# the name given is the only name that will be searched
if relativeImportIndex == 0:
module, returnError = self._InternalImportModule(name,
deferredImports)
# old style relative import (only possibility in Python 2.4 and prior)
# the name given is tried in all parents until a match is found and if
# no match is found, the global namespace is searched
elif relativeImportIndex < 0:
parent = self._DetermineParent(caller)
while parent is not None:
fullName = "%s.%s" % (parent.name, name)
module, returnError = self._InternalImportModule(fullName,
deferredImports)
if module is not None:
parent.globalNames[name] = None
return module
parent = self._GetParentByName(parent.name)
module, returnError = self._InternalImportModule(name,
deferredImports)
# new style relative import (available in Python 2.5 and up)
# the index indicates how many levels to traverse and only that level
# is searched for the named module
elif relativeImportIndex > 0:
parent = caller
if parent.path is not None:
relativeImportIndex -= 1
while parent is not None and relativeImportIndex > 0:
parent = self._GetParentByName(parent.name)
relativeImportIndex -= 1
if parent is None:
module = None
returnError = True
elif not name:
module = parent
else:
name = "%s.%s" % (parent.name, name)
module, returnError = self._InternalImportModule(name,
deferredImports)
# if module not found, track that fact
if module is None:
if caller is None:
raise ImportError, "No module named %s" % name
self._RunHook("missing", name, caller)
if returnError and name not in caller.ignoreNames:
callers = self._badModules.setdefault(name, {})
callers[caller.name] = None
return module
def _InternalImportModule(self, name, deferredImports):
"""Internal method used for importing a module which assumes that the
name given is an absolute name. None is returned if the module
cannot be found."""
try:
return self._modules[name], False
except KeyError:
pass
if name in self._builtinModules:
module = self._AddModule(name)
self._RunHook("load", module.name, module)
return module, False
pos = name.rfind(".")
if pos < 0:
path = self.path
searchName = name
parentModule = None
else:
parentName = name[:pos]
parentModule, returnError = \
self._InternalImportModule(parentName, deferredImports)
if parentModule is None:
return None, returnError
path = parentModule.path
searchName = name[pos + 1:]
if name in self.aliases:
actualName = self.aliases[name]
module, returnError = \
self._InternalImportModule(actualName, deferredImports)
self._modules[name] = module
return module, returnError
try:
fp, path, info = self._FindModule(searchName, path)
except ImportError:
self._modules[name] = None
return None, True
module = self._LoadModule(name, fp, path, info, deferredImports,
parentModule)
return module, False
def _LoadModule(self, name, fp, path, info, deferredImports,
parent = None):
"""Load the module, given the information acquired by the finder."""
suffix, mode, type = info
if type == imp.PKG_DIRECTORY:
return self._LoadPackage(name, path, parent, deferredImports)
module = self._AddModule(name)
module.file = path
module.parent = parent
if type == imp.PY_SOURCE:
module.code = compile(fp.read() + "\n", path, "exec")
elif type == imp.PY_COMPILED:
if isinstance(fp, str):
magic = fp[:4]
else:
magic = fp.read(4)
if magic != imp.get_magic():
raise ImportError, "Bad magic number in %s" % path
if isinstance(fp, str):
module.code = marshal.loads(fp[8:])
module.inZipFile = True
else:
fp.read(4)
module.code = marshal.load(fp)
self._RunHook("load", module.name, module)
if module.code is not None:
if self.replacePaths:
topLevelModule = module
while topLevelModule.parent is not None:
topLevelModule = topLevelModule.parent
module.code = self._ReplacePathsInCode(topLevelModule,
module.code)
self._ScanCode(module.code, module, deferredImports)
return module
def _LoadPackage(self, name, path, parent, deferredImports):
"""Load the package, given its name and path."""
module = self._AddModule(name)
module.path = [path]
fp, path, info = imp.find_module("__init__", module.path)
self._LoadModule(name, fp, path, info, deferredImports, parent)
return module
def _ReplacePathsInCode(self, topLevelModule, co):
"""Replace paths in the code as directed, returning a new code object
with the modified paths in place."""
origFileName = newFileName = os.path.normpath(co.co_filename)
for searchValue, replaceValue in self.replacePaths:
if searchValue == "*":
searchValue = os.path.dirname(topLevelModule.file)
if topLevelModule.path:
searchValue = os.path.dirname(searchValue)
if searchValue:
searchValue = searchValue + os.pathsep
elif not origFileName.startswith(searchValue):
continue
newFileName = replaceValue + origFileName[len(searchValue):]
break
constants = list(co.co_consts)
for i, value in enumerate(constants):
if isinstance(value, type(co)):
constants[i] = self._ReplacePathsInCode(topLevelModule, value)
return new.code(co.co_argcount, co.co_nlocals, co.co_stacksize,
co.co_flags, co.co_code, tuple(constants), co.co_names,
co.co_varnames, newFileName, co.co_name, co.co_firstlineno,
co.co_lnotab, co.co_freevars, co.co_cellvars)
def _RunHook(self, hookName, moduleName, *args):
"""Run hook for the given module if one is present."""
name = "%s_%s" % (hookName, moduleName.replace(".", "_"))
method = getattr(cx_Freeze.hooks, name, None)
if method is not None:
method(self, *args)
def _ScanCode(self, co, module, deferredImports):
"""Scan code, looking for imported modules and keeping track of the
constants that have been created in order to better tell which
modules are truly missing."""
opIndex = 0
arguments = []
code = co.co_code
numOps = len(code)
while opIndex < numOps:
op = ord(code[opIndex])
opIndex += 1
if op >= dis.HAVE_ARGUMENT:
opArg = ord(code[opIndex]) + ord(code[opIndex + 1]) * 256
opIndex += 2
if op == LOAD_CONST:
arguments.append(co.co_consts[opArg])
elif op == IMPORT_NAME:
name = co.co_names[opArg]
if len(arguments) == 2:
relativeImportIndex, fromList = arguments
else:
relativeImportIndex = -1
fromList, = arguments
if name not in module.excludeNames:
subModule = self._ImportModule(name, deferredImports,
module, relativeImportIndex)
if subModule is not None:
module.globalNames.update(subModule.globalNames)
if fromList and subModule.path is not None:
self._EnsureFromList(module, subModule, fromList,
deferredImports)
elif op == IMPORT_FROM:
opIndex += 3
elif op not in (BUILD_LIST, INPLACE_ADD):
if op in STORE_OPS:
name = co.co_names[opArg]
if name == "__all__":
module.allNames.extend(arguments)
module.globalNames[name] = None
arguments = []
for constant in co.co_consts:
if isinstance(constant, type(co)):
self._ScanCode(constant, module, deferredImports)
def AddAlias(self, name, aliasFor):
"""Add an alias for a particular module; when an attempt is made to
import a module using the alias name, import the actual name
instead."""
self.aliases[name] = aliasFor
def ExcludeModule(self, name):
"""Exclude the named module from the resulting frozen executable."""
self.excludes[name] = None
self._modules[name] = None
def IncludeFile(self, path, moduleName = None):
"""Include the named file as a module in the frozen executable."""
name, ext = os.path.splitext(os.path.basename(path))
if moduleName is None:
moduleName = name
info = (ext, "r", imp.PY_SOURCE)
deferredImports = []
module = self._LoadModule(moduleName, file(path, "U"), path, info,
deferredImports)
self._ImportDeferredImports(deferredImports)
return module
def IncludeFiles(self, sourcePath, targetPath):
"""Include the files in the given directory in the target build."""
self.includeFiles.append((sourcePath, targetPath))
def IncludeModule(self, name):
"""Include the named module in the frozen executable."""
deferredImports = []
module = self._ImportModule(name, deferredImports)
self._ImportDeferredImports(deferredImports)
return module
def IncludePackage(self, name):
"""Include the named package and any submodules in the frozen
executable."""
deferredImports = []
module = self._ImportModule(name, deferredImports)
if module.path:
self._ImportAllSubModules(module, deferredImports)
self._ImportDeferredImports(deferredImports)
return module
def ReportMissingModules(self):
if self._badModules:
print "Missing modules:"
names = self._badModules.keys()
names.sort()
for name in names:
callers = self._badModules[name].keys()
callers.sort()
print "?", name, "imported from", ", ".join(callers)
print
class Module(object):
def __init__(self, name):
self.name = name
self.file = None
self.path = None
self.code = None
self.parent = None
self.globalNames = {}
self.excludeNames = {}
self.ignoreNames = {}
self.allNames = []
self.inZipFile = False
def __repr__(self):
parts = ["name=%s" % repr(self.name)]
if self.file is not None:
parts.append("file=%s" % repr(self.file))
if self.path is not None:
parts.append("path=%s" % repr(self.path))
return "<Module %s>" % ", ".join(parts)
def AddGlobalName(self, name):
self.globalNames[name] = None
def ExcludeName(self, name):
self.excludeNames[name] = None
def IgnoreName(self, name):
self.ignoreNames[name] = None

View File

@ -0,0 +1,550 @@
"""
Base class for freezing scripts into executables.
"""
import datetime
import distutils.sysconfig
import imp
import marshal
import os
import shutil
import socket
import stat
import struct
import sys
import time
import zipfile
import cx_Freeze
import cx_Freeze.util
__all__ = [ "ConfigError", "ConstantsModule", "Executable", "Freezer" ]
if sys.platform == "win32":
pythonDll = "python%s%s.dll" % sys.version_info[:2]
GLOBAL_BIN_PATH_EXCLUDES = [cx_Freeze.util.GetSystemDir()]
GLOBAL_BIN_INCLUDES = [
pythonDll,
"gdiplus.dll",
"mfc71.dll",
"msvcp71.dll",
"msvcr71.dll"
]
GLOBAL_BIN_EXCLUDES = [
"comctl32.dll",
"oci.dll",
"cx_Logging.pyd"
]
else:
extension = distutils.sysconfig.get_config_var("SO")
pythonSharedLib = "libpython%s.%s%s" % \
(sys.version_info[:2] + (extension,))
GLOBAL_BIN_INCLUDES = [pythonSharedLib]
GLOBAL_BIN_EXCLUDES = [
"libclntsh.so",
"libwtc9.so"
]
GLOBAL_BIN_PATH_EXCLUDES = ["/lib", "/lib32", "/lib64", "/usr/lib",
"/usr/lib32", "/usr/lib64"]
# NOTE: the try: except: block in this code is not necessary under Python 2.4
# and higher and can be removed once support for Python 2.3 is no longer needed
EXTENSION_LOADER_SOURCE = \
"""
import imp, os, sys
found = False
for p in sys.path:
if not os.path.isdir(p):
continue
f = os.path.join(p, "%s")
if not os.path.exists(f):
continue
try:
m = imp.load_dynamic(__name__, f)
except ImportError:
del sys.modules[__name__]
raise
sys.modules[__name__] = m
found = True
break
if not found:
del sys.modules[__name__]
raise ImportError, "No module named %%s" %% __name__
"""
class Freezer(object):
def __init__(self, executables, constantsModules = [], includes = [],
excludes = [], packages = [], replacePaths = [], compress = None,
optimizeFlag = 0, copyDependentFiles = None, initScript = None,
base = None, path = None, createLibraryZip = None,
appendScriptToExe = None, appendScriptToLibrary = None,
targetDir = None, binIncludes = [], binExcludes = [],
binPathIncludes = [], binPathExcludes = [], icon = None,
includeFiles = []):
self.executables = executables
self.constantsModules = constantsModules
self.includes = includes
self.excludes = excludes
self.packages = packages
self.replacePaths = replacePaths
self.compress = compress
self.optimizeFlag = optimizeFlag
self.copyDependentFiles = copyDependentFiles
self.initScript = initScript
self.base = base
self.path = path
self.createLibraryZip = createLibraryZip
self.appendScriptToExe = appendScriptToExe
self.appendScriptToLibrary = appendScriptToLibrary
self.targetDir = targetDir
self.binIncludes = [os.path.normcase(n) \
for n in GLOBAL_BIN_INCLUDES + binIncludes]
self.binExcludes = [os.path.normcase(n) \
for n in GLOBAL_BIN_EXCLUDES + binExcludes]
self.binPathIncludes = [os.path.normcase(n) for n in binPathIncludes]
self.binPathExcludes = [os.path.normcase(n) \
for n in GLOBAL_BIN_PATH_EXCLUDES + binPathExcludes]
self.icon = icon
self.includeFiles = includeFiles
self._VerifyConfiguration()
def _CopyFile(self, source, target, copyDependentFiles,
includeMode = False):
normalizedSource = os.path.normcase(os.path.normpath(source))
normalizedTarget = os.path.normcase(os.path.normpath(target))
if normalizedTarget in self.filesCopied:
return
if normalizedSource == normalizedTarget:
return
self._RemoveFile(target)
targetDir = os.path.dirname(target)
self._CreateDirectory(targetDir)
print "copying", source, "->", target
shutil.copyfile(source, target)
if includeMode:
shutil.copymode(source, target)
self.filesCopied[normalizedTarget] = None
if copyDependentFiles:
for source in self._GetDependentFiles(source):
target = os.path.join(targetDir, os.path.basename(source))
self._CopyFile(source, target, copyDependentFiles)
def _CreateDirectory(self, path):
if not os.path.isdir(path):
print "creating directory", path
os.makedirs(path)
def _FreezeExecutable(self, exe):
if self.createLibraryZip:
finder = self.finder
else:
finder = self._GetModuleFinder(exe)
if exe.script is None:
scriptModule = None
else:
scriptModule = finder.IncludeFile(exe.script, exe.moduleName)
self._CopyFile(exe.base, exe.targetName, exe.copyDependentFiles,
includeMode = True)
if exe.icon is not None:
if sys.platform == "win32":
cx_Freeze.util.AddIcon(exe.targetName, exe.icon)
else:
targetName = os.path.join(os.path.dirname(exe.targetName),
os.path.basename(exe.icon))
self._CopyFile(exe.icon, targetName,
copyDependentFiles = False)
if not os.access(exe.targetName, os.W_OK):
mode = os.stat(exe.targetName).st_mode
os.chmod(exe.targetName, mode | stat.S_IWUSR)
if not exe.appendScriptToLibrary:
if exe.appendScriptToExe:
fileName = exe.targetName
else:
baseFileName, ext = os.path.splitext(exe.targetName)
fileName = baseFileName + ".zip"
self._RemoveFile(fileName)
if not self.createLibraryZip and exe.copyDependentFiles:
scriptModule = None
self._WriteModules(fileName, exe.initScript, finder, exe.compress,
exe.copyDependentFiles, scriptModule)
def _GetBaseFileName(self, argsSource = None):
if argsSource is None:
argsSource = self
name = argsSource.base
if name is None:
if argsSource.copyDependentFiles:
name = "Console"
else:
name = "ConsoleKeepPath"
argsSource.base = self._GetFileName("bases", name)
if argsSource.base is None:
raise ConfigError("no base named %s", name)
def _GetDependentFiles(self, path):
dependentFiles = self.dependentFiles.get(path)
if dependentFiles is None:
if sys.platform == "win32":
origPath = os.environ["PATH"]
os.environ["PATH"] = origPath + os.pathsep + \
os.pathsep.join(sys.path)
dependentFiles = cx_Freeze.util.GetDependentFiles(path)
os.environ["PATH"] = origPath
else:
dependentFiles = []
for line in os.popen('ldd "%s"' % path):
parts = line.strip().split(" => ")
if len(parts) != 2:
continue
dependentFile = parts[1]
if dependentFile == "not found":
print "WARNING: cannot find", parts[0]
continue
pos = dependentFile.find(" (")
if pos >= 0:
dependentFile = dependentFile[:pos].strip()
if dependentFile:
dependentFiles.append(dependentFile)
dependentFiles = self.dependentFiles[path] = \
[f for f in dependentFiles if self._ShouldCopyFile(f)]
return dependentFiles
def _GetFileName(self, dir, name):
if os.path.isabs(name):
return name
name = os.path.normcase(name)
fullDir = os.path.join(os.path.dirname(cx_Freeze.__file__), dir)
if os.path.isdir(fullDir):
for fileName in os.listdir(fullDir):
if name == os.path.splitext(os.path.normcase(fileName))[0]:
return os.path.join(fullDir, fileName)
def _GetInitScriptFileName(self, argsSource = None):
if argsSource is None:
argsSource = self
name = argsSource.initScript
if name is None:
if argsSource.copyDependentFiles:
name = "Console"
else:
name = "ConsoleKeepPath"
argsSource.initScript = self._GetFileName("initscripts", name)
if argsSource.initScript is None:
raise ConfigError("no initscript named %s", name)
def _GetModuleFinder(self, argsSource = None):
if argsSource is None:
argsSource = self
finder = cx_Freeze.ModuleFinder(self.includeFiles, argsSource.excludes,
argsSource.path, argsSource.replacePaths)
if argsSource.copyDependentFiles:
finder.IncludeModule("imp")
finder.IncludeModule("os")
finder.IncludeModule("sys")
if argsSource.compress:
finder.IncludeModule("zlib")
for name in argsSource.includes:
finder.IncludeModule(name)
for name in argsSource.packages:
finder.IncludePackage(name)
return finder
def _PrintReport(self, fileName, modules):
print "writing zip file", fileName
print
print " %-25s %s" % ("Name", "File")
print " %-25s %s" % ("----", "----")
for module in modules:
if module.path:
print "P",
else:
print "m",
print "%-25s" % module.name, module.file or ""
print
def _RemoveFile(self, path):
if os.path.exists(path):
os.chmod(path, 0777)
os.remove(path)
def _ShouldCopyFile(self, path):
dir, name = os.path.split(os.path.normcase(path))
parts = name.split(".")
tweaked = False
while True:
if not parts[-1].isdigit():
break
parts.pop(-1)
tweaked = True
if tweaked:
name = ".".join(parts)
if name in self.binIncludes:
return True
if name in self.binExcludes:
return False
for path in self.binPathIncludes:
if dir.startswith(path):
return True
for path in self.binPathExcludes:
if dir.startswith(path):
return False
return True
def _VerifyCanAppendToLibrary(self):
if not self.createLibraryZip:
raise ConfigError("script cannot be appended to library zip if "
"one is not being created")
def _VerifyConfiguration(self):
if self.compress is None:
self.compress = True
if self.copyDependentFiles is None:
self.copyDependentFiles = True
if self.createLibraryZip is None:
self.createLibraryZip = True
if self.appendScriptToExe is None:
self.appendScriptToExe = False
if self.appendScriptToLibrary is None:
self.appendScriptToLibrary = \
self.createLibraryZip and not self.appendScriptToExe
if self.targetDir is None:
self.targetDir = os.path.abspath("dist")
self._GetInitScriptFileName()
self._GetBaseFileName()
if self.path is None:
self.path = sys.path
if self.appendScriptToLibrary:
self._VerifyCanAppendToLibrary()
for sourceFileName, targetFileName in self.includeFiles:
if not os.path.exists(sourceFileName):
raise ConfigError("cannot find file/directory named %s",
sourceFileName)
if os.path.isabs(targetFileName):
raise ConfigError("target file/directory cannot be absolute")
for executable in self.executables:
executable._VerifyConfiguration(self)
def _WriteModules(self, fileName, initScript, finder, compress,
copyDependentFiles, scriptModule = None):
initModule = finder.IncludeFile(initScript, "cx_Freeze__init__")
if scriptModule is None:
for module in self.constantsModules:
module.Create(finder)
modules = [m for m in finder.modules \
if m.name not in self.excludeModules]
else:
modules = [initModule, scriptModule]
self.excludeModules[initModule.name] = None
self.excludeModules[scriptModule.name] = None
itemsToSort = [(m.name, m) for m in modules]
itemsToSort.sort()
modules = [m for n, m in itemsToSort]
self._PrintReport(fileName, modules)
if scriptModule is None:
finder.ReportMissingModules()
targetDir = os.path.dirname(fileName)
self._CreateDirectory(targetDir)
filesToCopy = []
if os.path.exists(fileName):
mode = "a"
else:
mode = "w"
outFile = zipfile.PyZipFile(fileName, mode, zipfile.ZIP_DEFLATED)
for module in modules:
if module.code is None and module.file is not None:
fileName = os.path.basename(module.file)
baseFileName, ext = os.path.splitext(fileName)
if baseFileName != module.name and module.name != "zlib":
if "." in module.name:
fileName = module.name + ext
generatedFileName = "ExtensionLoader_%s.py" % \
module.name.replace(".", "_")
module.code = compile(EXTENSION_LOADER_SOURCE % fileName,
generatedFileName, "exec")
target = os.path.join(targetDir, fileName)
filesToCopy.append((module, target))
if module.code is None:
continue
fileName = "/".join(module.name.split("."))
if module.path:
fileName += "/__init__"
if module.file is not None and os.path.exists(module.file):
mtime = os.stat(module.file).st_mtime
else:
mtime = time.time()
zipTime = time.localtime(mtime)[:6]
data = imp.get_magic() + struct.pack("<i", mtime) + \
marshal.dumps(module.code)
zinfo = zipfile.ZipInfo(fileName + ".pyc", zipTime)
if compress:
zinfo.compress_type = zipfile.ZIP_DEFLATED
outFile.writestr(zinfo, data)
origPath = os.environ["PATH"]
for module, target in filesToCopy:
try:
if module.parent is not None:
path = os.pathsep.join([origPath] + module.parent.path)
os.environ["PATH"] = path
self._CopyFile(module.file, target, copyDependentFiles)
finally:
os.environ["PATH"] = origPath
def Freeze(self):
self.finder = None
self.excludeModules = {}
self.dependentFiles = {}
self.filesCopied = {}
cx_Freeze.util.SetOptimizeFlag(self.optimizeFlag)
if self.createLibraryZip:
self.finder = self._GetModuleFinder()
for executable in self.executables:
self._FreezeExecutable(executable)
if self.createLibraryZip:
fileName = os.path.join(self.targetDir, "library.zip")
self._RemoveFile(fileName)
self._WriteModules(fileName, self.initScript, self.finder,
self.compress, self.copyDependentFiles)
for sourceFileName, targetFileName in self.includeFiles:
fullName = os.path.join(self.targetDir, targetFileName)
if os.path.isdir(sourceFileName):
for path, dirNames, fileNames in os.walk(sourceFileName):
shortPath = path[len(sourceFileName) + 1:]
if ".svn" in dirNames:
dirNames.remove(".svn")
if "CVS" in dirNames:
dirNames.remove("CVS")
for fileName in fileNames:
fullSourceName = os.path.join(path, fileName)
fullTargetName = os.path.join(self.targetDir,
targetFileName, shortPath, fileName)
self._CopyFile(fullSourceName, fullTargetName,
copyDependentFiles = False)
else:
self._CopyFile(sourceFileName, fullName,
copyDependentFiles = False)
class ConfigError(Exception):
def __init__(self, format, *args):
self.what = format % args
def __str__(self):
return self.what
class Executable(object):
def __init__(self, script, initScript = None, base = None, path = None,
targetDir = None, targetName = None, includes = None,
excludes = None, packages = None, replacePaths = None,
compress = None, copyDependentFiles = None,
appendScriptToExe = None, appendScriptToLibrary = None,
icon = None):
self.script = script
self.initScript = initScript
self.base = base
self.path = path
self.targetDir = targetDir
self.targetName = targetName
self.includes = includes
self.excludes = excludes
self.packages = packages
self.replacePaths = replacePaths
self.compress = compress
self.copyDependentFiles = copyDependentFiles
self.appendScriptToExe = appendScriptToExe
self.appendScriptToLibrary = appendScriptToLibrary
self.icon = icon
def __repr__(self):
return "<Executable script=%s>" % self.script
def _VerifyConfiguration(self, freezer):
if self.path is None:
self.path = freezer.path
if self.targetDir is None:
self.targetDir = freezer.targetDir
if self.includes is None:
self.includes = freezer.includes
if self.excludes is None:
self.excludes = freezer.excludes
if self.packages is None:
self.packages = freezer.packages
if self.replacePaths is None:
self.replacePaths = freezer.replacePaths
if self.compress is None:
self.compress = freezer.compress
if self.copyDependentFiles is None:
self.copyDependentFiles = freezer.copyDependentFiles
if self.appendScriptToExe is None:
self.appendScriptToExe = freezer.appendScriptToExe
if self.appendScriptToLibrary is None:
self.appendScriptToLibrary = freezer.appendScriptToLibrary
if self.initScript is None:
self.initScript = freezer.initScript
else:
freezer._GetInitScriptFileName(self)
if self.base is None:
self.base = freezer.base
else:
freezer._GetBaseFileName(self)
if self.appendScriptToLibrary:
freezer._VerifyCanAppendToLibrary()
if self.icon is None:
self.icon = freezer.icon
if self.script is not None:
name, ext = os.path.splitext(os.path.basename(self.script))
if self.appendScriptToLibrary:
self.moduleName = "%s__main__" % os.path.normcase(name)
else:
self.moduleName = "__main__"
if self.targetName is None:
baseName, ext = os.path.splitext(self.base)
self.targetName = name + ext
self.targetName = os.path.join(self.targetDir, self.targetName)
class ConstantsModule(object):
def __init__(self, releaseString = None, copyright = None,
moduleName = "BUILD_CONSTANTS", timeFormat = "%B %d, %Y %H:%M:%S"):
self.moduleName = moduleName
self.timeFormat = timeFormat
self.values = {}
self.values["BUILD_RELEASE_STRING"] = releaseString
self.values["BUILD_COPYRIGHT"] = copyright
def Create(self, finder):
"""Create the module which consists of declaration statements for each
of the values."""
today = datetime.datetime.today()
sourceTimestamp = 0
for module in finder.modules:
if module.file is None:
continue
if module.inZipFile:
continue
if not os.path.exists(module.file):
raise ConfigError("no file named %s", module.file)
timestamp = os.stat(module.file).st_mtime
sourceTimestamp = max(sourceTimestamp, timestamp)
sourceTimestamp = datetime.datetime.fromtimestamp(sourceTimestamp)
self.values["BUILD_TIMESTAMP"] = today.strftime(self.timeFormat)
self.values["BUILD_HOST"] = socket.gethostname().split(".")[0]
self.values["SOURCE_TIMESTAMP"] = \
sourceTimestamp.strftime(self.timeFormat)
module = finder._AddModule(self.moduleName)
sourceParts = []
names = self.values.keys()
names.sort()
for name in names:
value = self.values[name]
sourceParts.append("%s = %r" % (name, value))
source = "\n".join(sourceParts)
module.code = compile(source, "%s.py" % self.moduleName, "exec")

View File

@ -0,0 +1,281 @@
import os
import sys
def initialize(finder):
"""upon initialization of the finder, this routine is called to set up some
automatic exclusions for various platforms."""
finder.ExcludeModule("FCNTL")
finder.ExcludeModule("os.path")
if os.name == "nt":
finder.ExcludeModule("fcntl")
finder.ExcludeModule("grp")
finder.ExcludeModule("pwd")
finder.ExcludeModule("termios")
else:
finder.ExcludeModule("_winreg")
finder.ExcludeModule("msilib")
finder.ExcludeModule("msvcrt")
finder.ExcludeModule("nt")
if os.name not in ("os2", "ce"):
finder.ExcludeModule("ntpath")
finder.ExcludeModule("nturl2path")
finder.ExcludeModule("pythoncom")
finder.ExcludeModule("pywintypes")
finder.ExcludeModule("winerror")
finder.ExcludeModule("winsound")
finder.ExcludeModule("win32api")
finder.ExcludeModule("win32con")
finder.ExcludeModule("win32event")
finder.ExcludeModule("win32file")
finder.ExcludeModule("win32pdh")
finder.ExcludeModule("win32pipe")
finder.ExcludeModule("win32process")
finder.ExcludeModule("win32security")
finder.ExcludeModule("win32service")
finder.ExcludeModule("wx.activex")
if os.name != "posix":
finder.ExcludeModule("posix")
if os.name != "mac":
finder.ExcludeModule("Carbon")
finder.ExcludeModule("gestalt")
finder.ExcludeModule("ic")
finder.ExcludeModule("mac")
finder.ExcludeModule("MacOS")
finder.ExcludeModule("macpath")
finder.ExcludeModule("macurl2path")
if os.name != "nt":
finder.ExcludeModule("EasyDialogs")
if os.name != "os2":
finder.ExcludeModule("os2")
finder.ExcludeModule("os2emxpath")
finder.ExcludeModule("_emx_link")
if os.name != "ce":
finder.ExcludeModule("ce")
if os.name != "riscos":
finder.ExcludeModule("riscos")
finder.ExcludeModule("riscosenviron")
finder.ExcludeModule("riscospath")
finder.ExcludeModule("rourl2path")
if sys.platform[:4] != "java":
finder.ExcludeModule("java.lang")
finder.ExcludeModule("org.python.core")
def load_cElementTree(finder, module):
"""the cElementTree module implicitly loads the elementtree.ElementTree
module; make sure this happens."""
finder.IncludeModule("elementtree.ElementTree")
def load_ceODBC(finder, module):
"""the ceODBC module implicitly imports both datetime and decimal; make
sure this happens."""
finder.IncludeModule("datetime")
finder.IncludeModule("decimal")
def load_cx_Oracle(finder, module):
"""the cx_Oracle module implicitly imports datetime; make sure this
happens."""
finder.IncludeModule("datetime")
def load_docutils_frontend(finder, module):
"""The optik module is the old name for the optparse module; ignore the
module if it cannot be found."""
module.IgnoreName("optik")
def load_dummy_threading(finder, module):
"""the dummy_threading module plays games with the name of the threading
module for its own purposes; ignore that here"""
finder.ExcludeModule("_dummy_threading")
def load_email(finder, module):
"""the email package has a bunch of aliases as the submodule names were
all changed to lowercase in Python 2.5; mimic that here."""
if sys.version_info[:2] >= (2, 5):
for name in ("Charset", "Encoders", "Errors", "FeedParser",
"Generator", "Header", "Iterators", "Message", "Parser",
"Utils", "base64MIME", "quopriMIME"):
finder.AddAlias("email.%s" % name, "email.%s" % name.lower())
def load_ftplib(finder, module):
"""the ftplib module attempts to import the SOCKS module; ignore this
module if it cannot be found"""
module.IgnoreName("SOCKS")
def load_matplotlib(finder, module):
"""the matplotlib module requires data to be found in mpl-data in the
same directory as the frozen executable so oblige it"""
dir = os.path.join(module.path[0], "mpl-data")
finder.IncludeFiles(dir, "mpl-data")
def load_matplotlib_numerix(finder, module):
"""the numpy.numerix module loads a number of modules dynamically"""
for name in ("ma", "fft", "linear_algebra", "random_array", "mlab"):
finder.IncludeModule("%s.%s" % (module.name, name))
def load_numpy_linalg(finder, module):
"""the numpy.linalg module implicitly loads the lapack_lite module; make
sure this happens"""
finder.IncludeModule("numpy.linalg.lapack_lite")
def load_pty(finder, module):
"""The sgi module is not needed for this module to function."""
module.IgnoreName("sgi")
def load_pythoncom(finder, module):
"""the pythoncom module is actually contained in a DLL but since those
cannot be loaded directly in Python 2.5 and higher a special module is
used to perform that task; simply use that technique directly to
determine the name of the DLL and ensure it is included as a normal
extension; also load the pywintypes module which is implicitly
loaded."""
import pythoncom
module.file = pythoncom.__file__
module.code = None
finder.IncludeModule("pywintypes")
def load_pywintypes(finder, module):
"""the pywintypes module is actually contained in a DLL but since those
cannot be loaded directly in Python 2.5 and higher a special module is
used to perform that task; simply use that technique directly to
determine the name of the DLL and ensure it is included as a normal
extension."""
import pywintypes
module.file = pywintypes.__file__
module.code = None
def load_PyQt4_Qt(finder, module):
"""the PyQt4.Qt module is an extension module which imports a number of
other modules and injects their namespace into its own. It seems a
foolish way of doing things but perhaps there is some hidden advantage
to this technique over pure Python; ignore the absence of some of
the modules since not every installation includes all of them."""
finder.IncludeModule("PyQt4.QtCore")
finder.IncludeModule("PyQt4.QtGui")
finder.IncludeModule("sip")
for name in ("PyQt4.QtSvg", "PyQt4.Qsci", "PyQt4.QtAssistant",
"PyQt4.QtNetwork", "PyQt4.QtOpenGL", "PyQt4.QtScript", "PyQt4._qt",
"PyQt4.QtSql", "PyQt4.QtSvg", "PyQt4.QtTest", "PyQt4.QtXml"):
try:
finder.IncludeModule(name)
except ImportError:
pass
def load_Tkinter(finder, module):
"""the Tkinter module has data files that are required to be loaded so
ensure that they are copied into the directory that is expected at
runtime."""
import Tkinter
import _tkinter
tk = _tkinter.create()
tclDir = os.path.dirname(tk.call("info", "library"))
tclSourceDir = os.path.join(tclDir, "tcl%s" % _tkinter.TCL_VERSION)
tkSourceDir = os.path.join(tclDir, "tk%s" % _tkinter.TK_VERSION)
finder.IncludeFiles(tclSourceDir, "tcl")
finder.IncludeFiles(tkSourceDir, "tk")
def load_tempfile(finder, module):
"""the tempfile module attempts to load the fcntl and thread modules but
continues if these modules cannot be found; ignore these modules if they
cannot be found."""
module.IgnoreName("fcntl")
module.IgnoreName("thread")
def load_time(finder, module):
"""the time module implicitly loads _strptime; make sure this happens."""
finder.IncludeModule("_strptime")
def load_win32api(finder, module):
"""the win32api module implicitly loads the pywintypes module; make sure
this happens."""
finder.IncludeModule("pywintypes")
def load_win32com(finder, module):
"""the win32com package manipulates its search path at runtime to include
the sibling directory called win32comext; simulate that by changing the
search path in a similar fashion here."""
baseDir = os.path.dirname(os.path.dirname(module.file))
module.path.append(os.path.join(baseDir, "win32comext"))
def load_win32file(finder, module):
"""the win32api module implicitly loads the pywintypes module; make sure
this happens."""
finder.IncludeModule("pywintypes")
def load_xml(finder, module):
"""the builtin xml package attempts to load the _xmlplus module to see if
that module should take its role instead; ignore the failure to find
this module, though."""
module.IgnoreName("_xmlplus")
def load_xml_etree_cElementTree(finder, module):
"""the xml.etree.cElementTree module implicitly loads the
xml.etree.ElementTree module; make sure this happens."""
finder.IncludeModule("xml.etree.ElementTree")
def load_IPython(finder, module):
ipy = os.path.join(os.path.dirname(module.file), 'Extensions')
extensions = set([])
for m in os.listdir(ipy):
extensions.add(os.path.splitext(m)[0])
extensions.remove('__init__')
for m in extensions:
finder.IncludeModule('IPython.Extensions.'+m)
def load_lxml(finder, module):
finder.IncludeModule('lxml._elementpath')
def load_cherrypy(finder, module):
finder.IncludeModule('cherrypy.lib.encoding')
def missing_cElementTree(finder, caller):
"""the cElementTree has been incorporated into the standard library in
Python 2.5 so ignore its absence if it cannot found."""
if sys.version_info[:2] >= (2, 5):
caller.IgnoreName("cElementTree")
def missing_EasyDialogs(finder, caller):
"""the EasyDialogs module is not normally present on Windows but it also
may be so instead of excluding it completely, ignore it if it can't be
found"""
if sys.platform == "win32":
caller.IgnoreName("EasyDialogs")
def missing_readline(finder, caller):
"""the readline module is not normally present on Windows but it also may
be so instead of excluding it completely, ignore it if it can't be
found"""
if sys.platform == "win32":
caller.IgnoreName("readline")
def missing_xml_etree(finder, caller):
"""the xml.etree package is new for Python 2.5 but it is common practice
to use a try..except.. block in order to support versions earlier than
Python 2.5 transparently; ignore the absence of the package in this
situation."""
if sys.version_info[:2] < (2, 5):
caller.IgnoreName("xml.etree")

View File

@ -0,0 +1,171 @@
import optparse
import os
import shutil
import stat
import sys
import cx_Freeze
__all__ = ["main"]
USAGE = \
"""
%prog [options] [SCRIPT]
Freeze a Python script and all of its referenced modules to a base
executable which can then be distributed without requiring a Python
installation."""
VERSION = \
"""
%%prog %s
Copyright (c) 2007-2008 Colt Engineering. All rights reserved.
Copyright (c) 2001-2006 Computronix Corporation. All rights reserved.""" % \
cx_Freeze.version
def ParseCommandLine():
parser = optparse.OptionParser(version = VERSION.strip(),
usage = USAGE.strip())
parser.add_option("-O",
action = "count",
default = 0,
dest = "optimized",
help = "optimize generated bytecode as per PYTHONOPTIMIZE; "
"use -OO in order to remove doc strings")
parser.add_option("-c", "--compress",
action = "store_true",
dest = "compress",
help = "compress byte code in zip files")
parser.add_option("--base-name",
dest = "baseName",
metavar = "NAME",
help = "file on which to base the target file; if the name of the "
"file is not an absolute file name, the subdirectory bases "
"(rooted in the directory in which the freezer is found) "
"will be searched for a file matching the name")
parser.add_option("--init-script",
dest = "initScript",
metavar = "NAME",
help = "script which will be executed upon startup; if the name "
"of the file is not an absolute file name, the "
"subdirectory initscripts (rooted in the directory in "
"which the cx_Freeze package is found) will be searched "
"for a file matching the name")
parser.add_option("--target-dir", "--install-dir",
dest = "targetDir",
metavar = "DIR",
help = "the directory in which to place the target file and "
"any dependent files")
parser.add_option("--target-name",
dest = "targetName",
metavar = "NAME",
help = "the name of the file to create instead of the base name "
"of the script and the extension of the base binary")
parser.add_option("--no-copy-deps",
dest = "copyDeps",
default = True,
action = "store_false",
help = "do not copy the dependent files (extensions, shared "
"libraries, etc.) to the target directory; this also "
"modifies the default init script to ConsoleKeepPath.py "
"and means that the target executable requires a Python "
"installation to execute properly")
parser.add_option("--default-path",
action = "append",
dest = "defaultPath",
metavar = "DIRS",
help = "list of paths separated by the standard path separator "
"for the platform which will be used to initialize "
"sys.path prior to running the module finder")
parser.add_option("--include-path",
action = "append",
dest = "includePath",
metavar = "DIRS",
help = "list of paths separated by the standard path separator "
"for the platform which will be used to modify sys.path "
"prior to running the module finder")
parser.add_option("--replace-paths",
dest = "replacePaths",
metavar = "DIRECTIVES",
help = "replace all the paths in modules found in the given paths "
"with the given replacement string; multiple values are "
"separated by the standard path separator and each value "
"is of the form path=replacement_string; path can be * "
"which means all paths not already specified")
parser.add_option("--include-modules",
dest = "includeModules",
metavar = "NAMES",
help = "comma separated list of modules to include")
parser.add_option("--exclude-modules",
dest = "excludeModules",
metavar = "NAMES",
help = "comma separated list of modules to exclude")
parser.add_option("--ext-list-file",
dest = "extListFile",
metavar = "NAME",
help = "name of file in which to place the list of dependent files "
"which were copied into the target directory")
parser.add_option("-z", "--zip-include",
dest = "zipIncludes",
action = "append",
default = [],
metavar = "SPEC",
help = "name of file to add to the zip file or a specification of "
"the form name=arcname which will specify the archive name "
"to use; multiple --zip-include arguments can be used")
options, args = parser.parse_args()
if len(args) == 0:
options.script = None
elif len(args) == 1:
options.script, = args
else:
parser.error("only one script can be specified")
if not args and options.includeModules is None and options.copyDeps:
parser.error("script or a list of modules must be specified")
if not args and options.targetName is None:
parser.error("script or a target name must be specified")
if options.excludeModules:
options.excludeModules = options.excludeModules.split(",")
else:
options.excludeModules = []
if options.includeModules:
options.includeModules = options.includeModules.split(",")
else:
options.includeModules = []
replacePaths = []
if options.replacePaths:
for directive in options.replacePaths.split(os.pathsep):
fromPath, replacement = directive.split("=")
replacePaths.append((fromPath, replacement))
options.replacePaths = replacePaths
if options.defaultPath is not None:
sys.path = [p for mp in options.defaultPath \
for p in mp.split(os.pathsep)]
if options.includePath is not None:
paths = [p for mp in options.includePath for p in mp.split(os.pathsep)]
sys.path = paths + sys.path
if options.script is not None:
sys.path.insert(0, os.path.dirname(options.script))
return options
def main():
options = ParseCommandLine()
executables = [cx_Freeze.Executable(options.script,
targetName = options.targetName)]
freezer = cx_Freeze.Freezer(executables,
includes = options.includeModules,
excludes = options.excludeModules,
replacePaths = options.replacePaths,
compress = options.compress,
optimizeFlag = options.optimized,
copyDependentFiles = options.copyDeps,
initScript = options.initScript,
base = options.baseName,
path = None,
createLibraryZip = False,
appendScriptToExe = True,
targetDir = options.targetDir)
freezer.Freeze()

View File

@ -0,0 +1,337 @@
import distutils.command.bdist_msi
import msilib
import os
__all__ = [ "bdist_msi" ]
# force the remove existing products action to happen first since Windows
# installer appears to be braindead and doesn't handle files shared between
# different "products" very well
sequence = msilib.sequence.InstallExecuteSequence
for index, info in enumerate(sequence):
if info[0] == u'RemoveExistingProducts':
sequence[index] = (info[0], info[1], 1450)
class bdist_msi(distutils.command.bdist_msi.bdist_msi):
user_options = distutils.command.bdist_msi.bdist_msi.user_options + [
('add-to-path=', None, 'add target dir to PATH environment variable'),
('upgrade-code=', None, 'upgrade code to use')
]
x = y = 50
width = 370
height = 300
title = "[ProductName] Setup"
modeless = 1
modal = 3
def add_config(self, fullname):
initialTargetDir = self.get_initial_target_dir(fullname)
if self.add_to_path is None:
self.add_to_path = False
for executable in self.distribution.executables:
if os.path.basename(executable.base).startswith("Console"):
self.add_to_path = True
break
if self.add_to_path:
msilib.add_data(self.db, 'Environment',
[("E_PATH", "Path", r"[~];[TARGETDIR]", "TARGETDIR")])
msilib.add_data(self.db, 'CustomAction',
[("InitialTargetDir", 256 + 51, "TARGETDIR", initialTargetDir)
])
msilib.add_data(self.db, 'InstallExecuteSequence',
[("InitialTargetDir", 'TARGETDIR=""', 401)])
msilib.add_data(self.db, 'InstallUISequence',
[("PrepareDlg", None, 140),
("InitialTargetDir", 'TARGETDIR=""', 401),
("SelectDirectoryDlg", "not Installed", 1230),
("MaintenanceTypeDlg",
"Installed and not Resume and not Preselected", 1250),
("ProgressDlg", None, 1280)
])
def add_cancel_dialog(self):
dialog = msilib.Dialog(self.db, "CancelDlg", 50, 10, 260, 85, 3,
self.title, "No", "No", "No")
dialog.text("Text", 48, 15, 194, 30, 3,
"Are you sure you want to cancel [ProductName] installation?")
button = dialog.pushbutton("Yes", 72, 57, 56, 17, 3, "Yes", "No")
button.event("EndDialog", "Exit")
button = dialog.pushbutton("No", 132, 57, 56, 17, 3, "No", "Yes")
button.event("EndDialog", "Return")
def add_error_dialog(self):
dialog = msilib.Dialog(self.db, "ErrorDlg", 50, 10, 330, 101, 65543,
self.title, "ErrorText", None, None)
dialog.text("ErrorText", 50, 9, 280, 48, 3, "")
for text, x in [("No", 120), ("Yes", 240), ("Abort", 0),
("Cancel", 42), ("Ignore", 81), ("Ok", 159), ("Retry", 198)]:
button = dialog.pushbutton(text[0], x, 72, 81, 21, 3, text, None)
button.event("EndDialog", "Error%s" % text)
def add_exit_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "ExitDialog",
self.x, self.y, self.width, self.height, self.modal,
self.title, "Finish", "Finish", "Finish")
dialog.title("Completing the [ProductName] installer")
dialog.back("< Back", "Finish", active = False)
dialog.cancel("Cancel", "Back", active = False)
dialog.text("Description", 15, 235, 320, 20, 0x30003,
"Click the Finish button to exit the installer.")
button = dialog.next("Finish", "Cancel", name = "Finish")
button.event("EndDialog", "Return")
def add_fatal_error_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "FatalError",
self.x, self.y, self.width, self.height, self.modal,
self.title, "Finish", "Finish", "Finish")
dialog.title("[ProductName] installer ended prematurely")
dialog.back("< Back", "Finish", active = False)
dialog.cancel("Cancel", "Back", active = False)
dialog.text("Description1", 15, 70, 320, 80, 0x30003,
"[ProductName] setup ended prematurely because of an error. "
"Your system has not been modified. To install this program "
"at a later time, please run the installation again.")
dialog.text("Description2", 15, 155, 320, 20, 0x30003,
"Click the Finish button to exit the installer.")
button = dialog.next("Finish", "Cancel", name = "Finish")
button.event("EndDialog", "Exit")
def add_files_in_use_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "FilesInUse",
self.x, self.y, self.width, self.height, 19, self.title,
"Retry", "Retry", "Retry", bitmap = False)
dialog.text("Title", 15, 6, 200, 15, 0x30003,
r"{\DlgFontBold8}Files in Use")
dialog.text("Description", 20, 23, 280, 20, 0x30003,
"Some files that need to be updated are currently in use.")
dialog.text("Text", 20, 55, 330, 50, 3,
"The following applications are using files that need to be "
"updated by this setup. Close these applications and then "
"click Retry to continue the installation or Cancel to exit "
"it.")
dialog.control("List", "ListBox", 20, 107, 330, 130, 7,
"FileInUseProcess", None, None, None)
button = dialog.back("Exit", "Ignore", name = "Exit")
button.event("EndDialog", "Exit")
button = dialog.next("Ignore", "Retry", name = "Ignore")
button.event("EndDialog", "Ignore")
button = dialog.cancel("Retry", "Exit", name = "Retry")
button.event("EndDialog", "Retry")
def add_maintenance_type_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db,
"MaintenanceTypeDlg", self.x, self.y, self.width, self.height,
self.modal, self.title, "Next", "Next", "Cancel")
dialog.title("Welcome to the [ProductName] Setup Wizard")
dialog.text("BodyText", 15, 63, 330, 42, 3,
"Select whether you want to repair or remove [ProductName].")
group = dialog.radiogroup("RepairRadioGroup", 15, 108, 330, 60, 3,
"MaintenanceForm_Action", "", "Next")
group.add("Repair", 0, 18, 300, 17, "&Repair [ProductName]")
group.add("Remove", 0, 36, 300, 17, "Re&move [ProductName]")
dialog.back("< Back", None, active = False)
button = dialog.next("Finish", "Cancel")
button.event("[REINSTALL]", "ALL",
'MaintenanceForm_Action="Repair"', 5)
button.event("[Progress1]", "Repairing",
'MaintenanceForm_Action="Repair"', 6)
button.event("[Progress2]", "repairs",
'MaintenanceForm_Action="Repair"', 7)
button.event("Reinstall", "ALL",
'MaintenanceForm_Action="Repair"', 8)
button.event("[REMOVE]", "ALL",
'MaintenanceForm_Action="Remove"', 11)
button.event("[Progress1]", "Removing",
'MaintenanceForm_Action="Remove"', 12)
button.event("[Progress2]", "removes",
'MaintenanceForm_Action="Remove"', 13)
button.event("Remove", "ALL",
'MaintenanceForm_Action="Remove"', 14)
button.event("EndDialog", "Return",
'MaintenanceForm_Action<>"Change"', 20)
button = dialog.cancel("Cancel", "RepairRadioGroup")
button.event("SpawnDialog", "CancelDlg")
def add_prepare_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "PrepareDlg",
self.x, self.y, self.width, self.height, self.modeless,
self.title, "Cancel", "Cancel", "Cancel")
dialog.text("Description", 15, 70, 320, 40, 0x30003,
"Please wait while the installer prepares to guide you through"
"the installation.")
dialog.title("Welcome to the [ProductName] installer")
text = dialog.text("ActionText", 15, 110, 320, 20, 0x30003,
"Pondering...")
text.mapping("ActionText", "Text")
text = dialog.text("ActionData", 15, 135, 320, 30, 0x30003, None)
text.mapping("ActionData", "Text")
dialog.back("Back", None, active = False)
dialog.next("Next", None, active = False)
button = dialog.cancel("Cancel", None)
button.event("SpawnDialog", "CancelDlg")
def add_progress_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "ProgressDlg",
self.x, self.y, self.width, self.height, self.modeless,
self.title, "Cancel", "Cancel", "Cancel", bitmap = False)
dialog.text("Title", 20, 15, 200, 15, 0x30003,
r"{\DlgFontBold8}[Progress1] [ProductName]")
dialog.text("Text", 35, 65, 300, 30, 3,
"Please wait while the installer [Progress2] [ProductName].")
dialog.text("StatusLabel", 35, 100 ,35, 20, 3, "Status:")
text = dialog.text("ActionText", 70, 100, self.width - 70, 20, 3,
"Pondering...")
text.mapping("ActionText", "Text")
control = dialog.control("ProgressBar", "ProgressBar", 35, 120, 300,
10, 65537, None, "Progress done", None, None)
control.mapping("SetProgress", "Progress")
dialog.back("< Back", "Next", active = False)
dialog.next("Next >", "Cancel", active = False)
button = dialog.cancel("Cancel", "Back")
button.event("SpawnDialog", "CancelDlg")
def add_properties(self):
metadata = self.distribution.metadata
props = [
('DistVersion', metadata.get_version()),
('DefaultUIFont', 'DlgFont8'),
('ErrorDialog', 'ErrorDlg'),
('Progress1', 'Install'),
('Progress2', 'installs'),
('MaintenanceForm_Action', 'Repair')
]
email = metadata.author_email or metadata.maintainer_email
if email:
props.append(("ARPCONTACT", email))
if metadata.url:
props.append(("ARPURLINFOABOUT", metadata.url))
if self.upgrade_code is not None:
props.append(("UpgradeCode", self.upgrade_code))
msilib.add_data(self.db, 'Property', props)
def add_select_directory_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db,
"SelectDirectoryDlg", self.x, self.y, self.width, self.height,
self.modal, self.title, "Next", "Next", "Cancel")
dialog.title("Select destination directory")
dialog.back("< Back", None, active = False)
button = dialog.next("Next >", "Cancel")
button.event("SetTargetPath", "TARGETDIR", ordering = 1)
button.event("SpawnWaitDialog", "WaitForCostingDlg", ordering = 2)
button.event("EndDialog", "Return", ordering = 3)
button = dialog.cancel("Cancel", "DirectoryCombo")
button.event("SpawnDialog", "CancelDlg")
dialog.control("DirectoryCombo", "DirectoryCombo", 15, 70, 272, 80,
393219, "TARGETDIR", None, "DirectoryList", None)
dialog.control("DirectoryList", "DirectoryList", 15, 90, 308, 136, 3,
"TARGETDIR", None, "PathEdit", None)
dialog.control("PathEdit", "PathEdit", 15, 230, 306, 16, 3,
"TARGETDIR", None, "Next", None)
button = dialog.pushbutton("Up", 306, 70, 18, 18, 3, "Up", None)
button.event("DirectoryListUp", "0")
button = dialog.pushbutton("NewDir", 324, 70, 30, 18, 3, "New", None)
button.event("DirectoryListNew", "0")
def add_text_styles(self):
msilib.add_data(self.db, 'TextStyle',
[("DlgFont8", "Tahoma", 9, None, 0),
("DlgFontBold8", "Tahoma", 8, None, 1),
("VerdanaBold10", "Verdana", 10, None, 1),
("VerdanaRed9", "Verdana", 9, 255, 0)
])
def add_ui(self):
self.add_text_styles()
self.add_error_dialog()
self.add_fatal_error_dialog()
self.add_cancel_dialog()
self.add_exit_dialog()
self.add_user_exit_dialog()
self.add_files_in_use_dialog()
self.add_wait_for_costing_dialog()
self.add_prepare_dialog()
self.add_select_directory_dialog()
self.add_progress_dialog()
self.add_maintenance_type_dialog()
def add_upgrade_config(self, sversion):
if self.upgrade_code is not None:
msilib.add_data(self.db, 'Upgrade',
[(self.upgrade_code, None, sversion, None, 513, None,
"REMOVEOLDVERSION"),
(self.upgrade_code, sversion, None, None, 257, None,
"REMOVENEWVERSION")
])
def add_user_exit_dialog(self):
dialog = distutils.command.bdist_msi.PyDialog(self.db, "UserExit",
self.x, self.y, self.width, self.height, self.modal,
self.title, "Finish", "Finish", "Finish")
dialog.title("[ProductName] installer was interrupted")
dialog.back("< Back", "Finish", active = False)
dialog.cancel("Cancel", "Back", active = False)
dialog.text("Description1", 15, 70, 320, 80, 0x30003,
"[ProductName] setup was interrupted. Your system has not "
"been modified. To install this program at a later time, "
"please run the installation again.")
dialog.text("Description2", 15, 155, 320, 20, 0x30003,
"Click the Finish button to exit the installer.")
button = dialog.next("Finish", "Cancel", name = "Finish")
button.event("EndDialog", "Exit")
def add_wait_for_costing_dialog(self):
dialog = msilib.Dialog(self.db, "WaitForCostingDlg", 50, 10, 260, 85,
self.modal, self.title, "Return", "Return", "Return")
dialog.text("Text", 48, 15, 194, 30, 3,
"Please wait while the installer finishes determining your "
"disk space requirements.")
button = dialog.pushbutton("Return", 102, 57, 56, 17, 3, "Return",
None)
button.event("EndDialog", "Exit")
def get_initial_target_dir(self, fullname):
return r"[ProgramFilesFolder]\%s" % fullname
def get_installer_filename(self, fullname):
return os.path.join(self.dist_dir, "%s.msi" % fullname)
def initialize_options(self):
distutils.command.bdist_msi.bdist_msi.initialize_options(self)
self.upgrade_code = None
self.add_to_path = None
def run(self):
if not self.skip_build:
self.run_command('build')
install = self.reinitialize_command('install', reinit_subcommands = 1)
install.prefix = self.bdist_dir
install.skip_build = self.skip_build
install.warn_dir = 0
distutils.log.info("installing to %s", self.bdist_dir)
install.ensure_finalized()
install.run()
self.mkpath(self.dist_dir)
fullname = self.distribution.get_fullname()
filename = os.path.abspath(self.get_installer_filename(fullname))
if os.path.exists(filename):
os.unlink(filename)
metadata = self.distribution.metadata
author = metadata.author or metadata.maintainer or "UNKNOWN"
version = metadata.get_version()
sversion = "%d.%d.%d" % \
distutils.version.StrictVersion(version).version
self.db = msilib.init_database(filename, msilib.schema,
self.distribution.metadata.name, msilib.gen_uuid(), sversion,
author)
msilib.add_tables(self.db, msilib.sequence)
self.add_properties()
self.add_config(fullname)
self.add_upgrade_config(sversion)
self.add_ui()
self.add_files()
self.db.Commit()
if not self.keep_temp:
distutils.dir_util.remove_tree(self.bdist_dir,
dry_run = self.dry_run)

6
installer/cx_Freeze/cxfreeze Executable file
View File

@ -0,0 +1,6 @@
#!/usr/bin/python
from cx_Freeze import main
main()

View File

@ -0,0 +1,35 @@
#------------------------------------------------------------------------------
# Console.py
# Initialization script for cx_Freeze which manipulates the path so that the
# directory in which the executable is found is searched for extensions but
# no other directory is searched. It also sets the attribute sys.frozen so that
# the Win32 extensions behave as expected.
#------------------------------------------------------------------------------
import encodings
import os
import sys
import warnings
import zipimport
sys.frozen = True
sys.path = sys.path[:4]
os.environ["TCL_LIBRARY"] = os.path.join(DIR_NAME, "tcl")
os.environ["TK_LIBRARY"] = os.path.join(DIR_NAME, "tk")
m = __import__("__main__")
importer = zipimport.zipimporter(INITSCRIPT_ZIP_FILE_NAME)
if INITSCRIPT_ZIP_FILE_NAME != SHARED_ZIP_FILE_NAME:
moduleName = m.__name__
else:
name, ext = os.path.splitext(os.path.basename(os.path.normcase(FILE_NAME)))
moduleName = "%s__main__" % name
code = importer.get_code(moduleName)
exec code in m.__dict__
if sys.version_info[:2] >= (2, 5):
module = sys.modules.get("threading")
if module is not None:
module._shutdown()

View File

@ -0,0 +1,19 @@
#------------------------------------------------------------------------------
# ConsoleKeepPath.py
# Initialization script for cx_Freeze which leaves the path alone and does
# not set the sys.frozen attribute.
#------------------------------------------------------------------------------
import sys
import zipimport
m = __import__("__main__")
importer = zipimport.zipimporter(INITSCRIPT_ZIP_FILE_NAME)
code = importer.get_code(m.__name__)
exec code in m.__dict__
if sys.version_info[:2] >= (2, 5):
module = sys.modules.get("threading")
if module is not None:
module._shutdown()

View File

@ -0,0 +1,38 @@
#------------------------------------------------------------------------------
# ConsoleSetLibPath.py
# Initialization script for cx_Freeze which manipulates the path so that the
# directory in which the executable is found is searched for extensions but
# no other directory is searched. The environment variable LD_LIBRARY_PATH is
# manipulated first, however, to ensure that shared libraries found in the
# target directory are found. This requires a restart of the executable because
# the environment variable LD_LIBRARY_PATH is only checked at startup.
#------------------------------------------------------------------------------
import encodings
import os
import sys
import warnings
import zipimport
paths = os.environ.get("LD_LIBRARY_PATH", "").split(os.pathsep)
if DIR_NAME not in paths:
paths.insert(0, DIR_NAME)
os.environ["LD_LIBRARY_PATH"] = os.pathsep.join(paths)
os.execv(sys.executable, sys.argv)
sys.frozen = True
sys.path = sys.path[:4]
os.environ["TCL_LIBRARY"] = os.path.join(DIR_NAME, "tcl")
os.environ["TK_LIBRARY"] = os.path.join(DIR_NAME, "tk")
m = __import__("__main__")
importer = zipimport.zipimporter(INITSCRIPT_ZIP_FILE_NAME)
code = importer.get_code(m.__name__)
exec code in m.__dict__
if sys.version_info[:2] >= (2, 5):
module = sys.modules.get("threading")
if module is not None:
module._shutdown()

View File

@ -0,0 +1,20 @@
#------------------------------------------------------------------------------
# SharedLib.py
# Initialization script for cx_Freeze which behaves similarly to the one for
# console based applications but must handle the case where Python has already
# been initialized and another DLL of this kind has been loaded. As such it
# does not block the path unless sys.frozen is not already set.
#------------------------------------------------------------------------------
import encodings
import os
import sys
import warnings
if not hasattr(sys, "frozen"):
sys.frozen = True
sys.path = sys.path[:4]
os.environ["TCL_LIBRARY"] = os.path.join(DIR_NAME, "tcl")
os.environ["TK_LIBRARY"] = os.path.join(DIR_NAME, "tk")

View File

@ -0,0 +1,23 @@
#------------------------------------------------------------------------------
# SharedLibSource.py
# Initialization script for cx_Freeze which imports the site module (as per
# normal processing of a Python script) and then searches for a file with the
# same name as the shared library but with the extension .pth. The entries in
# this file are used to modify the path to use for subsequent imports.
#------------------------------------------------------------------------------
import os
import sys
import warnings
# the site module must be imported for normal behavior to take place; it is
# done dynamically so that cx_Freeze will not add all modules referenced by
# the site module to the frozen executable
__import__("site")
# now locate the pth file to modify the path appropriately
baseName, ext = os.path.splitext(FILE_NAME)
pathFileName = baseName + ".pth"
sys.path = [s.strip() for s in file(pathFileName).read().splitlines()] + \
sys.path

View File

@ -0,0 +1,7 @@
import sys
print "Hello from cx_Freeze Advanced #1"
print
module = __import__("testfreeze_1")

View File

@ -0,0 +1,7 @@
import sys
print "Hello from cx_Freeze Advanced #2"
print
module = __import__("testfreeze_2")

View File

@ -0,0 +1 @@
print "Test freeze module #1"

View File

@ -0,0 +1 @@
print "Test freeze module #2"

View File

@ -0,0 +1,31 @@
# An advanced setup script to create multiple executables and demonstrate a few
# of the features available to setup scripts
#
# hello.py is a very simple "Hello, world" type script which also displays the
# environment in which the script runs
#
# Run the build process by running the command 'python setup.py build'
#
# If everything works well you should find a subdirectory in the build
# subdirectory that contains the files needed to run the script without Python
import sys
from cx_Freeze import setup, Executable
executables = [
Executable("advanced_1.py"),
Executable("advanced_2.py")
]
buildOptions = dict(
compressed = True,
includes = ["testfreeze_1", "testfreeze_2"],
path = sys.path + ["modules"])
setup(
name = "advanced_cx_Freeze_sample",
version = "0.1",
description = "Advanced sample cx_Freeze script",
options = dict(build_exe = buildOptions),
executables = executables)

View File

@ -0,0 +1,27 @@
# A simple setup script to create an executable using matplotlib.
#
# test_matplotlib.py is a very simple matplotlib application that demonstrates
# its use.
#
# Run the build process by running the command 'python setup.py build'
#
# If everything works well you should find a subdirectory in the build
# subdirectory that contains the files needed to run the application
import cx_Freeze
import sys
base = None
if sys.platform == "win32":
base = "Win32GUI"
executables = [
cx_Freeze.Executable("test_matplotlib.py", base = base)
]
cx_Freeze.setup(
name = "test_matplotlib",
version = "0.1",
description = "Sample matplotlib script",
executables = executables)

View File

@ -0,0 +1,48 @@
from numpy import arange, sin, pi
import matplotlib
matplotlib.use('WXAgg')
from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as FigureCanvas
from matplotlib.backends.backend_wx import NavigationToolbar2Wx
from matplotlib.figure import Figure
from wx import *
class CanvasFrame(Frame):
def __init__(self):
Frame.__init__(self,None,-1, 'CanvasFrame',size=(550,350))
self.SetBackgroundColour(NamedColor("WHITE"))
self.figure = Figure()
self.axes = self.figure.add_subplot(111)
t = arange(0.0,3.0,0.01)
s = sin(2*pi*t)
self.axes.plot(t,s)
self.canvas = FigureCanvas(self, -1, self.figure)
self.sizer = BoxSizer(VERTICAL)
self.sizer.Add(self.canvas, 1, LEFT | TOP | GROW)
self.SetSizerAndFit(self.sizer)
self.add_toolbar()
def add_toolbar(self):
self.toolbar = NavigationToolbar2Wx(self.canvas)
self.toolbar.Realize()
if Platform == '__WXMAC__':
self.SetToolBar(self.toolbar)
else:
tw, th = self.toolbar.GetSizeTuple()
fw, fh = self.canvas.GetSizeTuple()
self.toolbar.SetSize(Size(fw, th))
self.sizer.Add(self.toolbar, 0, LEFT | EXPAND)
self.toolbar.update()
def OnPaint(self, event):
self.canvas.draw()
class App(App):
def OnInit(self):
'Create the main window and insert the custom frame'
frame = CanvasFrame()
frame.Show(True)
return True
app = App(0)
app.MainLoop()

View File

@ -0,0 +1,3 @@
print "importing pkg1"
from . import sub1
from . import pkg2

View File

@ -0,0 +1,3 @@
print "importing pkg1.pkg2"
from . import sub3
from .. import sub4

View File

@ -0,0 +1,3 @@
print "importing pkg1.pkg2.sub3"
from . import sub5
from .. import sub6

View File

@ -0,0 +1 @@
print "importing pkg1.pkg2.sub5"

View File

@ -0,0 +1,2 @@
print "importing pkg1.sub1"
from . import sub2

View File

@ -0,0 +1 @@
print "importing pkg1.sub2"

View File

@ -0,0 +1 @@
print 'importing pkg1.sub4'

View File

@ -0,0 +1 @@
print "importing pkg1.sub6"

View File

@ -0,0 +1 @@
import pkg1

View File

@ -0,0 +1,16 @@
# relimport.py is a very simple script that tests importing using relative
# imports (available in Python 2.5 and up)
#
# Run the build process by running the command 'python setup.py build'
#
# If everything works well you should find a subdirectory in the build
# subdirectory that contains the files needed to run the script without Python
from cx_Freeze import setup, Executable
setup(
name = "relimport",
version = "0.1",
description = "Sample cx_Freeze script for relative imports",
executables = [Executable("relimport.py")])

View File

@ -0,0 +1,19 @@
import sys
print "Hello from cx_Freeze"
print
print "sys.executable", sys.executable
print "sys.prefix", sys.prefix
print
print "ARGUMENTS:"
for a in sys.argv:
print a
print
print "PATH:"
for p in sys.path:
print p
print

View File

@ -0,0 +1,18 @@
# A very simple setup script to create a single executable
#
# hello.py is a very simple "Hello, world" type script which also displays the
# environment in which the script runs
#
# Run the build process by running the command 'python setup.py build'
#
# If everything works well you should find a subdirectory in the build
# subdirectory that contains the files needed to run the script without Python
from cx_Freeze import setup, Executable
setup(
name = "hello",
version = "0.1",
description = "Sample cx_Freeze script",
executables = [Executable("hello.py")])

View File

@ -0,0 +1,25 @@
# A simple setup script to create an executable running wxPython. This also
# demonstrates the method for creating a Windows executable that does not have
# an associated console.
#
# wxapp.py is a very simple "Hello, world" type wxPython application
#
# Run the build process by running the command 'python setup.py build'
#
# If everything works well you should find a subdirectory in the build
# subdirectory that contains the files needed to run the application
import sys
from cx_Freeze import setup, Executable
base = None
if sys.platform == "win32":
base = "Win32GUI"
setup(
name = "hello",
version = "0.1",
description = "Sample cx_Freeze wxPython script",
executables = [Executable("wxapp.py", base = base)])

View File

@ -0,0 +1,42 @@
import wx
class Frame(wx.Frame):
def __init__(self):
wx.Frame.__init__(self, parent = None, title = "Hello from cx_Freeze")
panel = wx.Panel(self)
closeMeButton = wx.Button(panel, -1, "Close Me")
wx.EVT_BUTTON(self, closeMeButton.GetId(), self.OnCloseMe)
wx.EVT_CLOSE(self, self.OnCloseWindow)
pushMeButton = wx.Button(panel, -1, "Push Me")
wx.EVT_BUTTON(self, pushMeButton.GetId(), self.OnPushMe)
sizer = wx.BoxSizer(wx.HORIZONTAL)
sizer.Add(closeMeButton, flag = wx.ALL, border = 20)
sizer.Add(pushMeButton, flag = wx.ALL, border = 20)
panel.SetSizer(sizer)
topSizer = wx.BoxSizer(wx.VERTICAL)
topSizer.Add(panel, flag = wx.ALL | wx.EXPAND)
topSizer.Fit(self)
def OnCloseMe(self, event):
self.Close(True)
def OnPushMe(self, event):
1 / 0
def OnCloseWindow(self, event):
self.Destroy()
class App(wx.App):
def OnInit(self):
frame = Frame()
frame.Show(True)
self.SetTopWindow(frame)
return True
app = App(1)
app.MainLoop()

197
installer/cx_Freeze/setup.py Executable file
View File

@ -0,0 +1,197 @@
"""
Distutils script for cx_Freeze.
"""
import distutils.command.bdist_rpm
import distutils.command.build_ext
import distutils.command.build_scripts
import distutils.command.install
import distutils.command.install_data
import distutils.sysconfig
import os
import sys
from distutils.core import setup
from distutils.extension import Extension
class bdist_rpm(distutils.command.bdist_rpm.bdist_rpm):
# rpm automatically byte compiles all Python files in a package but we
# don't want that to happen for initscripts and samples so we tell it to
# ignore those files
def _make_spec_file(self):
specFile = distutils.command.bdist_rpm.bdist_rpm._make_spec_file(self)
specFile.insert(0, "%define _unpackaged_files_terminate_build 0%{nil}")
return specFile
def run(self):
distutils.command.bdist_rpm.bdist_rpm.run(self)
specFile = os.path.join(self.rpm_base, "SPECS",
"%s.spec" % self.distribution.get_name())
queryFormat = "%{name}-%{version}-%{release}.%{arch}.rpm"
command = "rpm -q --qf '%s' --specfile %s" % (queryFormat, specFile)
origFileName = os.popen(command).read()
parts = origFileName.split("-")
parts.insert(2, "py%s%s" % sys.version_info[:2])
newFileName = "-".join(parts)
self.move_file(os.path.join("dist", origFileName),
os.path.join("dist", newFileName))
class build_ext(distutils.command.build_ext.build_ext):
def build_extension(self, ext):
if ext.name.find("bases") < 0:
distutils.command.build_ext.build_ext.build_extension(self, ext)
return
os.environ["LD_RUN_PATH"] = "${ORIGIN}:${ORIGIN}/../lib"
objects = self.compiler.compile(ext.sources,
output_dir = self.build_temp,
include_dirs = ext.include_dirs,
debug = self.debug,
depends = ext.depends)
fileName = os.path.splitext(self.get_ext_filename(ext.name))[0]
fullName = os.path.join(self.build_lib, fileName)
libraryDirs = ext.library_dirs or []
libraries = self.get_libraries(ext)
extraArgs = ext.extra_link_args or []
if sys.platform != "win32":
vars = distutils.sysconfig.get_config_vars()
libraryDirs.append(vars["LIBPL"])
libraries.append("python%s.%s" % sys.version_info[:2])
if vars["LINKFORSHARED"]:
extraArgs.extend(vars["LINKFORSHARED"].split())
if vars["LIBS"]:
extraArgs.extend(vars["LIBS"].split())
if vars["LIBM"]:
extraArgs.append(vars["LIBM"])
if vars["BASEMODLIBS"]:
extraArgs.extend(vars["BASEMODLIBS"].split())
if vars["LOCALMODLIBS"]:
extraArgs.extend(vars["LOCALMODLIBS"].split())
extraArgs.append("-s")
self.compiler.link_executable(objects, fullName,
libraries = libraries,
library_dirs = libraryDirs,
runtime_library_dirs = ext.runtime_library_dirs,
extra_postargs = extraArgs,
debug = self.debug)
def get_ext_filename(self, name):
fileName = distutils.command.build_ext.build_ext.get_ext_filename(self,
name)
if name.find("bases") < 0:
return fileName
ext = self.compiler.exe_extension or ""
return os.path.splitext(fileName)[0] + ext
class build_scripts(distutils.command.build_scripts.build_scripts):
def copy_scripts(self):
distutils.command.build_scripts.build_scripts.copy_scripts(self)
if sys.platform == "win32":
for script in self.scripts:
batFileName = os.path.join(self.build_dir, script + ".bat")
fullScriptName = r"%s\Scripts\%s" % \
(os.path.dirname(sys.executable), script)
command = "%s %s %%1 %%2 %%3 %%4 %%5 %%6 %%7 %%8 %%9" % \
(sys.executable, fullScriptName)
file(batFileName, "w").write("@echo off\n\n%s" % command)
class install(distutils.command.install.install):
def get_sub_commands(self):
subCommands = distutils.command.install.install.get_sub_commands(self)
subCommands.append("install_packagedata")
return subCommands
class install_packagedata(distutils.command.install_data.install_data):
def run(self):
installCommand = self.get_finalized_command("install")
installDir = getattr(installCommand, "install_lib")
sourceDirs = ["samples", "initscripts"]
while sourceDirs:
sourceDir = sourceDirs.pop(0)
targetDir = os.path.join(installDir, "cx_Freeze", sourceDir)
self.mkpath(targetDir)
for name in os.listdir(sourceDir):
if name == "build" or name.startswith("."):
continue
fullSourceName = os.path.join(sourceDir, name)
if os.path.isdir(fullSourceName):
sourceDirs.append(fullSourceName)
else:
fullTargetName = os.path.join(targetDir, name)
self.copy_file(fullSourceName, fullTargetName)
self.outfiles.append(fullTargetName)
commandClasses = dict(
build_ext = build_ext,
build_scripts = build_scripts,
bdist_rpm = bdist_rpm,
install = install,
install_packagedata = install_packagedata)
if sys.platform == "win32":
libraries = ["imagehlp"]
else:
libraries = []
utilModule = Extension("cx_Freeze.util", ["source/util.c"],
libraries = libraries)
depends = ["source/bases/Common.c"]
if sys.platform == "win32":
if sys.version_info[:2] >= (2, 6):
extraSources = ["source/bases/manifest.rc"]
else:
extraSources = ["source/bases/dummy.rc"]
else:
extraSources = []
console = Extension("cx_Freeze.bases.Console",
["source/bases/Console.c"] + extraSources, depends = depends)
consoleKeepPath = Extension("cx_Freeze.bases.ConsoleKeepPath",
["source/bases/ConsoleKeepPath.c"] + extraSources, depends = depends)
extensions = [utilModule, console, consoleKeepPath]
if sys.platform == "win32":
gui = Extension("cx_Freeze.bases.Win32GUI",
["source/bases/Win32GUI.c"] + extraSources,
depends = depends, extra_link_args = ["-mwindows"])
extensions.append(gui)
docFiles = "LICENSE.txt README.txt HISTORY.txt doc/cx_Freeze.html"
classifiers = [
"Development Status :: 5 - Production/Stable",
"Intended Audience :: Developers",
"License :: OSI Approved :: Python Software Foundation License",
"Natural Language :: English",
"Operating System :: OS Independent",
"Programming Language :: C",
"Programming Language :: Python",
"Topic :: Software Development :: Build Tools",
"Topic :: Software Development :: Libraries :: Python Modules",
"Topic :: System :: Software Distribution",
"Topic :: Utilities"
]
setup(name = "cx_Freeze",
description = "create standalone executables from Python scripts",
long_description = "create standalone executables from Python scripts",
version = "4.0.1",
cmdclass = commandClasses,
options = dict(bdist_rpm = dict(doc_files = docFiles),
install = dict(optimize = 1)),
ext_modules = extensions,
packages = ['cx_Freeze'],
maintainer="Anthony Tuininga",
maintainer_email="anthony.tuininga@gmail.com",
url = "http://cx-freeze.sourceforge.net",
scripts = ["cxfreeze"],
classifiers = classifiers,
keywords = "freeze",
license = "Python Software Foundation License")

View File

@ -0,0 +1,262 @@
//-----------------------------------------------------------------------------
// Common.c
// Routines which are common to running frozen executables.
//-----------------------------------------------------------------------------
#include <compile.h>
#include <eval.h>
#include <osdefs.h>
// global variables (used for simplicity)
static PyObject *g_FileName = NULL;
static PyObject *g_DirName = NULL;
static PyObject *g_ExclusiveZipFileName = NULL;
static PyObject *g_SharedZipFileName = NULL;
static PyObject *g_InitScriptZipFileName = NULL;
//-----------------------------------------------------------------------------
// GetDirName()
// Return the directory name of the given path.
//-----------------------------------------------------------------------------
static int GetDirName(
const char *path, // path to calculate dir name for
PyObject **dirName) // directory name (OUT)
{
int i;
for (i = strlen(path); i > 0 && path[i] != SEP; --i);
*dirName = PyString_FromStringAndSize(path, i);
if (!*dirName)
return FatalError("cannot create string for directory name");
return 0;
}
//-----------------------------------------------------------------------------
// SetExecutableName()
// Set the script to execute and calculate the directory in which the
// executable is found as well as the exclusive (only for this executable) and
// shared zip file names.
//-----------------------------------------------------------------------------
static int SetExecutableName(
const char *fileName) // script to execute
{
char temp[MAXPATHLEN + 12], *ptr;
#ifndef WIN32
char linkData[MAXPATHLEN + 1];
struct stat statData;
size_t linkSize, i;
PyObject *dirName;
#endif
// store file name
g_FileName = PyString_FromString(fileName);
if (!g_FileName)
return FatalError("cannot create string for file name");
#ifndef WIN32
for (i = 0; i < 25; i++) {
if (lstat(fileName, &statData) < 0) {
PyErr_SetFromErrnoWithFilename(PyExc_OSError, (char*) fileName);
return FatalError("unable to stat file");
}
if (!S_ISLNK(statData.st_mode))
break;
linkSize = readlink(fileName, linkData, sizeof(linkData));
if (linkSize < 0) {
PyErr_SetFromErrnoWithFilename(PyExc_OSError, (char*) fileName);
return FatalError("unable to stat file");
}
if (linkData[0] == '/') {
Py_DECREF(g_FileName);
g_FileName = PyString_FromStringAndSize(linkData, linkSize);
} else {
if (GetDirName(PyString_AS_STRING(g_FileName), &dirName) < 0)
return -1;
if (PyString_GET_SIZE(dirName) + linkSize + 1 > MAXPATHLEN) {
Py_DECREF(dirName);
return FatalError("cannot dereference link, path too large");
}
strcpy(temp, PyString_AS_STRING(dirName));
strcat(temp, "/");
strcat(temp, linkData);
Py_DECREF(g_FileName);
g_FileName = PyString_FromString(temp);
}
if (!g_FileName)
return FatalError("cannot create string for linked file name");
fileName = PyString_AS_STRING(g_FileName);
}
#endif
// calculate and store directory name
if (GetDirName(fileName, &g_DirName) < 0)
return -1;
// calculate and store exclusive zip file name
strcpy(temp, fileName);
ptr = temp + strlen(temp) - 1;
while (ptr > temp && *ptr != SEP && *ptr != '.')
ptr--;
if (*ptr == '.')
*ptr = '\0';
strcat(temp, ".zip");
g_ExclusiveZipFileName = PyString_FromString(temp);
if (!g_ExclusiveZipFileName)
return FatalError("cannot create string for exclusive zip file name");
// calculate and store shared zip file name
strcpy(temp, PyString_AS_STRING(g_DirName));
ptr = temp + strlen(temp);
*ptr++ = SEP;
strcpy(ptr, "library.zip");
g_SharedZipFileName = PyString_FromString(temp);
if (!g_SharedZipFileName)
return FatalError("cannot create string for shared zip file name");
return 0;
}
//-----------------------------------------------------------------------------
// SetPathToSearch()
// Set the path to search. This includes the file (for those situations where
// a zip file is attached to the executable itself), the directory where the
// executable is found (to search for extensions), the exclusive zip file
// name and the shared zip file name.
//-----------------------------------------------------------------------------
static int SetPathToSearch(void)
{
PyObject *pathList;
pathList = PySys_GetObject("path");
if (!pathList)
return FatalError("cannot acquire sys.path");
if (PyList_Insert(pathList, 0, g_FileName) < 0)
return FatalError("cannot insert file name into sys.path");
if (PyList_Insert(pathList, 1, g_DirName) < 0)
return FatalError("cannot insert directory name into sys.path");
if (PyList_Insert(pathList, 2, g_ExclusiveZipFileName) < 0)
return FatalError("cannot insert exclusive zip name into sys.path");
if (PyList_Insert(pathList, 3, g_SharedZipFileName) < 0)
return FatalError("cannot insert shared zip name into sys.path");
return 0;
}
//-----------------------------------------------------------------------------
// GetImporterHelper()
// Helper which is used to locate the importer for the initscript.
//-----------------------------------------------------------------------------
static PyObject *GetImporterHelper(
PyObject *module, // zipimport module
PyObject *fileName) // name of file to search
{
PyObject *importer;
importer = PyObject_CallMethod(module, "zipimporter", "O", fileName);
if (importer)
g_InitScriptZipFileName = fileName;
else
PyErr_Clear();
return importer;
}
//-----------------------------------------------------------------------------
// GetImporter()
// Return the importer which will be used for importing the initialization
// script. The executable itself is searched first, followed by the exclusive
// zip file and finally by the shared zip file.
//-----------------------------------------------------------------------------
static int GetImporter(
PyObject **importer) // importer (OUT)
{
PyObject *module;
module = PyImport_ImportModule("zipimport");
if (!module)
return FatalError("cannot import zipimport module");
*importer = GetImporterHelper(module, g_FileName);
if (!*importer) {
*importer = GetImporterHelper(module, g_ExclusiveZipFileName);
if (!*importer)
*importer = GetImporterHelper(module, g_SharedZipFileName);
}
Py_DECREF(module);
if (!*importer)
return FatalError("cannot get zipimporter instance");
return 0;
}
//-----------------------------------------------------------------------------
// PopulateInitScriptDict()
// Return the dictionary used by the initialization script.
//-----------------------------------------------------------------------------
static int PopulateInitScriptDict(
PyObject *dict) // dictionary to populate
{
if (!dict)
return FatalError("unable to create temporary dictionary");
if (PyDict_SetItemString(dict, "__builtins__", PyEval_GetBuiltins()) < 0)
return FatalError("unable to set __builtins__");
if (PyDict_SetItemString(dict, "FILE_NAME", g_FileName) < 0)
return FatalError("unable to set FILE_NAME");
if (PyDict_SetItemString(dict, "DIR_NAME", g_DirName) < 0)
return FatalError("unable to set DIR_NAME");
if (PyDict_SetItemString(dict, "EXCLUSIVE_ZIP_FILE_NAME",
g_ExclusiveZipFileName) < 0)
return FatalError("unable to set EXCLUSIVE_ZIP_FILE_NAME");
if (PyDict_SetItemString(dict, "SHARED_ZIP_FILE_NAME",
g_SharedZipFileName) < 0)
return FatalError("unable to set SHARED_ZIP_FILE_NAME");
if (PyDict_SetItemString(dict, "INITSCRIPT_ZIP_FILE_NAME",
g_InitScriptZipFileName) < 0)
return FatalError("unable to set INITSCRIPT_ZIP_FILE_NAME");
return 0;
}
//-----------------------------------------------------------------------------
// ExecuteScript()
// Execute the script found within the file.
//-----------------------------------------------------------------------------
static int ExecuteScript(
const char *fileName) // name of file containing Python code
{
PyObject *importer, *dict, *code, *temp;
if (SetExecutableName(fileName) < 0)
return -1;
if (SetPathToSearch() < 0)
return -1;
importer = NULL;
if (GetImporter(&importer) < 0)
return -1;
// create and populate dictionary for initscript module
dict = PyDict_New();
if (PopulateInitScriptDict(dict) < 0) {
Py_XDECREF(dict);
Py_DECREF(importer);
return -1;
}
// locate and execute script
code = PyObject_CallMethod(importer, "get_code", "s", "cx_Freeze__init__");
Py_DECREF(importer);
if (!code)
return FatalError("unable to locate initialization module");
temp = PyEval_EvalCode( (PyCodeObject*) code, dict, dict);
Py_DECREF(code);
Py_DECREF(dict);
if (!temp)
return FatalScriptError();
Py_DECREF(temp);
return 0;
}

View File

@ -0,0 +1,72 @@
//-----------------------------------------------------------------------------
// Console.c
// Main routine for frozen programs which run in a console.
//-----------------------------------------------------------------------------
#include <Python.h>
#ifdef __WIN32__
#include <windows.h>
#endif
//-----------------------------------------------------------------------------
// FatalError()
// Prints a fatal error.
//-----------------------------------------------------------------------------
static int FatalError(
const char *message) // message to print
{
PyErr_Print();
Py_FatalError(message);
return -1;
}
//-----------------------------------------------------------------------------
// FatalScriptError()
// Prints a fatal error in the initialization script.
//-----------------------------------------------------------------------------
static int FatalScriptError(void)
{
PyErr_Print();
return -1;
}
#include "Common.c"
//-----------------------------------------------------------------------------
// main()
// Main routine for frozen programs.
//-----------------------------------------------------------------------------
int main(int argc, char **argv)
{
const char *fileName;
char *encoding;
// initialize Python
Py_NoSiteFlag = 1;
Py_FrozenFlag = 1;
Py_IgnoreEnvironmentFlag = 1;
encoding = getenv("PYTHONIOENCODING");
if (encoding != NULL) {
Py_FileSystemDefaultEncoding = strndup(encoding, 100);
}
Py_SetPythonHome("");
Py_SetProgramName(argv[0]);
fileName = Py_GetProgramFullPath();
Py_Initialize();
PySys_SetArgv(argc, argv);
// do the work
if (ExecuteScript(fileName) < 0)
return 1;
Py_Finalize();
return 0;
}

View File

@ -0,0 +1,60 @@
//-----------------------------------------------------------------------------
// ConsoleKeepPath.c
// Main routine for frozen programs which need a Python installation to do
// their work.
//-----------------------------------------------------------------------------
#include <Python.h>
#ifdef __WIN32__
#include <windows.h>
#endif
//-----------------------------------------------------------------------------
// FatalError()
// Prints a fatal error.
//-----------------------------------------------------------------------------
static int FatalError(
const char *message) // message to print
{
PyErr_Print();
Py_FatalError(message);
return -1;
}
//-----------------------------------------------------------------------------
// FatalScriptError()
// Prints a fatal error in the initialization script.
//-----------------------------------------------------------------------------
static int FatalScriptError(void)
{
PyErr_Print();
return -1;
}
#include "Common.c"
//-----------------------------------------------------------------------------
// main()
// Main routine for frozen programs.
//-----------------------------------------------------------------------------
int main(int argc, char **argv)
{
const char *fileName;
// initialize Python
Py_SetProgramName(argv[0]);
fileName = Py_GetProgramFullPath();
Py_Initialize();
PySys_SetArgv(argc, argv);
// do the work
if (ExecuteScript(fileName) < 0)
return 1;
Py_Finalize();
return 0;
}

View File

@ -0,0 +1,242 @@
//-----------------------------------------------------------------------------
// Win32GUI.c
// Main routine for frozen programs written for the Win32 GUI subsystem.
//-----------------------------------------------------------------------------
#include <Python.h>
#include <windows.h>
//-----------------------------------------------------------------------------
// FatalError()
// Handle a fatal error.
//-----------------------------------------------------------------------------
static int FatalError(
char *a_Message) // message to display
{
MessageBox(NULL, a_Message, "cx_Freeze Fatal Error", MB_ICONERROR);
Py_Finalize();
return -1;
}
//-----------------------------------------------------------------------------
// StringifyObject()
// Stringify a Python object.
//-----------------------------------------------------------------------------
static char *StringifyObject(
PyObject *object, // object to stringify
PyObject **stringRep) // string representation
{
if (object) {
*stringRep = PyObject_Str(object);
if (*stringRep)
return PyString_AS_STRING(*stringRep);
return "Unable to stringify";
}
// object is NULL
*stringRep = NULL;
return "None";
}
//-----------------------------------------------------------------------------
// FatalPythonErrorNoTraceback()
// Handle a fatal Python error without traceback.
//-----------------------------------------------------------------------------
static int FatalPythonErrorNoTraceback(
PyObject *origType, // exception type
PyObject *origValue, // exception value
char *message) // message to display
{
PyObject *typeStrRep, *valueStrRep, *origTypeStrRep, *origValueStrRep;
char *totalMessage, *typeStr, *valueStr, *origTypeStr, *origValueStr;
PyObject *type, *value, *traceback;
int totalMessageLength;
char *messageFormat;
// fetch error and string representations of the error
PyErr_Fetch(&type, &value, &traceback);
origTypeStr = StringifyObject(origType, &origTypeStrRep);
origValueStr = StringifyObject(origValue, &origValueStrRep);
typeStr = StringifyObject(type, &typeStrRep);
valueStr = StringifyObject(value, &valueStrRep);
// fill out the message to be displayed
messageFormat = "Type: %s\nValue: %s\nOther Type: %s\nOtherValue: %s\n%s";
totalMessageLength = strlen(origTypeStr) + strlen(origValueStr) +
strlen(typeStr) + strlen(valueStr) + strlen(message) +
strlen(messageFormat) + 1;
totalMessage = malloc(totalMessageLength);
if (!totalMessage)
return FatalError("Out of memory!");
sprintf(totalMessage, messageFormat, typeStr, valueStr, origTypeStr,
origValueStr, message);
// display the message
MessageBox(NULL, totalMessage,
"cx_Freeze: Python error in main script (traceback unavailable)",
MB_ICONERROR);
free(totalMessage);
return -1;
}
//-----------------------------------------------------------------------------
// ArgumentValue()
// Return a suitable argument value by replacing NULL with Py_None.
//-----------------------------------------------------------------------------
static PyObject *ArgumentValue(
PyObject *object) // argument to massage
{
if (object) {
Py_INCREF(object);
return object;
}
Py_INCREF(Py_None);
return Py_None;
}
//-----------------------------------------------------------------------------
// HandleSystemExitException()
// Handles a system exit exception differently. If an integer value is passed
// through then that becomes the exit value; otherwise the string value of the
// value passed through is displayed in a message box.
//-----------------------------------------------------------------------------
static void HandleSystemExitException()
{
PyObject *type, *value, *traceback, *valueStr;
int exitCode = 0;
char *message;
PyErr_Fetch(&type, &value, &traceback);
if (PyInstance_Check(value)) {
PyObject *code = PyObject_GetAttrString(value, "code");
if (code) {
Py_DECREF(value);
value = code;
if (value == Py_None)
Py_Exit(0);
}
}
if (PyInt_Check(value))
exitCode = PyInt_AsLong(value);
else {
message = StringifyObject(value, &valueStr);
MessageBox(NULL, message, "cx_Freeze: Application Terminated",
MB_ICONERROR);
Py_XDECREF(valueStr);
exitCode = 1;
}
Py_Exit(exitCode);
}
//-----------------------------------------------------------------------------
// FatalScriptError()
// Handle a fatal Python error with traceback.
//-----------------------------------------------------------------------------
static int FatalScriptError()
{
PyObject *type, *value, *traceback, *argsTuple, *module, *method, *result;
int tracebackLength, i;
char *tracebackStr;
// if a system exception, handle it specially
if (PyErr_ExceptionMatches(PyExc_SystemExit))
HandleSystemExitException();
// get the exception details
PyErr_Fetch(&type, &value, &traceback);
// import the traceback module
module = PyImport_ImportModule("traceback");
if (!module)
return FatalPythonErrorNoTraceback(type, value,
"Cannot import traceback module.");
// get the format_exception method
method = PyObject_GetAttrString(module, "format_exception");
Py_DECREF(module);
if (!method)
return FatalPythonErrorNoTraceback(type, value,
"Cannot get format_exception method.");
// create a tuple for the arguments
argsTuple = PyTuple_New(3);
if (!argsTuple) {
Py_DECREF(method);
return FatalPythonErrorNoTraceback(type, value,
"Cannot create arguments tuple for traceback.");
}
PyTuple_SET_ITEM(argsTuple, 0, ArgumentValue(type));
PyTuple_SET_ITEM(argsTuple, 1, ArgumentValue(value));
PyTuple_SET_ITEM(argsTuple, 2, ArgumentValue(traceback));
// call the format_exception method
result = PyObject_CallObject(method, argsTuple);
Py_DECREF(method);
Py_DECREF(argsTuple);
if (!result)
return FatalPythonErrorNoTraceback(type, value,
"Failed calling format_exception method.");
// determine length of string representation of formatted traceback
tracebackLength = 1;
for (i = 0; i < PyList_GET_SIZE(result); i++)
tracebackLength += PyString_GET_SIZE(PyList_GET_ITEM(result, i));
// create a string representation of the formatted traceback
tracebackStr = malloc(tracebackLength);
if (!tracebackStr) {
Py_DECREF(result);
return FatalError("Out of memory!");
}
tracebackStr[0] = '\0';
for (i = 0; i < PyList_GET_SIZE(result); i++)
strcat(tracebackStr, PyString_AS_STRING(PyList_GET_ITEM(result, i)));
Py_DECREF(result);
// bring up the error
MessageBox(NULL, tracebackStr, "cx_Freeze: Python error in main script",
MB_ICONERROR);
Py_Finalize();
return 1;
}
#include "Common.c"
//-----------------------------------------------------------------------------
// WinMain()
// Main routine for the executable in Windows.
//-----------------------------------------------------------------------------
int WINAPI WinMain(
HINSTANCE instance, // handle to application
HINSTANCE prevInstance, // previous handle to application
LPSTR commandLine, // command line
int showFlag) // show flag
{
const char *fileName;
// initialize Python
Py_NoSiteFlag = 1;
Py_FrozenFlag = 1;
Py_IgnoreEnvironmentFlag = 1;
Py_SetPythonHome("");
Py_SetProgramName(__argv[0]);
fileName = Py_GetProgramFullPath();
Py_Initialize();
PySys_SetArgv(__argc, __argv);
// do the work
if (ExecuteScript(fileName) < 0)
return 1;
// terminate Python
Py_Finalize();
return 0;
}

View File

@ -0,0 +1,5 @@
STRINGTABLE
{
1, "Just to ensure that buggy EndUpdateResource doesn't fall over."
}

View File

@ -0,0 +1,3 @@
#include "dummy.rc"
1 24 source/bases/manifest.txt

View File

@ -0,0 +1,418 @@
//-----------------------------------------------------------------------------
// util.c
// Shared library for use by cx_Freeze.
//-----------------------------------------------------------------------------
#include <Python.h>
#ifdef WIN32
#include <windows.h>
#include <imagehlp.h>
#pragma pack(2)
typedef struct {
BYTE bWidth; // Width, in pixels, of the image
BYTE bHeight; // Height, in pixels, of the image
BYTE bColorCount; // Number of colors in image
BYTE bReserved; // Reserved ( must be 0)
WORD wPlanes; // Color Planes
WORD wBitCount; // Bits per pixel
DWORD dwBytesInRes; // How many bytes in this resource?
DWORD dwImageOffset; // Where in the file is this image?
} ICONDIRENTRY;
typedef struct {
WORD idReserved; // Reserved (must be 0)
WORD idType; // Resource Type (1 for icons)
WORD idCount; // How many images?
ICONDIRENTRY idEntries[0]; // An entry for each image
} ICONDIR;
typedef struct {
BYTE bWidth; // Width, in pixels, of the image
BYTE bHeight; // Height, in pixels, of the image
BYTE bColorCount; // Number of colors in image
BYTE bReserved; // Reserved ( must be 0)
WORD wPlanes; // Color Planes
WORD wBitCount; // Bits per pixel
DWORD dwBytesInRes; // How many bytes in this resource?
WORD nID; // resource ID
} GRPICONDIRENTRY;
typedef struct {
WORD idReserved; // Reserved (must be 0)
WORD idType; // Resource Type (1 for icons)
WORD idCount; // How many images?
GRPICONDIRENTRY idEntries[0]; // An entry for each image
} GRPICONDIR;
#endif
//-----------------------------------------------------------------------------
// Globals
//-----------------------------------------------------------------------------
#ifdef WIN32
static PyObject *g_BindErrorException = NULL;
static PyObject *g_ImageNames = NULL;
#endif
#ifdef WIN32
//-----------------------------------------------------------------------------
// BindStatusRoutine()
// Called by BindImageEx() at various points. This is used to determine the
// dependency tree which is later examined by cx_Freeze.
//-----------------------------------------------------------------------------
static BOOL __stdcall BindStatusRoutine(
IMAGEHLP_STATUS_REASON reason, // reason called
PSTR imageName, // name of image being examined
PSTR dllName, // name of DLL
ULONG virtualAddress, // computed virtual address
ULONG parameter) // parameter (value depends on reason)
{
char fileName[MAX_PATH + 1];
switch (reason) {
case BindImportModule:
if (!SearchPath(NULL, dllName, NULL, sizeof(fileName), fileName,
NULL))
return FALSE;
Py_INCREF(Py_None);
if (PyDict_SetItemString(g_ImageNames, fileName, Py_None) < 0)
return FALSE;
break;
default:
break;
}
return TRUE;
}
//-----------------------------------------------------------------------------
// GetFileData()
// Return the data for the given file.
//-----------------------------------------------------------------------------
static int GetFileData(
const char *fileName, // name of file to read
char **data) // pointer to data (OUT)
{
DWORD numberOfBytesRead, dataSize;
HANDLE file;
file = CreateFile(fileName, GENERIC_READ, FILE_SHARE_READ, NULL,
OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
if (file == INVALID_HANDLE_VALUE)
return -1;
dataSize = GetFileSize(file, NULL);
if (dataSize == INVALID_FILE_SIZE) {
CloseHandle(file);
return -1;
}
*data = PyMem_Malloc(dataSize);
if (!*data) {
CloseHandle(file);
return -1;
}
if (!ReadFile(file, *data, dataSize, &numberOfBytesRead, NULL)) {
CloseHandle(file);
return -1;
}
CloseHandle(file);
return 0;
}
//-----------------------------------------------------------------------------
// CreateGroupIconResource()
// Return the group icon resource given the icon file data.
//-----------------------------------------------------------------------------
static GRPICONDIR *CreateGroupIconResource(
ICONDIR *iconDir, // icon information
DWORD *resourceSize) // size of resource (OUT)
{
GRPICONDIR *groupIconDir;
int i;
*resourceSize = sizeof(GRPICONDIR) +
sizeof(GRPICONDIRENTRY) * iconDir->idCount;
groupIconDir = PyMem_Malloc(*resourceSize);
if (!groupIconDir)
return NULL;
groupIconDir->idReserved = iconDir->idReserved;
groupIconDir->idType = iconDir->idType;
groupIconDir->idCount = iconDir->idCount;
for (i = 0; i < iconDir->idCount; i++) {
groupIconDir->idEntries[i].bWidth = iconDir->idEntries[i].bWidth;
groupIconDir->idEntries[i].bHeight = iconDir->idEntries[i].bHeight;
groupIconDir->idEntries[i].bColorCount =
iconDir->idEntries[i].bColorCount;
groupIconDir->idEntries[i].bReserved = iconDir->idEntries[i].bReserved;
groupIconDir->idEntries[i].wPlanes = iconDir->idEntries[i].wPlanes;
groupIconDir->idEntries[i].wBitCount = iconDir->idEntries[i].wBitCount;
groupIconDir->idEntries[i].dwBytesInRes =
iconDir->idEntries[i].dwBytesInRes;
groupIconDir->idEntries[i].nID = i + 1;
}
return groupIconDir;
}
//-----------------------------------------------------------------------------
// ExtAddIcon()
// Add the icon as a resource to the specified file.
//-----------------------------------------------------------------------------
static PyObject *ExtAddIcon(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
char *executableName, *iconName, *data, *iconData;
GRPICONDIR *groupIconDir;
DWORD resourceSize;
ICONDIR *iconDir;
BOOL succeeded;
HANDLE handle;
int i;
if (!PyArg_ParseTuple(args, "ss", &executableName, &iconName))
return NULL;
// begin updating the executable
handle = BeginUpdateResource(executableName, FALSE);
if (!handle) {
PyErr_SetExcFromWindowsErrWithFilename(PyExc_WindowsError,
GetLastError(), executableName);
return NULL;
}
// first attempt to get the data from the icon file
data = NULL;
succeeded = TRUE;
groupIconDir = NULL;
if (GetFileData(iconName, &data) < 0)
succeeded = FALSE;
iconDir = (ICONDIR*) data;
// next, attempt to add a group icon resource
if (succeeded) {
groupIconDir = CreateGroupIconResource(iconDir, &resourceSize);
if (groupIconDir)
succeeded = UpdateResource(handle, RT_GROUP_ICON,
MAKEINTRESOURCE(1),
MAKELANGID(LANG_NEUTRAL, SUBLANG_NEUTRAL),
groupIconDir, resourceSize);
else succeeded = FALSE;
}
// next, add each icon as a resource
if (succeeded) {
for (i = 0; i < iconDir->idCount; i++) {
iconData = &data[iconDir->idEntries[i].dwImageOffset];
resourceSize = iconDir->idEntries[i].dwBytesInRes;
succeeded = UpdateResource(handle, RT_ICON, MAKEINTRESOURCE(i + 1),
MAKELANGID(LANG_NEUTRAL, SUBLANG_NEUTRAL), iconData,
resourceSize);
if (!succeeded)
break;
}
}
// finish writing the resource (or discarding the changes upon an error)
if (!EndUpdateResource(handle, !succeeded)) {
if (succeeded) {
succeeded = FALSE;
PyErr_SetExcFromWindowsErrWithFilename(PyExc_WindowsError,
GetLastError(), executableName);
}
}
// clean up
if (groupIconDir)
PyMem_Free(groupIconDir);
if (data)
PyMem_Free(data);
if (!succeeded)
return NULL;
Py_INCREF(Py_None);
return Py_None;
}
//-----------------------------------------------------------------------------
// ExtBeginUpdateResource()
// Wrapper for BeginUpdateResource().
//-----------------------------------------------------------------------------
static PyObject *ExtBeginUpdateResource(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
BOOL deleteExistingResources;
char *fileName;
HANDLE handle;
deleteExistingResources = TRUE;
if (!PyArg_ParseTuple(args, "s|i", &fileName, &deleteExistingResources))
return NULL;
handle = BeginUpdateResource(fileName, deleteExistingResources);
if (!handle) {
PyErr_SetExcFromWindowsErrWithFilename(PyExc_WindowsError,
GetLastError(), fileName);
return NULL;
}
return PyInt_FromLong((long) handle);
}
//-----------------------------------------------------------------------------
// ExtUpdateResource()
// Wrapper for UpdateResource().
//-----------------------------------------------------------------------------
static PyObject *ExtUpdateResource(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
int resourceType, resourceId, resourceDataSize;
char *resourceData;
HANDLE handle;
if (!PyArg_ParseTuple(args, "iiis#", &handle, &resourceType, &resourceId,
&resourceData, &resourceDataSize))
return NULL;
if (!UpdateResource(handle, MAKEINTRESOURCE(resourceType),
MAKEINTRESOURCE(resourceId),
MAKELANGID(LANG_NEUTRAL, SUBLANG_NEUTRAL), resourceData,
resourceDataSize)) {
PyErr_SetExcFromWindowsErr(PyExc_WindowsError, GetLastError());
return NULL;
}
Py_INCREF(Py_None);
return Py_None;
}
//-----------------------------------------------------------------------------
// ExtEndUpdateResource()
// Wrapper for EndUpdateResource().
//-----------------------------------------------------------------------------
static PyObject *ExtEndUpdateResource(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
BOOL discardChanges;
HANDLE handle;
discardChanges = FALSE;
if (!PyArg_ParseTuple(args, "i|i", &handle, &discardChanges))
return NULL;
if (!EndUpdateResource(handle, discardChanges)) {
PyErr_SetExcFromWindowsErr(PyExc_WindowsError, GetLastError());
return NULL;
}
Py_INCREF(Py_None);
return Py_None;
}
//-----------------------------------------------------------------------------
// ExtGetDependentFiles()
// Return a list of files that this file depends on.
//-----------------------------------------------------------------------------
static PyObject *ExtGetDependentFiles(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
PyObject *results;
char *imageName;
if (!PyArg_ParseTuple(args, "s", &imageName))
return NULL;
g_ImageNames = PyDict_New();
if (!g_ImageNames)
return NULL;
if (!BindImageEx(BIND_NO_BOUND_IMPORTS | BIND_NO_UPDATE | BIND_ALL_IMAGES,
imageName, NULL, NULL, BindStatusRoutine)) {
Py_DECREF(g_ImageNames);
PyErr_SetExcFromWindowsErrWithFilename(g_BindErrorException,
GetLastError(), imageName);
return NULL;
}
results = PyDict_Keys(g_ImageNames);
Py_DECREF(g_ImageNames);
return results;
}
//-----------------------------------------------------------------------------
// ExtGetSystemDir()
// Return the Windows directory (C:\Windows for example).
//-----------------------------------------------------------------------------
static PyObject *ExtGetSystemDir(
PyObject *self, // passthrough argument
PyObject *args) // arguments (ignored)
{
char dir[MAX_PATH + 1];
if (GetSystemDirectory(dir, sizeof(dir)))
return PyString_FromString(dir);
PyErr_SetExcFromWindowsErr(PyExc_RuntimeError, GetLastError());
return NULL;
}
#endif
//-----------------------------------------------------------------------------
// ExtSetOptimizeFlag()
// Set the optimize flag as needed.
//-----------------------------------------------------------------------------
static PyObject *ExtSetOptimizeFlag(
PyObject *self, // passthrough argument
PyObject *args) // arguments
{
if (!PyArg_ParseTuple(args, "i", &Py_OptimizeFlag))
return NULL;
Py_INCREF(Py_None);
return Py_None;
}
//-----------------------------------------------------------------------------
// Methods
//-----------------------------------------------------------------------------
static PyMethodDef g_ModuleMethods[] = {
{ "SetOptimizeFlag", ExtSetOptimizeFlag, METH_VARARGS },
#ifdef WIN32
{ "BeginUpdateResource", ExtBeginUpdateResource, METH_VARARGS },
{ "UpdateResource", ExtUpdateResource, METH_VARARGS },
{ "EndUpdateResource", ExtEndUpdateResource, METH_VARARGS },
{ "AddIcon", ExtAddIcon, METH_VARARGS },
{ "GetDependentFiles", ExtGetDependentFiles, METH_VARARGS },
{ "GetSystemDir", ExtGetSystemDir, METH_NOARGS },
#endif
{ NULL }
};
//-----------------------------------------------------------------------------
// initutil()
// Initialization routine for the shared libary.
//-----------------------------------------------------------------------------
void initutil(void)
{
PyObject *module;
module = Py_InitModule("cx_Freeze.util", g_ModuleMethods);
if (!module)
return;
#ifdef WIN32
g_BindErrorException = PyErr_NewException("cx_Freeze.util.BindError",
NULL, NULL);
if (!g_BindErrorException)
return;
if (PyModule_AddObject(module, "BindError", g_BindErrorException) < 0)
return;
#endif
}

View File

@ -1,4 +1,5 @@
#!/usr/bin/env python
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
__docformat__ = 'restructuredtext en'
@ -6,162 +7,216 @@ __docformat__ = 'restructuredtext en'
'''
Create linux binary.
'''
import glob, sys, subprocess, tarfile, os, re, py_compile, shutil
HOME = '/home/kovid'
PYINSTALLER = os.path.expanduser('~/build/pyinstaller')
CALIBREPREFIX = '___'
PDFTOHTML = '/usr/bin/pdftohtml'
LIBUNRAR = '/usr/lib/libunrar.so'
QTDIR = '/usr/lib/qt4'
QTDLLS = ('QtCore', 'QtGui', 'QtNetwork', 'QtSvg', 'QtXml', 'QtWebKit')
EXTRAS = ('/usr/lib/python2.5/site-packages/PIL', os.path.expanduser('~/ipython/IPython'))
SQLITE = '/usr/lib/libsqlite3.so.0'
DBUS = '/usr/lib/libdbus-1.so.3'
LIBMNG = '/usr/lib/libmng.so.1'
LIBZ = '/lib/libz.so.1'
LIBBZ2 = '/lib/libbz2.so.1'
LIBUSB = '/usr/lib/libusb.so'
LIBPOPPLER = '/usr/lib/libpoppler.so.3'
LIBXML2 = '/usr/lib/libxml2.so.2'
LIBXSLT = '/usr/lib/libxslt.so.1'
LIBEXSLT = '/usr/lib/libexslt.so.0'
def freeze():
import glob, sys, subprocess, tarfile, os, re, textwrap, shutil, cStringIO, bz2, codecs
from contextlib import closing
from cx_Freeze import Executable, setup
from calibre.constants import __version__, __appname__
from calibre.linux import entry_points
from calibre import walk
from calibre.web.feeds.recipes import recipe_modules
import calibre
CALIBRESRC = os.path.join(CALIBREPREFIX, 'src')
CALIBREPLUGINS = os.path.join(CALIBRESRC, 'calibre', 'plugins')
QTDIR = '/usr/lib/qt4'
QTDLLS = ('QtCore', 'QtGui', 'QtNetwork', 'QtSvg', 'QtXml', 'QtWebKit')
binary_excludes = ['libGLcore*', 'libGL*', 'libnvidia*']
binary_includes = [
'/usr/bin/pdftohtml',
'/usr/lib/libunrar.so',
'/usr/lib/libsqlite3.so.0',
'/usr/lib/libsqlite3.so.0',
'/usr/lib/libmng.so.1',
'/lib/libz.so.1',
'/lib/libbz2.so.1',
'/lib/libbz2.so.1',
'/usr/lib/libpoppler.so.4',
'/usr/lib/libxml2.so.2',
'/usr/lib/libxslt.so.1',
'/usr/lib/libxslt.so.1',
'/usr/lib/libMagickWand.so',
'/usr/lib/libMagickCore.so',
]
binary_includes += [os.path.join(QTDIR, 'lib%s.so.4'%x) for x in QTDLLS]
d = os.path.dirname
CALIBRESRC = d(d(d(os.path.abspath(calibre.__file__))))
CALIBREPLUGINS = os.path.join(CALIBRESRC, 'src', 'calibre', 'plugins')
FREEZE_DIR = os.path.join(CALIBRESRC, 'build', 'cx_freeze')
DIST_DIR = os.path.join(CALIBRESRC, 'dist')
os.chdir(CALIBRESRC)
print 'Freezing calibre located at', CALIBRESRC
sys.path.insert(0, os.path.join(CALIBRESRC, 'src'))
entry_points = entry_points['console_scripts'] + entry_points['gui_scripts']
entry_points = ['calibre_postinstall=calibre.linux:binary_install',
'calibre-parallel=calibre.parallel:main'] + entry_points
executables = {}
for ep in entry_points:
executables[ep.split('=')[0].strip()] = (ep.split('=')[1].split(':')[0].strip(),
ep.split(':')[-1].strip())
if os.path.exists(FREEZE_DIR):
shutil.rmtree(FREEZE_DIR)
os.makedirs(FREEZE_DIR)
if not os.path.exists(DIST_DIR):
os.makedirs(DIST_DIR)
includes = [x[0] for x in executables.values()]
excludes = ['matplotlib', "Tkconstants", "Tkinter", "tcl", "_imagingtk",
"ImageTk", "FixTk", 'wx', 'PyQt4.QtAssistant', 'PyQt4.QtOpenGL.so',
'PyQt4.QtScript.so', 'PyQt4.QtSql.so', 'PyQt4.QtTest.so', 'qt',
'glib', 'gobject']
packages = ['calibre', 'encodings', 'cherrypy', 'cssutils', 'xdg']
includes += ['calibre.web.feeds.recipes.'+r for r in recipe_modules]
LOADER = '/tmp/loader.py'
open(LOADER, 'wb').write('# This script is never actually used.\nimport sys')
INIT_SCRIPT = '/tmp/init.py'
open(INIT_SCRIPT, 'wb').write(textwrap.dedent('''
## Load calibre module specified in the environment variable CALIBRE_CX_EXE
## Also restrict sys.path to the executables' directory and add the
## executables directory to LD_LIBRARY_PATH
import encodings
import os
import sys
import warnings
import zipimport
import locale
import codecs
enc = locale.getdefaultlocale()[1]
if not enc:
enc = locale.nl_langinfo(locale.CODESET)
enc = codecs.lookup(enc if enc else 'UTF-8').name
sys.setdefaultencoding(enc)
paths = os.environ.get('LD_LIBRARY_PATH', '').split(os.pathsep)
if DIR_NAME not in paths or not sys.getfilesystemencoding():
paths.insert(0, DIR_NAME)
os.environ['LD_LIBRARY_PATH'] = os.pathsep.join(paths)
os.environ['PYTHONIOENCODING'] = enc
os.execv(sys.executable, sys.argv)
sys.path = sys.path[:3]
sys.frozen = True
sys.frozen_path = DIR_NAME
executables = %(executables)s
exe = os.environ.get('CALIBRE_CX_EXE', False)
ret = 1
if not exe:
print >>sys.stderr, 'Invalid invocation of calibre loader. CALIBRE_CX_EXE not set'
elif exe not in executables:
print >>sys.stderr, 'Invalid invocation of calibre loader. CALIBRE_CX_EXE=%%s is unknown'%%exe
else:
from PyQt4.QtCore import QCoreApplication
QCoreApplication.setLibraryPaths([sys.frozen_path, os.path.join(sys.frozen_path, "qtplugins")])
sys.argv[0] = exe
module, func = executables[exe]
module = __import__(module, fromlist=[1])
func = getattr(module, func)
ret = func()
module = sys.modules.get("threading")
if module is not None:
module._shutdown()
sys.exit(ret)
''')%dict(executables=repr(executables)))
sys.argv = ['freeze', 'build_exe']
setup(
name = __appname__,
version = __version__,
executables = [Executable(script=LOADER, targetName='loader', compress=False)],
options = { 'build_exe' :
{
'build_exe' : os.path.join(CALIBRESRC, 'build/cx_freeze'),
'optimize' : 2,
'excludes' : excludes,
'includes' : includes,
'packages' : packages,
'init_script' : INIT_SCRIPT,
'copy_dependent_files' : True,
'create_shared_zip' : False,
}
}
)
def copy_binary(src, dest_dir):
dest = os.path.join(dest_dir, os.path.basename(src))
if not os.path.exists(dest_dir):
os.makedirs(dest_dir)
shutil.copyfile(os.path.realpath(src), dest)
shutil.copymode(os.path.realpath(src), dest)
for f in binary_includes:
copy_binary(f, FREEZE_DIR)
for pat in binary_excludes:
matches = glob.glob(os.path.join(FREEZE_DIR, pat))
for f in matches:
os.remove(f)
print 'Adding calibre plugins...'
os.makedirs(os.path.join(FREEZE_DIR, 'plugins'))
for f in glob.glob(os.path.join(CALIBREPLUGINS, '*.so')):
copy_binary(f, os.path.join(FREEZE_DIR, 'plugins'))
print 'Adding Qt plugins...'
plugdir = os.path.join(QTDIR, 'plugins')
for dirpath, dirnames, filenames in os.walk(plugdir):
for f in filenames:
if not f.endswith('.so') or 'designer' in dirpath or 'codecs' in dirpath or 'sqldrivers' in dirpath:
continue
f = os.path.join(dirpath, f)
dest_dir = dirpath.replace(plugdir, os.path.join(FREEZE_DIR, 'qtplugins'))
copy_binary(f, dest_dir)
sys.path.insert(0, CALIBRESRC)
from calibre import __version__
from calibre.parallel import PARALLEL_FUNCS
from calibre.web.feeds.recipes import recipes
hiddenimports = list(map(lambda x: x[0], PARALLEL_FUNCS.values()))
hiddenimports += ['PyQt4.QtWebKit']
hiddenimports += ['lxml._elementpath', 'keyword', 'codeop', 'commands', 'shlex', 'pydoc']
hiddenimports += map(lambda x: x.__module__, recipes)
open(os.path.join(PYINSTALLER, 'hooks', 'hook-calibre.parallel.py'), 'wb').write('hiddenimports = %s'%repr(hiddenimports))
def run_pyinstaller(args=sys.argv):
subprocess.check_call(('/usr/bin/sudo', 'chown', '-R', 'kovid:users', glob.glob('/usr/lib/python*/site-packages/')[-1]))
subprocess.check_call('rm -rf %(py)s/dist/* %(py)s/build/*'%dict(py=PYINSTALLER), shell=True)
cp = HOME+'/build/'+os.path.basename(os.getcwd())
spec = open(os.path.join(PYINSTALLER, 'calibre', 'calibre.spec'), 'wb')
raw = re.sub(r'CALIBREPREFIX\s+=\s+\'___\'', 'CALIBREPREFIX = '+repr(cp),
open(__file__).read())
spec.write(raw)
spec.close()
os.chdir(PYINSTALLER)
shutil.rmtree('calibre/dist')
os.mkdir('calibre/dist')
subprocess.check_call('python -OO Build.py calibre/calibre.spec', shell=True)
print 'Creating launchers'
for exe in executables:
path = os.path.join(FREEZE_DIR, exe)
open(path, 'wb').write(textwrap.dedent('''\
#!/bin/sh
export CALIBRE_CX_EXE=%s
path=`readlink -e $0`
base=`dirname $path`
loader=$base/loader
export LD_LIBRARY_PATH=$base:$LD_LIBRARY_PATH
$loader "$@"
''')%exe)
os.chmod(path, 0755)
exes = list(executables.keys())
exes.remove('calibre_postinstall')
exes.remove('calibre-parallel')
open(os.path.join(FREEZE_DIR, 'manifest'), 'wb').write('\n'.join(exes))
print 'Creating archive...'
dist = open(os.path.join(DIST_DIR, 'calibre-%s-i686.tar.bz2'%__version__), 'wb')
with closing(tarfile.open(fileobj=dist, mode='w:bz2',
format=tarfile.PAX_FORMAT)) as tf:
for f in walk(FREEZE_DIR):
name = f.replace(FREEZE_DIR, '')[1:]
if name:
tf.add(f, name)
dist.flush()
dist.seek(0, 2)
print 'Archive %s created: %.2f MB'%(dist.name, dist.tell()/(1024.**2))
return 0
if __name__ == '__main__' and 'freeze.py' in __file__:
sys.exit(run_pyinstaller())
loader = os.path.join(os.path.expanduser('~/temp'), 'calibre_installer_loader.py')
if not os.path.exists(loader):
open(loader, 'wb').write('''
import sys, os
sys.frozen_path = os.getcwd()
os.chdir(os.environ.get("ORIGWD", "."))
sys.path.insert(0, os.path.join(sys.frozen_path, "library.pyz"))
sys.path.insert(0, sys.frozen_path)
from PyQt4.QtCore import QCoreApplication
QCoreApplication.setLibraryPaths([sys.frozen_path, os.path.join(sys.frozen_path, "qtplugins")])
''')
excludes = ['gtk._gtk', 'gtk.glade', 'qt', 'matplotlib.nxutils', 'matplotlib._cntr',
'matplotlib.ttconv', 'matplotlib._image', 'matplotlib.ft2font',
'matplotlib._transforms', 'matplotlib._agg', 'matplotlib.backends._backend_agg',
'matplotlib.axes', 'matplotlib', 'matplotlib.pyparsing',
'TKinter', 'atk', 'gobject._gobject', 'pango', 'PIL', 'Image', 'IPython']
sys.path.insert(0, CALIBRESRC)
from calibre.linux import entry_points
executables, scripts = ['calibre_postinstall', 'calibre-parallel'], \
[os.path.join(CALIBRESRC, 'calibre', 'linux.py'), os.path.join(CALIBRESRC, 'calibre', 'parallel.py')]
for entry in entry_points['console_scripts'] + entry_points['gui_scripts']:
fields = entry.split('=')
executables.append(fields[0].strip())
scripts.append(os.path.join(CALIBRESRC, *map(lambda x: x.strip(), fields[1].split(':')[0].split('.')))+'.py')
analyses = [Analysis([os.path.join(HOMEPATH,'support/_mountzlib.py'), os.path.join(HOMEPATH,'support/useUnicode.py'), loader, script],
pathex=[PYINSTALLER, CALIBRESRC], excludes=excludes) for script in scripts]
pyz = TOC()
binaries = TOC()
for a in analyses:
pyz = a.pure + pyz
binaries = a.binaries + binaries
pyz = PYZ(pyz, name='library.pyz')
built_executables = []
for script, exe, a in zip(scripts, executables, analyses):
built_executables.append(EXE(PYZ(TOC()),
a.scripts+[('O','','OPTION'),],
exclude_binaries=1,
name=os.path.join('buildcalibre', exe),
debug=False,
strip=True,
upx=False,
excludes=excludes,
console=1))
print 'Adding plugins...'
for f in glob.glob(os.path.join(CALIBREPLUGINS, '*.so')):
binaries += [('plugins/'+os.path.basename(f), f, 'BINARY')]
for f in glob.glob(os.path.join(CALIBREPLUGINS, '*.so.*')):
binaries += [(os.path.basename(f), f, 'BINARY')]
print 'Adding external programs...'
binaries += [('pdftohtml', PDFTOHTML, 'BINARY'),
('libunrar.so', LIBUNRAR, 'BINARY')]
print 'Adding external libraries...'
binaries += [ (os.path.basename(x), x, 'BINARY') for x in (SQLITE, DBUS,
LIBMNG, LIBZ, LIBBZ2, LIBUSB, LIBPOPPLER, LIBXML2, LIBXSLT, LIBEXSLT)]
qt = []
for dll in QTDLLS:
path = os.path.join(QTDIR, 'lib'+dll+'.so.4')
qt.append((os.path.basename(path), path, 'BINARY'))
binaries += qt
plugins = []
plugdir = os.path.join(QTDIR, 'plugins')
for dirpath, dirnames, filenames in os.walk(plugdir):
for f in filenames:
if not f.endswith('.so') or 'designer' in dirpath or 'codcs' in dirpath or 'sqldrivers' in dirpath : continue
f = os.path.join(dirpath, f)
plugins.append(('qtplugins/'+f.replace(plugdir, ''), f, 'BINARY'))
binaries += plugins
manifest = '/tmp/manifest'
open(manifest, 'wb').write('\n'.join(executables))
version = '/tmp/version'
open(version, 'wb').write(__version__)
coll = COLLECT(binaries, pyz,
[('manifest', manifest, 'DATA'), ('version', version, 'DATA')],
*built_executables,
**dict(strip=True,
upx=False,
excludes=excludes,
name='dist'))
os.chdir(os.path.join(HOMEPATH, 'calibre', 'dist'))
for folder in EXTRAS:
subprocess.check_call('cp -rf %s .'%folder, shell=True)
print 'Building tarball...'
tbz2 = 'calibre-%s-i686.tar.bz2'%__version__
tf = tarfile.open(os.path.join('/tmp', tbz2), 'w:bz2')
for f in os.listdir('.'):
tf.add(f)
if __name__ == '__main__':
freeze()

View File

@ -177,7 +177,7 @@ _check_symlinks_prescript()
def fix_python_dependencies(self, files):
for f in files:
subprocess.check_call(['/usr/bin/install_name_tool', '-change', '/Library/Frameworks/Python.framework/Versions/2.5/Python', '@executable_path/../Frameworks/Python.framework/Versions/2.5/Python', f])
subprocess.check_call(['/usr/bin/install_name_tool', '-change', '/Library/Frameworks/Python.framework/Versions/2.6/Python', '@executable_path/../Frameworks/Python.framework/Versions/2.6/Python', f])
def fix_misc_dependencies(self, files):
for path in files:
@ -247,10 +247,13 @@ _check_symlinks_prescript()
print 'Adding pdftohtml'
os.link(os.path.expanduser('~/pdftohtml'), os.path.join(frameworks_dir, 'pdftohtml'))
print 'Adding plugins'
module_dir = os.path.join(resource_dir, 'lib', 'python2.5', 'lib-dynload')
module_dir = os.path.join(resource_dir, 'lib', 'python2.6', 'lib-dynload')
print 'Adding fontconfig'
for f in glob.glob(os.path.expanduser('~/fontconfig-bundled/*')):
os.link(f, os.path.join(frameworks_dir, os.path.basename(f)))
dest = os.path.join(frameworks_dir, os.path.basename(f))
if os.path.exists(dest):
os.remove(dest)
os.link(f, dest)
dst = os.path.join(resource_dir, 'fonts')
if os.path.exists(dst):
shutil.rmtree(dst)
@ -258,7 +261,7 @@ _check_symlinks_prescript()
print
print 'Adding IPython'
dst = os.path.join(resource_dir, 'lib', 'python2.5', 'IPython')
dst = os.path.join(resource_dir, 'lib', 'python2.6', 'IPython')
if os.path.exists(dst): shutil.rmtree(dst)
shutil.copytree(os.path.expanduser('~/build/ipython/IPython'), dst)
@ -280,6 +283,7 @@ _check_symlinks_prescript()
f = open(launcher_path, 'r')
src = f.read()
f.close()
src = src.replace('import Image', 'from PIL import Image')
src = re.sub('(_run\s*\(.*?.py.*?\))', cs+'%s'%(
'''
sys.frameworks_dir = os.path.join(os.path.dirname(os.environ['RESOURCEPATH']), 'Frameworks')
@ -290,7 +294,7 @@ sys.frameworks_dir = os.path.join(os.path.dirname(os.environ['RESOURCEPATH']), '
f.close()
print
print 'Adding main scripts to site-packages'
f = zipfile.ZipFile(os.path.join(self.dist_dir, APPNAME+'.app', 'Contents', 'Resources', 'lib', 'python2.5', 'site-packages.zip'), 'a', zipfile.ZIP_DEFLATED)
f = zipfile.ZipFile(os.path.join(self.dist_dir, APPNAME+'.app', 'Contents', 'Resources', 'lib', 'python2.6', 'site-packages.zip'), 'a', zipfile.ZIP_DEFLATED)
for script in scripts['gui']+scripts['console']:
f.write(script, script.partition('/')[-1])
f.close()
@ -322,7 +326,8 @@ def main():
'genshi', 'calibre.web.feeds.recipes.*',
'calibre.ebooks.lrf.any.*', 'calibre.ebooks.lrf.feeds.*',
'keyword', 'codeop', 'pydoc', 'readline',
'BeautifulSoup'],
'BeautifulSoup'
],
'packages' : ['PIL', 'Authorization', 'lxml'],
'excludes' : ['IPython'],
'plist' : { 'CFBundleGetInfoString' : '''calibre, an E-book management application.'''

View File

@ -6,13 +6,13 @@ __docformat__ = 'restructuredtext en'
'''
Freeze app into executable using py2exe.
'''
QT_DIR = 'C:\\Qt\\4.4.1'
QT_DIR = 'C:\\Qt\\4.4.3'
LIBUSB_DIR = 'C:\\libusb'
LIBUNRAR = 'C:\\Program Files\\UnrarDLL\\unrar.dll'
PDFTOHTML = 'C:\\pdftohtml\\pdftohtml.exe'
IMAGEMAGICK_DIR = 'C:\\ImageMagick'
FONTCONFIG_DIR = 'C:\\fontconfig'
VC90 = r'C:\Program Files\Microsoft Visual Studio 9.0\VC\redist\x86\Microsoft.VC90.CRT'
import sys, os, py2exe, shutil, zipfile, glob, subprocess, re
from distutils.core import setup
@ -65,6 +65,8 @@ class BuildEXE(py2exe.build_exe.py2exe):
shutil.copyfile(f, os.path.join(self.dist_dir, os.path.basename(f)))
for f in glob.glob(os.path.join(BASE_DIR, 'src', 'calibre', 'plugins', '*.pyd')):
shutil.copyfile(f, os.path.join(tgt, os.path.basename(f)))
for f in glob.glob(os.path.join(BASE_DIR, 'src', 'calibre', 'plugins', '*.manifest')):
shutil.copyfile(f, os.path.join(tgt, os.path.basename(f)))
shutil.copyfile('LICENSE', os.path.join(self.dist_dir, 'LICENSE'))
print
print 'Adding QtXml4.dll'
@ -115,12 +117,17 @@ class BuildEXE(py2exe.build_exe.py2exe):
shutil.copytree(f, tgt)
else:
shutil.copyfile(f, tgt)
print
print 'Doing DLL redirection' # See http://msdn.microsoft.com/en-us/library/ms682600(VS.85).aspx
for f in glob.glob(os.path.join(PY2EXE_DIR, '*.exe')):
open(f + '.local', 'w').write('\n')
print
print 'Adding Windows runtime dependencies...'
for f in glob.glob(os.path.join(VC90, '*')):
shutil.copyfile(f, os.path.join(PY2EXE_DIR, os.path.basename(f)))
@classmethod
def manifest(cls, prog):
@ -142,17 +149,17 @@ def main(args=sys.argv):
{'script' : scripts['gui'][0],
'dest_base' : APPNAME,
'icon_resources' : [(1, ICONS[0])],
'other_resources' : [BuildEXE.manifest(APPNAME)],
#'other_resources' : [BuildEXE.manifest(APPNAME)],
},
{'script' : scripts['gui'][1],
'dest_base' : 'lrfviewer',
'icon_resources' : [(1, ICONS[1])],
'other_resources' : [BuildEXE.manifest('lrfviewer')],
#'other_resources' : [BuildEXE.manifest('lrfviewer')],
},
{'script' : scripts['gui'][2],
'dest_base' : 'ebook-viewer',
'icon_resources' : [(1, ICONS[1])],
'other_resources' : [BuildEXE.manifest('ebook-viewer')],
#'other_resources' : [BuildEXE.manifest('ebook-viewer')],
},
],
console = console,
@ -162,12 +169,12 @@ def main(args=sys.argv):
'includes' : [
'sip', 'pkg_resources', 'PyQt4.QtSvg',
'mechanize', 'ClientForm', 'wmi',
'win32file', 'pythoncom', 'rtf2xml',
'win32file', 'pythoncom',
'win32process', 'win32api', 'msvcrt',
'win32event', 'calibre.ebooks.lrf.any.*',
'calibre.ebooks.lrf.feeds.*',
'genshi', 'BeautifulSoup',
'path', 'pydoc', 'IPython.Extensions.*',
'BeautifulSoup', 'pyreadline',
'pydoc', 'IPython.Extensions.*',
'calibre.web.feeds.recipes.*',
'PyQt4.QtWebKit', 'PyQt4.QtNetwork',
],

View File

@ -6,7 +6,7 @@ __docformat__ = 'restructuredtext en'
'''
Build PyQt extensions. Integrates with distutils (but uses the PyQt build system).
'''
from distutils.core import Extension
from distutils.core import Extension as _Extension
from distutils.command.build_ext import build_ext as _build_ext
from distutils.dep_util import newer_group
from distutils import log
@ -15,12 +15,23 @@ import sipconfig, os, sys, string, glob, shutil
from PyQt4 import pyqtconfig
iswindows = 'win32' in sys.platform
QMAKE = os.path.expanduser('~/qt/bin/qmake') if 'darwin' in sys.platform else'qmake'
WINDOWS_PYTHON = ['C:/Python25/libs']
WINDOWS_PYTHON = ['C:/Python26/libs']
OSX_SDK = '/Developer/SDKs/MacOSX10.4u.sdk'
def replace_suffix(path, new_suffix):
return os.path.splitext(path)[0] + new_suffix
class Extension(_Extension):
pass
if iswindows:
from distutils import msvc9compiler
msvc = msvc9compiler.MSVCCompiler()
msvc.initialize()
nmake = msvc.find_exe('nmake.exe')
rc = msvc.find_exe('rc.exe')
class PyQtExtension(Extension):
def __init__(self, name, sources, sip_sources, **kw):
@ -37,9 +48,7 @@ class PyQtExtension(Extension):
class build_ext(_build_ext):
def make(self, makefile):
make = 'make'
if iswindows:
make = 'mingw32-make'
make = nmake if iswindows else 'make'
self.spawn([make, '-f', makefile])
def build_qt_objects(self, ext, bdir):
@ -65,12 +74,13 @@ CONFIG += x86 ppc
open(name+'.pro', 'wb').write(pro)
self.spawn([QMAKE, '-o', 'Makefile.qt', name+'.pro'])
self.make('Makefile.qt')
pat = 'release\\*.o' if iswindows else '*.o'
pat = 'release\\*.obj' if iswindows else '*.o'
return map(os.path.abspath, glob.glob(pat))
finally:
os.chdir(cwd)
def build_sbf(self, sip, sbf, bdir):
print '\tBuilding spf...'
sip_bin = self.sipcfg.sip_bin
self.spawn([sip_bin,
"-c", bdir,
@ -100,9 +110,7 @@ CONFIG += x86 ppc
def build_extension(self, ext):
self.inplace = True # Causes extensions to be built in the source tree
if not isinstance(ext, PyQtExtension):
return _build_ext.build_extension(self, ext)
fullname = self.get_ext_fullname(ext.name)
if self.inplace:
# ignore build-lib -- put the compiled extension into
@ -119,7 +127,38 @@ CONFIG += x86 ppc
else:
ext_filename = os.path.join(self.build_lib,
self.get_ext_filename(fullname))
bdir = os.path.abspath(os.path.join(self.build_temp, fullname))
bdir = os.path.abspath(os.path.join(self.build_temp, fullname))
if not os.path.exists(bdir):
os.makedirs(bdir)
if not isinstance(ext, PyQtExtension):
if not iswindows:
return _build_ext.build_extension(self, ext)
c_sources = [f for f in ext.sources if os.path.splitext(f)[1].lower() in ('.c', '.cpp', '.cxx')]
compile_args = '/c /nologo /Ox /MD /W3 /GX /DNDEBUG'.split()
compile_args += ext.extra_compile_args
self.swig_opts = ''
inc_dirs = self.include_dirs + [x.replace('/', '\\') for x in ext.include_dirs]
cc = [msvc.cc] + compile_args + ['-I%s'%x for x in list(set(inc_dirs))]
objects = []
for f in c_sources:
o = os.path.join(bdir, os.path.basename(f)+'.obj')
objects.append(o)
compiler = cc + ['/Tc'+f, '/Fo'+o]
self.spawn(compiler)
out = os.path.join(bdir, base+'.pyd')
linker = [msvc.linker] + '/DLL /nologo /INCREMENTAL:NO'.split()
linker += ['/LIBPATH:'+x for x in self.library_dirs]
linker += [x+'.lib' for x in ext.libraries]
linker += ['/EXPORT:init'+base] + objects + ['/OUT:'+out]
self.spawn(linker)
for src in (out, out+'.manifest'):
shutil.copyfile(src, os.path.join('src', 'calibre', 'plugins', os.path.basename(src)))
return
if not os.path.exists(bdir):
os.makedirs(bdir)
ext.sources2 = map(os.path.abspath, ext.sources)

View File

@ -46,10 +46,11 @@ main_functions = {
}
if __name__ == '__main__':
from setuptools import setup, find_packages, Extension
from setuptools import setup, find_packages
from setuptools.command.build_py import build_py as _build_py, convert_path
from distutils.command.build import build as _build
from distutils.core import Command as _Command
from pyqtdistutils import PyQtExtension, build_ext
from pyqtdistutils import PyQtExtension, build_ext, Extension
import subprocess, glob
def newer(targets, sources):
@ -65,6 +66,25 @@ if __name__ == '__main__':
newest_source, oldest_target = max(stimes), min(ttimes)
return newest_source > oldest_target
class build_py(_build_py):
def find_data_files(self, package, src_dir):
"""
Return filenames for package's data files in 'src_dir'
Modified to treat data file specs as paths not globs
"""
globs = (self.package_data.get('', [])
+ self.package_data.get(package, []))
files = self.manifest_files.get(package, [])[:]
for pattern in globs:
# Each pattern has to be converted to a platform-specific path
pattern = os.path.join(src_dir, convert_path(pattern))
next = glob.glob(pattern)
files.extend(next if next else [pattern])
return self.exclude_data_files(package, src_dir, files)
class Command(_Command):
user_options = []
def initialize_options(self): pass
@ -146,6 +166,7 @@ if __name__ == '__main__':
metadata_sqlite = 'library/metadata_sqlite.sql',
jquery = 'gui2/viewer/jquery.js',
jquery_scrollTo = 'gui2/viewer/jquery_scrollTo.js',
html_css = 'ebooks/lit/html.css',
)
DEST = os.path.join('src', APPNAME, 'resources.py')
@ -251,6 +272,7 @@ if __name__ == '__main__':
description='''Compile all GUI forms and images'''
PATH = os.path.join('src', APPNAME, 'gui2')
IMAGES_DEST = os.path.join(PATH, 'images_rc.py')
QRC = os.path.join(PATH, 'images.qrc')
@classmethod
def find_forms(cls):
@ -330,9 +352,9 @@ if __name__ == '__main__':
c = cls.form_to_compiled_form(form)
if os.path.exists(c):
os.remove(c)
images = cls.IMAGES_DEST
if os.path.exists(images):
os.remove(images)
for x in (cls.IMAGES_DEST, cls.QRC):
if os.path.exists(x):
os.remove(x)
class clean(Command):
description='''Delete all computer generated files in the source tree'''
@ -348,32 +370,34 @@ if __name__ == '__main__':
os.remove(f)
for root, dirs, files in os.walk('.'):
for name in files:
if name.endswith('~') or \
name.endswith('.pyc') or \
name.endswith('.pyo'):
os.remove(os.path.join(root, name))
for t in ('.pyc', '.pyo', '~'):
if name.endswith(t):
os.remove(os.path.join(root, name))
break
for dir in 'build', 'dist':
for f in os.listdir(dir):
if os.path.isdir(dir + os.sep + f):
shutil.rmtree(dir + os.sep + f)
else:
os.remove(dir + os.sep + f)
for dir in ('build', 'dist', os.path.join('src', 'calibre.egg-info')):
shutil.rmtree(dir, ignore_errors=True)
class build(_build):
sub_commands = \
[
sub_commands = [
('resources', lambda self : 'CALIBRE_BUILDBOT' not in os.environ.keys()),
('translations', lambda self : 'CALIBRE_BUILDBOT' not in os.environ.keys()),
('gui', lambda self : 'CALIBRE_BUILDBOT' not in os.environ.keys()),
] + _build.sub_commands
('build_ext', lambda self: True),
('build_py', lambda self: True),
('build_clib', _build.has_c_libraries),
('build_scripts', _build.has_scripts),
]
entry_points['console_scripts'].append('calibre_postinstall = calibre.linux:post_install')
ext_modules = [
Extension('calibre.plugins.lzx',
sources=['src/calibre/utils/lzx/lzxmodule.c',
'src/calibre/utils/lzx/lzxd.c'],
'src/calibre/utils/lzx/compressor.c',
'src/calibre/utils/lzx/lzxd.c',
'src/calibre/utils/lzx/lzc.c',
'src/calibre/utils/lzx/lzxc.c'],
include_dirs=['src/calibre/utils/lzx']),
Extension('calibre.plugins.msdes',
@ -391,13 +415,18 @@ if __name__ == '__main__':
ext_modules.append(Extension('calibre.plugins.winutil',
sources=['src/calibre/utils/windows/winutil.c'],
libraries=['shell32', 'setupapi'],
include_dirs=['C:/WinDDK/6001.18001/inc/api/'])
)
include_dirs=['C:/WinDDK/6001.18001/inc/api/',
'C:/WinDDK/6001.18001/inc/crt/'],
extra_compile_args=['/X']
))
if isosx:
ext_modules.append(Extension('calibre.plugins.usbobserver',
sources=['src/calibre/devices/usbobserver/usbobserver.c'])
sources=['src/calibre/devices/usbobserver/usbobserver.c'],
extra_link_args=['-framework', 'IOKit'])
)
plugins = ['plugins/%s.so'%(x.name.rpartition('.')[-1]) for x in ext_modules]
setup(
name = APPNAME,
packages = find_packages('src'),
@ -406,8 +435,7 @@ if __name__ == '__main__':
author = 'Kovid Goyal',
author_email = 'kovid@kovidgoyal.net',
url = 'http://%s.kovidgoyal.net'%APPNAME,
package_data = {'calibre':['plugins/*']},
include_package_data = True,
package_data = {'calibre':plugins},
entry_points = entry_points,
zip_safe = False,
options = { 'bdist_egg' : {'exclude_source_files': True,}, },
@ -448,7 +476,8 @@ if __name__ == '__main__':
],
cmdclass = {
'build_ext' : build_ext,
'build' : build,
'build' : build,
'build_py' : build_py,
'pot' : pot,
'manual' : manual,
'resources' : resources,

View File

@ -13,13 +13,19 @@ from calibre.startup import plugins, winutil, winutilerror
from calibre.constants import iswindows, isosx, islinux, isfrozen, \
terminal_controller, preferred_encoding, \
__appname__, __version__, __author__, \
win32event, win32api, winerror, fcntl
win32event, win32api, winerror, fcntl, \
filesystem_encoding
import mechanize
mimetypes.add_type('application/epub+zip', '.epub')
mimetypes.add_type('text/x-sony-bbeb+xml', '.lrs')
mimetypes.add_type('application/x-sony-bbeb', '.lrf')
def to_unicode(raw, encoding='utf-8', errors='strict'):
if isinstance(raw, unicode):
return raw
return raw.decode(encoding, errors)
def unicode_path(path, abs=False):
if not isinstance(path, unicode):
path = path.decode(sys.getfilesystemencoding())
@ -36,6 +42,28 @@ def osx_version():
return int(m.group(1)), int(m.group(2)), int(m.group(3))
_filename_sanitize = re.compile(r'[\xae\0\\|\?\*<":>\+\[\]/]')
def sanitize_file_name(name, substitute='_', as_unicode=False):
'''
Sanitize the filename `name`. All invalid characters are replaced by `substitute`.
The set of invalid characters is the union of the invalid characters in Windows,
OS X and Linux. Also removes leading an trailing whitespace.
**WARNING:** This function also replaces path separators, so only pass file names
and not full paths to it.
*NOTE:* This function always returns byte strings, not unicode objects. The byte strings
are encoded in the filesystem encoding of the platform, or UTF-8.
'''
if isinstance(name, unicode):
name = name.encode(filesystem_encoding, 'ignore')
one = _filename_sanitize.sub(substitute, name)
one = re.sub(r'\s', ' ', one).strip()
one = re.sub(r'^\.+$', '_', one)
if as_unicode:
one = one.decode(filesystem_encoding)
return one
class CommandLineError(Exception):
pass
@ -196,13 +224,6 @@ class CurrentDir(object):
def __exit__(self, *args):
os.chdir(self.cwd)
def sanitize_file_name(name):
'''
Remove characters that are illegal in filenames from name.
Also remove path separators. All illegal characters are replaced by
underscores.
'''
return re.sub(r'\s', ' ', re.sub(r'[\xae"\'\|\~\:\?\\\/]|^-', '_', name.strip()))
def detect_ncpus():
"""Detects the number of effective CPUs in the system"""
@ -317,7 +338,12 @@ class LoggingInterface:
def ___log(self, func, msg, args, kwargs):
args = [msg] + list(args)
for i in range(len(args)):
if isinstance(args[i], unicode):
if not isinstance(args[i], basestring):
continue
if sys.version_info[:2] > (2, 5):
if not isinstance(args[i], unicode):
args[i] = args[i].decode(preferred_encoding, 'replace')
elif isinstance(args[i], unicode):
args[i] = args[i].encode(preferred_encoding, 'replace')
func(*args, **kwargs)
@ -355,7 +381,13 @@ def strftime(fmt, t=time.localtime()):
fmt = fmt.encode('mbcs')
return plugins['winutil'][0].strftime(fmt, t)
return time.strftime(fmt, t).decode(preferred_encoding, 'replace')
def my_unichr(num):
try:
return unichr(num)
except ValueError:
return u'?'
def entity_to_unicode(match, exceptions=[], encoding='cp1252'):
'''
@param match: A match object such that '&'+match.group(1)';' is the entity.
@ -371,7 +403,7 @@ def entity_to_unicode(match, exceptions=[], encoding='cp1252'):
if ent.startswith(u'#x'):
num = int(ent[2:], 16)
if encoding is None or num > 255:
return unichr(num)
return my_unichr(num)
return chr(num).decode(encoding)
if ent.startswith(u'#'):
try:
@ -379,13 +411,13 @@ def entity_to_unicode(match, exceptions=[], encoding='cp1252'):
except ValueError:
return '&'+ent+';'
if encoding is None or num > 255:
return unichr(num)
return my_unichr(num)
try:
return chr(num).decode(encoding)
except UnicodeDecodeError:
return unichr(num)
return my_unichr(num)
try:
return unichr(name2codepoint[ent])
return my_unichr(name2codepoint[ent])
except KeyError:
return '&'+ent+';'

View File

@ -2,7 +2,7 @@ __license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
__docformat__ = 'restructuredtext en'
__appname__ = 'calibre'
__version__ = '0.4.109'
__version__ = '0.4.125'
__author__ = "Kovid Goyal <kovid@kovidgoyal.net>"
'''
Various run time constants.
@ -29,6 +29,10 @@ winerror = __import__('winerror') if iswindows else None
win32api = __import__('win32api') if iswindows else None
fcntl = None if iswindows else __import__('fcntl')
filesystem_encoding = sys.getfilesystemencoding()
if filesystem_encoding is None: filesystem_encoding = 'utf-8'
################################################################################
plugins = None
if plugins is None:

View File

@ -0,0 +1,223 @@
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
import sys
from calibre.ptempfile import PersistentTemporaryFile
from calibre.constants import __version__, __author__
class Plugin(object):
'''
A calibre plugin. Useful members include:
* ``self.plugin_path``: Stores path to the zip file that contains
this plugin or None if it is a builtin
plugin
* ``self.site_customization``: Stores a customization string entered
by the user.
Methods that should be overridden in sub classes:
* :meth:`initialize`
* :meth:`customization_help`
Useful methods:
* :meth:`temporary_file`
'''
#: List of platforms this plugin works on
#: For example: ``['windows', 'osx', 'linux']
supported_platforms = []
#: The name of this plugin
name = 'Trivial Plugin'
#: The version of this plugin as a 3-tuple (major, minor, revision)
version = (1, 0, 0)
#: A short string describing what this plugin does
description = _('Does absolutely nothing')
#: The author of this plugin
author = _('Unknown')
#: When more than one plugin exists for a filetype,
#: the plugins are run in order of decreasing priority
#: i.e. plugins with higher priority will be run first.
#: The highest possible priority is ``sys.maxint``.
#: Default pririty is 1.
priority = 1
#: The earliest version of calibre this plugin requires
minimum_calibre_version = (0, 4, 118)
#: If False, the user will not be able to disable this plugin. Use with
#: care.
can_be_disabled = True
#: The type of this plugin. Used for categorizing plugins in the
#: GUI
type = _('Base')
def __init__(self, plugin_path):
self.plugin_path = plugin_path
self.site_customization = None
def initialize(self):
'''
Called once when calibre plugins are initialized. Plugins are re-initialized
every time a new plugin is added.
Perform any plugin specific initialization here, such as extracting
resources from the plugin zip file. The path to the zip file is
available as ``self.plugin_path``.
Note that ``self.site_customization`` is **not** available at this point.
'''
pass
def customization_help(self, gui=False):
'''
Return a string giving help on how to customize this plugin.
By default raise a :class:`NotImplementedError`, which indicates that
the plugin does not require customization.
If you re-implement this method in your subclass, the user will
be asked to enter a string as customization for this plugin.
The customization string will be available as
``self.site_customization``.
Site customization could be anything, for example, the path to
a needed binary on the user's computer.
:param gui: If True return HTML help, otherwise return plain text help.
'''
raise NotImplementedError
def temporary_file(self, suffix):
'''
Return a file-like object that is a temporary file on the file system.
This file will remain available even after being closed and will only
be removed on interpreter shutdown. Use the ``name`` member of the
returned object to access the full path to the created temporary file.
:param suffix: The suffix that the temporary file will have.
'''
return PersistentTemporaryFile(suffix)
def is_customizable(self):
try:
self.customization_help()
return True
except NotImplementedError:
return False
def __enter__(self, *args):
if self.plugin_path is not None:
sys.path.insert(0, self.plugin_path)
def __exit__(self, *args):
if self.plugin_path in sys.path:
sys.path.remove(self.plugin_path)
class FileTypePlugin(Plugin):
'''
A plugin that is associated with a particular set of file types.
'''
#: Set of file types for which this plugin should be run
#: For example: ``set(['lit', 'mobi', 'prc'])``
file_types = set([])
#: If True, this plugin is run when books are added
#: to the database
on_import = False
#: If True, this plugin is run whenever an any2* tool
#: is used, on the file passed to the any2* tool.
on_preprocess = False
#: If True, this plugin is run after an any2* tool is
#: used, on the final file produced by the tool.
on_postprocess = False
type = _('File type')
def run(self, path_to_ebook):
'''
Run the plugin. Must be implemented in subclasses.
It should perform whatever modifications are required
on the ebook and return the absolute path to the
modified ebook. If no modifications are needed, it should
return the path to the original ebook. If an error is encountered
it should raise an Exception. The default implementation
simply return the path to the original ebook.
The modified ebook file should be created with the
:meth:`temporary_file` method.
:param path_to_ebook: Absolute path to the ebook.
:return: Absolute path to the modified ebook.
'''
# Default implementation does nothing
return path_to_ebook
class MetadataReaderPlugin(Plugin):
'''
A plugin that implements reading metadata from a set of file types.
'''
#: Set of file types for which this plugin should be run
#: For example: ``set(['lit', 'mobi', 'prc'])``
file_types = set([])
supported_platforms = ['windows', 'osx', 'linux']
version = tuple(map(int, (__version__.split('.'))[:3]))
author = 'Kovid Goyal'
type = _('Metadata reader')
def get_metadata(self, stream, type):
'''
Return metadata for the file represented by stream (a file like object
that supports reading). Raise an exception when there is an error
with the input data.
:param type: The type of file. Guaranteed to be one of the entries
in :attr:`file_types`.
:return: A :class:`calibre.ebooks.metadata.MetaInformation` object
'''
return None
class MetadataWriterPlugin(Plugin):
'''
A plugin that implements reading metadata from a set of file types.
'''
#: Set of file types for which this plugin should be run
#: For example: ``set(['lit', 'mobi', 'prc'])``
file_types = set([])
supported_platforms = ['windows', 'osx', 'linux']
version = tuple(map(int, (__version__.split('.'))[:3]))
author = 'Kovid Goyal'
type = _('Metadata writer')
def set_metadata(self, stream, mi, type):
'''
Set metadata for the file represented by stream (a file like object
that supports reading). Raise an exception when there is an error
with the input data.
:param type: The type of file. Guaranteed to be one of the entries
in :attr:`file_types`.
:param mi: A :class:`calibre.ebooks.metadata.MetaInformation` object
'''
pass

View File

@ -0,0 +1,205 @@
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
import textwrap, os
from calibre.customize import FileTypePlugin, MetadataReaderPlugin, MetadataWriterPlugin
from calibre.constants import __version__
class HTML2ZIP(FileTypePlugin):
name = 'HTML to ZIP'
author = 'Kovid Goyal'
description = textwrap.dedent(_('''\
Follow all local links in an HTML file and create a ZIP \
file containing all linked files. This plugin is run \
every time you add an HTML file to the library.\
'''))
version = tuple(map(int, (__version__.split('.'))[:3]))
file_types = set(['html', 'htm', 'xhtml', 'xhtm'])
supported_platforms = ['windows', 'osx', 'linux']
on_import = True
def run(self, htmlfile):
of = self.temporary_file('_plugin_html2zip.zip')
from calibre.ebooks.html import gui_main as html2oeb
html2oeb(htmlfile, of)
return of.name
class RTFMetadataReader(MetadataReaderPlugin):
name = 'Read RTF metadata'
file_types = set(['rtf'])
description = _('Read metadata from %s files')%'RTF'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.rtf import get_metadata
return get_metadata(stream)
class FB2MetadataReader(MetadataReaderPlugin):
name = 'Read FB2 metadata'
file_types = set(['fb2'])
description = _('Read metadata from %s files')%'FB2'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.fb2 import get_metadata
return get_metadata(stream)
class LRFMetadataReader(MetadataReaderPlugin):
name = 'Read LRF metadata'
file_types = set(['lrf'])
description = _('Read metadata from %s files')%'LRF'
def get_metadata(self, stream, ftype):
from calibre.ebooks.lrf.meta import get_metadata
return get_metadata(stream)
class PDFMetadataReader(MetadataReaderPlugin):
name = 'Read PDF metadata'
file_types = set(['pdf'])
description = _('Read metadata from %s files')%'PDF'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.pdf import get_metadata
return get_metadata(stream)
class LITMetadataReader(MetadataReaderPlugin):
name = 'Read LIT metadata'
file_types = set(['lit'])
description = _('Read metadata from %s files')%'LIT'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.lit import get_metadata
return get_metadata(stream)
class IMPMetadataReader(MetadataReaderPlugin):
name = 'Read IMP metadata'
file_types = set(['imp'])
description = _('Read metadata from %s files')%'IMP'
author = 'Ashish Kulkarni'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.imp import get_metadata
return get_metadata(stream)
class RBMetadataReader(MetadataReaderPlugin):
name = 'Read RB metadata'
file_types = set(['rb'])
description = _('Read metadata from %s files')%'RB'
author = 'Ashish Kulkarni'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.rb import get_metadata
return get_metadata(stream)
class EPUBMetadataReader(MetadataReaderPlugin):
name = 'Read EPUB metadata'
file_types = set(['epub'])
description = _('Read metadata from %s files')%'EPUB'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.epub import get_metadata
return get_metadata(stream)
class HTMLMetadataReader(MetadataReaderPlugin):
name = 'Read HTML metadata'
file_types = set(['html'])
description = _('Read metadata from %s files')%'HTML'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.html import get_metadata
return get_metadata(stream)
class MOBIMetadataReader(MetadataReaderPlugin):
name = 'Read MOBI metadata'
file_types = set(['mobi'])
description = _('Read metadata from %s files')%'MOBI'
def get_metadata(self, stream, ftype):
from calibre.ebooks.mobi.reader import get_metadata
return get_metadata(stream)
class ODTMetadataReader(MetadataReaderPlugin):
name = 'Read ODT metadata'
file_types = set(['odt'])
description = _('Read metadata from %s files')%'ODT'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.odt import get_metadata
return get_metadata(stream)
class LRXMetadataReader(MetadataReaderPlugin):
name = 'Read LRX metadata'
file_types = set(['lrx'])
description = _('Read metadata from %s files')%'LRX'
def get_metadata(self, stream, ftype):
from calibre.ebooks.metadata.lrx import get_metadata
return get_metadata(stream)
class ComicMetadataReader(MetadataReaderPlugin):
name = 'Read comic metadata'
file_types = set(['cbr', 'cbz'])
description = _('Extract cover from comic files')
def get_metadata(self, stream, ftype):
if ftype == 'cbr':
from calibre.libunrar import extract_member as extract_first
else:
from calibre.libunzip import extract_member as extract_first
from calibre.ebooks.metadata import MetaInformation
ret = extract_first(stream)
mi = MetaInformation(None, None)
if ret is not None:
path, data = ret
ext = os.path.splitext(path)[1][1:]
mi.cover_data = (ext.lower(), data)
return mi
class EPUBMetadataWriter(MetadataWriterPlugin):
name = 'Set EPUB metadata'
file_types = set(['epub'])
description = _('Set metadata in EPUB files')
def set_metadata(self, stream, mi, type):
from calibre.ebooks.metadata.epub import set_metadata
set_metadata(stream, mi)
class LRFMetadataWriter(MetadataWriterPlugin):
name = 'Set LRF metadata'
file_types = set(['lrf'])
description = _('Set metadata in LRF files')
def set_metadata(self, stream, mi, type):
from calibre.ebooks.lrf.meta import set_metadata
set_metadata(stream, mi)
class RTFMetadataWriter(MetadataWriterPlugin):
name = 'Set RTF metadata'
file_types = set(['rtf'])
description = _('Set metadata in RTF files')
def set_metadata(self, stream, mi, type):
from calibre.ebooks.metadata.rtf import set_metadata
set_metadata(stream, mi)
plugins = [HTML2ZIP]
plugins += [x for x in list(locals().values()) if isinstance(x, type) and \
x.__name__.endswith('MetadataReader')]
plugins += [x for x in list(locals().values()) if isinstance(x, type) and \
x.__name__.endswith('MetadataWriter')]

320
src/calibre/customize/ui.py Normal file
View File

@ -0,0 +1,320 @@
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
import os, shutil, traceback, functools, sys
from calibre.customize import Plugin, FileTypePlugin, MetadataReaderPlugin, \
MetadataWriterPlugin
from calibre.customize.builtins import plugins as builtin_plugins
from calibre.constants import __version__, iswindows, isosx
from calibre.ebooks.metadata import MetaInformation
from calibre.utils.config import make_config_dir, Config, ConfigProxy, \
plugin_dir, OptionParser
version = tuple([int(x) for x in __version__.split('.')])
platform = 'linux'
if iswindows:
platform = 'windows'
if isosx:
platform = 'osx'
from zipfile import ZipFile
def _config():
c = Config('customize')
c.add_opt('plugins', default={}, help=_('Installed plugins'))
c.add_opt('filetype_mapping', default={}, help=_('Mapping for filetype plugins'))
c.add_opt('plugin_customization', default={}, help=_('Local plugin customization'))
c.add_opt('disabled_plugins', default=set([]), help=_('Disabled plugins'))
return ConfigProxy(c)
config = _config()
class InvalidPlugin(ValueError):
pass
class PluginNotFound(ValueError):
pass
def load_plugin(path_to_zip_file):
'''
Load plugin from zip file or raise InvalidPlugin error
:return: A :class:`Plugin` instance.
'''
print 'Loading plugin from', path_to_zip_file
if not os.access(path_to_zip_file, os.R_OK):
raise PluginNotFound
zf = ZipFile(path_to_zip_file)
for name in zf.namelist():
if name.lower().endswith('plugin.py'):
locals = {}
exec zf.read(name) in locals
for x in locals.values():
if isinstance(x, type) and issubclass(x, Plugin):
if x.minimum_calibre_version > version or \
platform not in x.supported_platforms:
continue
return x
raise InvalidPlugin(_('No valid plugin found in ')+path_to_zip_file)
_initialized_plugins = []
_on_import = {}
_on_preprocess = {}
_on_postprocess = {}
def reread_filetype_plugins():
global _on_import
global _on_preprocess
global _on_postprocess
_on_import = {}
_on_preprocess = {}
_on_postprocess = {}
for plugin in _initialized_plugins:
if isinstance(plugin, FileTypePlugin):
for ft in plugin.file_types:
if plugin.on_import:
if not _on_import.has_key(ft):
_on_import[ft] = []
_on_import[ft].append(plugin)
if plugin.on_preprocess:
if not _on_preprocess.has_key(ft):
_on_preprocess[ft] = []
_on_preprocess[ft].append(plugin)
if plugin.on_postprocess:
if not _on_postprocess.has_key(ft):
_on_postprocess[ft] = []
_on_postprocess[ft].append(plugin)
_metadata_readers = {}
_metadata_writers = {}
def reread_metadata_plugins():
global _metadata_readers
global _metadata_writers
_metadata_readers = {}
for plugin in _initialized_plugins:
if isinstance(plugin, MetadataReaderPlugin):
for ft in plugin.file_types:
_metadata_readers[ft] = plugin
elif isinstance(plugin, MetadataWriterPlugin):
for ft in plugin.file_types:
_metadata_writers[ft] = plugin
def get_file_type_metadata(stream, ftype):
mi = MetaInformation(None, None)
try:
plugin = _metadata_readers[ftype.lower().strip()]
if not is_disabled(plugin):
with plugin:
mi = plugin.get_metadata(stream, ftype.lower().strip())
except:
pass
return mi
def set_file_type_metadata(stream, mi, ftype):
try:
plugin = _metadata_writers[ftype.lower().strip()]
if not is_disabled(plugin):
with plugin:
plugin.set_metadata(stream, mi, ftype.lower().strip())
except:
traceback.print_exc()
def _run_filetype_plugins(path_to_file, ft=None, occasion='preprocess'):
occasion = {'import':_on_import, 'preprocess':_on_preprocess,
'postprocess':_on_postprocess}[occasion]
customization = config['plugin_customization']
if ft is None:
ft = os.path.splitext(path_to_file)[-1].lower().replace('.', '')
nfp = path_to_file
for plugin in occasion.get(ft, []):
if is_disabled(plugin):
continue
plugin.site_customization = customization.get(plugin.name, '')
with plugin:
try:
nfp = plugin.run(path_to_file)
except:
print 'Running file type plugin %s failed with traceback:'%plugin.name
traceback.print_exc()
x = lambda j : os.path.normpath(os.path.normcase(j))
if occasion == 'postprocess' and x(nfp) != x(path_to_file):
shutil.copyfile(nfp, path_to_file)
nfp = path_to_file
return nfp
run_plugins_on_import = functools.partial(_run_filetype_plugins,
occasion='import')
run_plugins_on_preprocess = functools.partial(_run_filetype_plugins,
occasion='preprocess')
run_plugins_on_postprocess = functools.partial(_run_filetype_plugins,
occasion='postprocess')
def initialize_plugin(plugin, path_to_zip_file):
try:
return plugin(path_to_zip_file)
except Exception:
print 'Failed to initialize plugin:', plugin.name, plugin.version
tb = traceback.format_exc()
raise InvalidPlugin((_('Initialization of plugin %s failed with traceback:')
%tb) + '\n'+tb)
def add_plugin(path_to_zip_file):
make_config_dir()
plugin = load_plugin(path_to_zip_file)
plugin = initialize_plugin(plugin, path_to_zip_file)
plugins = config['plugins']
zfp = os.path.join(plugin_dir, plugin.name+'.zip')
if os.path.exists(zfp):
os.remove(zfp)
shutil.copyfile(path_to_zip_file, zfp)
plugins[plugin.name] = zfp
config['plugins'] = plugins
initialize_plugins()
return plugin
def remove_plugin(plugin_or_name):
name = getattr(plugin_or_name, 'name', plugin_or_name)
plugins = config['plugins']
removed = False
if name in plugins.keys():
removed = True
zfp = plugins[name]
if os.path.exists(zfp):
os.remove(zfp)
plugins.pop(name)
config['plugins'] = plugins
initialize_plugins()
return removed
def is_disabled(plugin):
return plugin.name in config['disabled_plugins']
def find_plugin(name):
for plugin in _initialized_plugins:
if plugin.name == name:
return plugin
def disable_plugin(plugin_or_name):
x = getattr(plugin_or_name, 'name', plugin_or_name)
plugin = find_plugin(x)
if not plugin.can_be_disabled:
raise ValueError('Plugin %s cannot be disabled'%x)
dp = config['disabled_plugins']
dp.add(x)
config['disabled_plugins'] = dp
def enable_plugin(plugin_or_name):
x = getattr(plugin_or_name, 'name', plugin_or_name)
dp = config['disabled_plugins']
if x in dp:
dp.remove(x)
config['disabled_plugins'] = dp
def initialize_plugins():
global _initialized_plugins
_initialized_plugins = []
for zfp in list(config['plugins'].values()) + builtin_plugins:
try:
try:
plugin = load_plugin(zfp) if not isinstance(zfp, type) else zfp
except PluginNotFound:
continue
plugin = initialize_plugin(plugin, zfp if not isinstance(zfp, type) else zfp)
_initialized_plugins.append(plugin)
except:
print 'Failed to initialize plugin...'
traceback.print_exc()
_initialized_plugins.sort(cmp=lambda x,y:cmp(x.priority, y.priority), reverse=True)
reread_filetype_plugins()
reread_metadata_plugins()
initialize_plugins()
def option_parser():
parser = OptionParser(usage=_('''\
%prog options
Customize calibre by loading external plugins.
'''))
parser.add_option('-a', '--add-plugin', default=None,
help=_('Add a plugin by specifying the path to the zip file containing it.'))
parser.add_option('-r', '--remove-plugin', default=None,
help=_('Remove a custom plugin by name. Has no effect on builtin plugins'))
parser.add_option('--customize-plugin', default=None,
help=_('Customize plugin. Specify name of plugin and customization string separated by a comma.'))
parser.add_option('-l', '--list-plugins', default=False, action='store_true',
help=_('List all installed plugins'))
parser.add_option('--enable-plugin', default=None,
help=_('Enable the named plugin'))
parser.add_option('--disable-plugin', default=None,
help=_('Disable the named plugin'))
return parser
def initialized_plugins():
return _initialized_plugins
def customize_plugin(plugin, custom):
d = config['plugin_customization']
d[plugin.name] = custom.strip()
config['plugin_customization'] = d
def plugin_customization(plugin):
return config['plugin_customization'].get(plugin.name, '')
def main(args=sys.argv):
parser = option_parser()
if len(args) < 2:
parser.print_help()
return 1
opts, args = parser.parse_args(args)
if opts.add_plugin is not None:
plugin = add_plugin(opts.add_plugin)
print 'Plugin added:', plugin.name, plugin.version
if opts.remove_plugin is not None:
if remove_plugin(opts.remove_plugin):
print 'Plugin removed'
else:
print 'No custom pluginnamed', opts.remove_plugin
if opts.customize_plugin is not None:
name, custom = opts.customize_plugin.split(',')
plugin = find_plugin(name.strip())
if plugin is None:
print 'No plugin with the name %s exists'%name
return 1
customize_plugin(plugin, custom)
if opts.enable_plugin is not None:
enable_plugin(opts.enable_plugin.strip())
if opts.disable_plugin is not None:
disable_plugin(opts.disable_plugin.strip())
if opts.list_plugins:
fmt = '%-15s%-20s%-15s%-15s%s'
print fmt%tuple(('Type|Name|Version|Disabled|Site Customization'.split('|')))
print
for plugin in initialized_plugins():
print fmt%(
plugin.type, plugin.name,
plugin.version, is_disabled(plugin),
plugin_customization(plugin)
)
print '\t', plugin.description
if plugin.is_customizable():
print '\t', plugin.customization_help()
print
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@ -21,8 +21,10 @@ Run an embedded python interpreter.
'Module specifications are of the form full.name.of.module,path_to_module.py', default=None
)
parser.add_option('-c', '--command', help='Run python code.', default=None)
parser.add_option('-g', '--gui', default=False, action='store_true',
help='Run the GUI',)
parser.add_option('--migrate', action='store_true', default=False,
help='Migrate old database. Needs two arguments. Path to library1.db and path to new library folder.', default=False)
help='Migrate old database. Needs two arguments. Path to library1.db and path to new library folder.')
return parser
def update_zipfile(zipfile, mod, path):
@ -72,7 +74,10 @@ def migrate(old, new):
def main(args=sys.argv):
opts, args = option_parser().parse_args(args)
if opts.update_module:
if opts.gui:
from calibre.gui2.main import main
main(['calibre'])
elif opts.update_module:
mod, path = opts.update_module.partition(',')[0], opts.update_module.partition(',')[-1]
update_module(mod, os.path.expanduser(path))
elif opts.command:

View File

@ -1,5 +1,6 @@
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
'''
Device drivers.
'''
@ -8,8 +9,9 @@ def devices():
from calibre.devices.prs500.driver import PRS500
from calibre.devices.prs505.driver import PRS505
from calibre.devices.prs700.driver import PRS700
from calibre.devices.cybookg3.driver import CYBOOKG3
#from calibre.devices.kindle.driver import KINDLE
return (PRS500, PRS505, PRS700)
return (PRS500, PRS505, PRS700, CYBOOKG3)
import time

View File

View File

@ -0,0 +1,78 @@
__license__ = 'GPL v3'
__copyright__ = '2009, John Schember <john at nachtimwald.com'
'''
'''
import os, fnmatch, time
from calibre.devices.interface import BookList as _BookList
EBOOK_DIR = "eBooks"
EBOOK_TYPES = ['mobi', 'prc', 'html', 'pdf', 'rtf', 'txt']
class Book(object):
def __init__(self, path, title, authors):
self.title = title
self.authors = authors
self.size = os.path.getsize(path)
self.datetime = time.gmtime(os.path.getctime(path))
self.path = path
self.thumbnail = None
self.tags = []
@apply
def thumbnail():
return None
def __str__(self):
""" Return a utf-8 encoded string with title author and path information """
return self.title.encode('utf-8') + " by " + \
self.authors.encode('utf-8') + " at " + self.path.encode('utf-8')
class BookList(_BookList):
def __init__(self, mountpath):
self._mountpath = mountpath
_BookList.__init__(self)
self.return_books(mountpath)
def return_books(self, mountpath):
# Get all books in all directories under the root EBOOK_DIR directory
for path, dirs, files in os.walk(os.path.join(mountpath, EBOOK_DIR)):
# Filter out anything that isn't in the list of supported ebook types
for book_type in EBOOK_TYPES:
for filename in fnmatch.filter(files, '*.%s' % (book_type)):
book_title = ''
book_author = ''
# Calibre uses a specific format for file names. They take the form
# title_-_author_number.extention We want to see if the file name is
# in this format.
if fnmatch.fnmatchcase(filename, '*_-_*.*'):
# Get the title and author from the file name
title, sep, author = filename.rpartition('_-_')
author, sep, ext = author.rpartition('_')
book_title = title.replace('_', ' ')
book_author = author.replace('_', ' ')
# if the filename did not match just set the title to
# the filename without the extension
else:
book_title = os.path.splitext(filename)[0].replace('_', ' ')
self.append(Book(os.path.join(path, filename), book_title, book_author))
def add_book(self, path, title):
self.append(Book(path, title, ""))
def remove_book(self, path):
for book in self:
if path.endswith(book.path):
self.remove(book)
break
def supports_tags(self):
''' Return True if the the device supports tags (collections) for this book list. '''
return False
def set_tags(self, book, tags):
pass

View File

@ -0,0 +1,325 @@
__license__ = 'GPL v3'
__copyright__ = '2009, John Schember <john at nachtimwald.com'
'''
Device driver for Bookeen's Cybook Gen 3
'''
import os, fnmatch, shutil, time
from itertools import cycle
from calibre.devices.interface import Device
from calibre.devices.errors import DeviceError, FreeSpaceError
from calibre.devices.cybookg3.books import BookList, EBOOK_DIR, EBOOK_TYPES
from calibre import iswindows, islinux, isosx, __appname__
class CYBOOKG3(Device):
# Ordered list of supported formats
FORMATS = EBOOK_TYPES
VENDOR_ID = 0x0bda
PRODUCT_ID = 0x0703
BCD = 0x110
#THUMBNAIL_HEIGHT = 68 # Height for thumbnails on device
MAIN_MEMORY_VOLUME_LABEL = 'Cybook Gen 3 Main Memory'
STORAGE_CARD_VOLUME_LABEL = 'Cybook Gen 3 Storage Card'
FDI_TEMPLATE = \
'''
<device>
<match key="info.category" string="volume">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.vendor_id" int="%(vendor_id)s">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.product_id" int="%(product_id)s">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.device_revision_bcd" int="%(bcd)s">
<match key="volume.is_partition" bool="false">
<merge key="volume.label" type="string">%(main_memory)s</merge>
<merge key="%(app)s.mainvolume" type="string">%(deviceclass)s</merge>
</match>
</match>
</match>
</match>
</match>
</device>
<device>
<match key="info.category" string="volume">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.vendor_id" int="%(vendor_id)s">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.product_id" int="%(product_id)s">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.device_revision_bcd" int="%(bcd)s">
<match key="volume.is_partition" bool="true">
<merge key="volume.label" type="string">%(storage_card)s</merge>
<merge key="%(app)s.cardvolume" type="string">%(deviceclass)s</merge>
</match>
</match>
</match>
</match>
</match>
</device>
'''
def __init__(self, key='-1', log_packets=False, report_progress=None) :
self._main_prefix = self._card_prefix = None
@classmethod
def get_fdi(cls):
return cls.FDI_TEMPLATE%dict(
app=__appname__,
deviceclass=cls.__name__,
vendor_id=hex(cls.VENDOR_ID),
product_id=hex(cls.PRODUCT_ID),
bcd=hex(cls.BCD),
main_memory=cls.MAIN_MEMORY_VOLUME_LABEL,
storage_card=cls.STORAGE_CARD_VOLUME_LABEL,
)
def set_progress_reporter(self, report_progress):
self.report_progress = report_progress
def get_device_information(self, end_session=True):
"""
Ask device for device information. See L{DeviceInfoQuery}.
@return: (device name, device version, software version on device, mime type)
"""
return (self.__class__.__name__, '', '', '')
def card_prefix(self, end_session=True):
return self._card_prefix
@classmethod
def _windows_space(cls, prefix):
if prefix is None:
return 0, 0
win32file = __import__('win32file', globals(), locals(), [], -1)
try:
sectors_per_cluster, bytes_per_sector, free_clusters, total_clusters = \
win32file.GetDiskFreeSpace(prefix[:-1])
except Exception, err:
if getattr(err, 'args', [None])[0] == 21: # Disk not ready
time.sleep(3)
sectors_per_cluster, bytes_per_sector, free_clusters, total_clusters = \
win32file.GetDiskFreeSpace(prefix[:-1])
else: raise
mult = sectors_per_cluster * bytes_per_sector
return total_clusters * mult, free_clusters * mult
def total_space(self, end_session=True):
msz = csz = 0
print self._main_prefix
if not iswindows:
if self._main_prefix is not None:
stats = os.statvfs(self._main_prefix)
msz = stats.f_frsize * (stats.f_blocks + stats.f_bavail - stats.f_bfree)
if self._card_prefix is not None:
stats = os.statvfs(self._card_prefix)
csz = stats.f_frsize * (stats.f_blocks + stats.f_bavail - stats.f_bfree)
else:
msz = self._windows_space(self._main_prefix)[0]
csz = self._windows_space(self._card_prefix)[0]
return (msz, 0, csz)
def free_space(self, end_session=True):
msz = csz = 0
if not iswindows:
if self._main_prefix is not None:
stats = os.statvfs(self._main_prefix)
msz = stats.f_frsize * stats.f_bavail
if self._card_prefix is not None:
stats = os.statvfs(self._card_prefix)
csz = stats.f_frsize * stats.f_bavail
else:
msz = self._windows_space(self._main_prefix)[1]
csz = self._windows_space(self._card_prefix)[1]
return (msz, 0, csz)
def books(self, oncard=False, end_session=True):
if oncard and self._card_prefix is None:
return []
prefix = self._card_prefix if oncard else self._main_prefix
bl = BookList(prefix)
return bl
def upload_books(self, files, names, on_card=False, end_session=True):
if on_card and not self._card_prefix:
raise ValueError(_('The reader has no storage card connected.'))
if not on_card:
path = os.path.join(self._main_prefix, EBOOK_DIR)
else:
path = os.path.join(self._card_prefix, EBOOK_DIR)
sizes = map(os.path.getsize, files)
size = sum(sizes)
if on_card and size > self.free_space()[2] - 1024*1024:
raise FreeSpaceError("There is insufficient free space "+\
"on the storage card")
if not on_card and size > self.free_space()[0] - 2*1024*1024:
raise FreeSpaceError("There is insufficient free space " +\
"in main memory")
paths = []
names = iter(names)
for infile in files:
filepath = os.path.join(path, names.next())
paths.append(filepath)
shutil.copy2(infile, filepath)
return zip(paths, cycle([on_card]))
@classmethod
def add_books_to_metadata(cls, locations, metadata, booklists):
for location in locations:
path = location[0]
on_card = 1 if location[1] else 0
booklists[on_card].add_book(path, os.path.basename(path))
def delete_books(self, paths, end_session=True):
for path in paths:
if os.path.exists(path):
# Delete the ebook
os.unlink(path)
filepath, ext = os.path.splitext(path)
basepath, filename = os.path.split(filepath)
# Delete the ebook auxiliary file
if os.path.exists(filepath + '.mbp'):
os.unlink(filepath + '.mbp')
# Delete the thumbnails file auto generated for the ebook
for p, d, files in os.walk(basepath):
for filen in fnmatch.filter(files, filename + "*.t2b"):
os.unlink(os.path.join(p, filen))
@classmethod
def remove_books_from_metadata(cls, paths, booklists):
for path in paths:
for bl in booklists:
bl.remove_book(path)
def sync_booklists(self, booklists, end_session=True):
# There is no meta data on the device to update. The device is treated
# as a mass storage device and does not use a meta data xml file like
# the Sony Readers.
pass
def get_file(self, path, outfile, end_session=True):
path = self.munge_path(path)
src = open(path, 'rb')
shutil.copyfileobj(src, outfile, 10*1024*1024)
def munge_path(self, path):
if path.startswith('/') and not (path.startswith(self._main_prefix) or \
(self._card_prefix and path.startswith(self._card_prefix))):
path = self._main_prefix + path[1:]
elif path.startswith('card:'):
path = path.replace('card:', self._card_prefix[:-1])
return path
def _windows_match_device(self, device_id):
device_id = device_id.upper()
vid, pid = hex(cls.VENDOR_ID)[2:], hex(cls.PRODUCT_ID)[2:]
while len(vid) < 4: vid = '0' + vid
while len(pid) < 4: pid = '0' + pid
if 'VID_'+vid in device_id and 'PID_'+pid in device_id:
return True
return False
# This only supports Windows >= 2000
def open_windows(self):
drives = []
wmi = __import__('wmi', globals(), locals(), [], -1)
c = wmi.WMI()
for drive in c.Win32_DiskDrive():
if self._windows_match_device(str(drive.PNPDeviceID)):
if drive.Partitions == 0:
continue
try:
partition = drive.associators("Win32_DiskDriveToDiskPartition")[0]
logical_disk = partition.associators('Win32_LogicalDiskToPartition')[0]
prefix = logical_disk.DeviceID+os.sep
drives.append((drive.Index, prefix))
except IndexError:
continue
if not drives:
raise DeviceError(_('Unable to detect the %s disk drive. Try rebooting.')%self.__class__.__name__)
drives.sort(cmp=lambda a, b: cmp(a[0], b[0]))
self._main_prefix = drives[0][1]
if len(drives) > 1:
self._card_prefix = drives[1][1]
def open_osx(self):
raise NotImplementedError()
def open_linux(self):
import dbus
bus = dbus.SystemBus()
hm = dbus.Interface(bus.get_object("org.freedesktop.Hal", "/org/freedesktop/Hal/Manager"), "org.freedesktop.Hal.Manager")
def conditional_mount(dev):
mmo = bus.get_object("org.freedesktop.Hal", dev)
label = mmo.GetPropertyString('volume.label', dbus_interface='org.freedesktop.Hal.Device')
is_mounted = mmo.GetPropertyString('volume.is_mounted', dbus_interface='org.freedesktop.Hal.Device')
mount_point = mmo.GetPropertyString('volume.mount_point', dbus_interface='org.freedesktop.Hal.Device')
fstype = mmo.GetPropertyString('volume.fstype', dbus_interface='org.freedesktop.Hal.Device')
if is_mounted:
return str(mount_point)
mmo.Mount(label, fstype, ['umask=077', 'uid='+str(os.getuid()), 'sync'],
dbus_interface='org.freedesktop.Hal.Device.Volume')
return os.path.normpath('/media/'+label)+'/'
mm = hm.FindDeviceStringMatch(__appname__+'.mainvolume', self.__class__.__name__)
if not mm:
raise DeviceError(_('Unable to detect the %s disk drive. Try rebooting.')%(self.__class__.__name__,))
self._main_prefix = None
for dev in mm:
try:
self._main_prefix = conditional_mount(dev)+os.sep
break
except dbus.exceptions.DBusException:
continue
if not self._main_prefix:
raise DeviceError('Could not open device for reading. Try a reboot.')
self._card_prefix = None
cards = hm.FindDeviceStringMatch(__appname__+'.cardvolume', self.__class__.__name__)
for dev in cards:
try:
self._card_prefix = conditional_mount(dev)+os.sep
break
except:
import traceback
print traceback
continue
def open(self):
time.sleep(5)
self._main_prefix = self._card_prefix = None
if islinux:
try:
self.open_linux()
except DeviceError:
time.sleep(3)
self.open_linux()
if iswindows:
try:
self.open_windows()
except DeviceError:
time.sleep(3)
self.open_windows()
if isosx:
try:
self.open_osx()
except DeviceError:
time.sleep(3)
self.open_osx()

View File

@ -39,6 +39,18 @@ class Device(object):
'''Return the FDI description of this device for HAL on linux.'''
return ''
def open(self):
'''
Perform any device specific initialization. Called after the device is
detected but before any other functions that communicate with the device.
For example: For devices that present themselves as USB Mass storage
devices, this method would be responsible for mounting the device or
if the device has been automounted, for finding out where it has been
mounted. The driver for the PRS505 has a implementation of this function
that should serve as a good example for USB Mass storage devices.
'''
raise NotImplementedError()
def set_progress_reporter(self, report_progress):
'''
@param report_progress: Function that is called with a % progress

View File

@ -4,17 +4,20 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
This module provides a thin ctypes based wrapper around libusb.
"""
from ctypes import cdll, POINTER, byref, pointer, Structure, \
from ctypes import cdll, POINTER, byref, pointer, Structure as _Structure, \
c_ubyte, c_ushort, c_int, c_char, c_void_p, c_byte, c_uint
from errno import EBUSY, ENOMEM
from calibre import iswindows, isosx, load_library, isfrozen
from calibre import iswindows, isosx, load_library
_libusb_name = 'libusb'
PATH_MAX = 511 if iswindows else 1024 if isosx else 4096
if iswindows:
Structure._pack_ = 1
class Structure(_Structure):
_pack_ = 1
_libusb_name = 'libusb0'
else:
Structure = _Structure
try:
try:

View File

@ -60,7 +60,7 @@ class Book(object):
rpath = book_metadata_field("path")
id = book_metadata_field("id", formatter=int)
sourceid = book_metadata_field("sourceid", formatter=int)
size = book_metadata_field("size", formatter=int)
size = book_metadata_field("size", formatter=lambda x : int(float(x)))
# When setting this attribute you must use an epoch
datetime = book_metadata_field("date", formatter=strptime, setter=strftime)

View File

@ -19,10 +19,10 @@ class File(object):
self.is_readonly = not os.access(path, os.W_OK)
self.ctime = stats.st_ctime
self.wtime = stats.st_mtime
self.size = stats.st_size
if path.endswith(os.sep):
self.size = stats.st_size
if path.endswith(os.sep):
path = path[:-1]
self.path = path
self.path = path
self.name = os.path.basename(path)
@ -64,9 +64,11 @@ class PRS505(Device):
<match key="info.category" string="volume">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.vendor_id" int="%(vendor_id)s">
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.product_id" int="%(product_id)s">
<match key="volume.is_partition" bool="true">
<merge key="volume.label" type="string">%(storage_card)s</merge>
<merge key="%(app)s.cardvolume" type="string">%(deviceclass)s</merge>
<match key="@info.parent:@info.parent:@info.parent:@info.parent:usb.device_revision_bcd" int="%(bcd)s">
<match key="volume.is_partition" bool="true">
<merge key="volume.label" type="string">%(storage_card)s</merge>
<merge key="%(app)s.cardvolume" type="string">%(deviceclass)s</merge>
</match>
</match>
</match>
</match>
@ -106,7 +108,10 @@ class PRS505(Device):
@classmethod
def get_osx_mountpoints(cls, raw=None):
if raw is None:
raw = subprocess.Popen('ioreg -w 0 -S -c IOMedia'.split(),
ioreg = '/usr/sbin/ioreg'
if not os.access(ioreg, os.X_OK):
ioreg = 'ioreg'
raw = subprocess.Popen((ioreg+' -w 0 -S -c IOMedia').split(),
stdout=subprocess.PIPE).stdout.read()
lines = raw.splitlines()
names = {}

View File

@ -43,7 +43,8 @@ class DeviceScanner(object):
if iswindows:
for device_id in self.devices:
vid, pid = 'vid_%4.4x'%device.VENDOR_ID, 'pid_%4.4x'%device.PRODUCT_ID
if vid in device_id and pid in device_id:
rev = ('rev_%4.4x'%device.BCD).replace('a', ':') # Bug in winutil.get_usb_devices converts a to :
if vid in device_id and pid in device_id and rev in device_id:
return True
return False
else:

View File

@ -1,4 +1,4 @@
#!/usr/bin/env python
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
__docformat__ = 'restructuredtext en'
@ -6,10 +6,12 @@ __docformat__ = 'restructuredtext en'
'''
Conversion to EPUB.
'''
import sys, textwrap, re
import sys, textwrap, re, os, uuid
from itertools import cycle
from calibre.utils.config import Config, StringConfig
from calibre.utils.zipfile import ZipFile, ZIP_STORED
from calibre.ebooks.html import config as common_config, tostring
from lxml import etree
class DefaultProfile(object):
@ -36,6 +38,38 @@ def rules(stylesheets):
if r.type == r.STYLE_RULE:
yield r
def decrypt_font(key, path):
raw = open(path, 'rb').read()
crypt = raw[:1024]
key = cycle(iter(key))
decrypt = ''.join([chr(ord(x)^key.next()) for x in crypt])
with open(path, 'wb') as f:
f.write(decrypt)
f.write(raw[1024:])
def process_encryption(encfile, opf):
key = None
m = re.search(r'(?i)(urn:uuid:[0-9a-f-]+)', open(opf, 'rb').read())
if m:
key = m.group(1)
key = list(map(ord, uuid.UUID(key).bytes))
try:
root = etree.parse(encfile)
for em in root.xpath('descendant::*[contains(name(), "EncryptionMethod")]'):
algorithm = em.get('Algorithm', '')
if algorithm != 'http://ns.adobe.com/pdf/enc#RC':
return False
cr = em.getparent().xpath('descendant::*[contains(name(), "CipherReference")]')[0]
uri = cr.get('URI')
path = os.path.abspath(os.path.join(os.path.dirname(encfile), '..', *uri.split('/')))
if os.path.exists(path):
decrypt_font(key, path)
return True
except:
import traceback
traceback.print_exc()
return False
def initialize_container(path_to_container, opf_name='metadata.opf'):
'''
Create an empty EPUB document, with a default skeleton.
@ -54,10 +88,10 @@ def initialize_container(path_to_container, opf_name='metadata.opf'):
zf.writestr('META-INF/container.xml', CONTAINER)
return zf
def config(defaults=None):
def config(defaults=None, name='epub'):
desc = _('Options to control the conversion to EPUB')
if defaults is None:
c = Config('epub', desc)
c = Config(name, desc)
else:
c = StringConfig(defaults, desc)

View File

@ -12,12 +12,13 @@ from contextlib import nested
from calibre import extract, walk
from calibre.ebooks import DRMError
from calibre.ebooks.epub import config as common_config
from calibre.ebooks.epub import config as common_config, process_encryption
from calibre.ebooks.epub.from_html import convert as html2epub, find_html_index
from calibre.ptempfile import TemporaryDirectory
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.utils.zipfile import ZipFile
from calibre.customize.ui import run_plugins_on_preprocess
def lit2opf(path, tdir, opts):
from calibre.ebooks.lit.reader import LitReader
@ -30,7 +31,7 @@ def lit2opf(path, tdir, opts):
def mobi2opf(path, tdir, opts):
from calibre.ebooks.mobi.reader import MobiReader
print 'Exploding MOBI file:', path
print 'Exploding MOBI file:', path.encode('utf-8') if isinstance(path, unicode) else path
reader = MobiReader(path)
reader.extract_content(tdir)
files = list(walk(tdir))
@ -72,12 +73,19 @@ def epub2opf(path, tdir, opts):
zf = ZipFile(path)
zf.extractall(tdir)
opts.chapter_mark = 'none'
if os.path.exists(os.path.join(tdir, 'META-INF', 'encryption.xml')):
raise DRMError(os.path.basename(path))
encfile = os.path.join(tdir, 'META-INF', 'encryption.xml')
opf = None
for f in walk(tdir):
if f.lower().endswith('.opf'):
return f
raise ValueError('%s is not a valid EPUB file'%path)
opf = f
break
if opf and os.path.exists(encfile):
if not process_encryption(encfile, opf):
raise DRMError(os.path.basename(path))
if opf is None:
raise ValueError('%s is not a valid EPUB file'%path)
return opf
def odt2epub(path, tdir, opts):
from calibre.ebooks.odt.to_oeb import Extract
@ -109,7 +117,9 @@ def unarchive(path, tdir):
return f, ext
return find_html_index(files)
def any2epub(opts, path, notification=None):
def any2epub(opts, path, notification=None, create_epub=True,
oeb_cover=False, extract_to=None):
path = run_plugins_on_preprocess(path)
ext = os.path.splitext(path)[1]
if not ext:
raise ValueError('Unknown file type: '+path)
@ -132,7 +142,9 @@ def any2epub(opts, path, notification=None):
raise ValueError('Conversion from %s is not supported'%ext.upper())
print 'Creating EPUB file...'
html2epub(path, opts, notification=notification)
html2epub(path, opts, notification=notification,
create_epub=create_epub, oeb_cover=oeb_cover,
extract_to=extract_to)
def config(defaults=None):
return common_config(defaults=defaults)
@ -141,14 +153,14 @@ def config(defaults=None):
def formats():
return ['html', 'rar', 'zip', 'oebzip']+list(MAP.keys())
def option_parser():
return config().option_parser(usage=_('''\
USAGE = _('''\
%%prog [options] filename
Convert any of a large number of ebook formats to an epub file. Supported formats are: %s
''')%formats()
)
Convert any of a large number of ebook formats to a %s file. Supported formats are: %s
''')
def option_parser(usage=USAGE):
return config().option_parser(usage=usage%('EPUB', formats()))
def main(args=sys.argv):
parser = option_parser()

View File

@ -32,14 +32,15 @@ Conversion of HTML/OPF files follows several stages:
* The EPUB container is created.
'''
import os, sys, cStringIO, logging, re
import os, sys, cStringIO, logging, re, functools, shutil
from lxml.etree import XPath
from lxml import html, etree
from PyQt4.Qt import QApplication, QPixmap
from calibre.ebooks.html import Processor, merge_metadata, get_filelist,\
opf_traverse, create_metadata, rebase_toc
from calibre.ebooks.epub import config as common_config
opf_traverse, create_metadata, rebase_toc, Link, parser
from calibre.ebooks.epub import config as common_config, tostring
from calibre.ptempfile import TemporaryDirectory
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata.opf2 import OPF
@ -47,7 +48,46 @@ from calibre.ebooks.epub import initialize_container, PROFILES
from calibre.ebooks.epub.split import split
from calibre.ebooks.epub.fonts import Rationalizer
from calibre.constants import preferred_encoding
from calibre import walk
from calibre.customize.ui import run_plugins_on_postprocess
from calibre import walk, CurrentDir, to_unicode
content = functools.partial(os.path.join, u'content')
def remove_bad_link(element, attribute, link, pos):
if attribute is not None:
if element.tag in ['link']:
element.getparent().remove(element)
else:
element.set(attribute, '')
del element.attrib[attribute]
def check_links(opf_path, pretty_print):
'''
Find and remove all invalid links in the HTML files
'''
logger = logging.getLogger('html2epub')
logger.info('\tChecking files for bad links...')
pathtoopf = os.path.abspath(opf_path)
with CurrentDir(os.path.dirname(pathtoopf)):
opf = OPF(open(pathtoopf, 'rb'), os.path.dirname(pathtoopf))
html_files = []
for item in opf.itermanifest():
if 'html' in item.get('media-type', '').lower():
f = item.get('href').split('/')[-1].decode('utf-8')
html_files.append(os.path.abspath(content(f)))
for path in html_files:
base = os.path.dirname(path)
root = html.fromstring(open(content(path), 'rb').read(), parser=parser)
for element, attribute, link, pos in list(root.iterlinks()):
link = to_unicode(link)
plink = Link(link, base)
bad = False
if plink.path is not None and not os.path.exists(plink.path):
bad = True
if bad:
remove_bad_link(element, attribute, link, pos)
open(content(path), 'wb').write(tostring(root, pretty_print))
def find_html_index(files):
'''
@ -83,6 +123,10 @@ class HTMLProcessor(Processor, Rationalizer):
if opts.verbose > 2:
self.debug_tree('nocss')
if hasattr(self.body, 'xpath'):
for script in list(self.body.xpath('descendant::script')):
script.getparent().remove(script)
def convert_image(self, img):
rpath = img.get('src', '')
path = os.path.join(os.path.dirname(self.save_path()), *rpath.split('/'))
@ -165,23 +209,22 @@ TITLEPAGE = '''\
</head>
<body>
<div>
<img src="%s" alt="cover" />
<img src="%s" alt="cover" style="height: 100%%" />
</div>
</body>
</html>
'''
def create_cover_image(src, dest, screen_size):
from PyQt4.Qt import QApplication, QImage, Qt
if QApplication.instance() is None:
app = QApplication([])
app
im = QImage()
def create_cover_image(src, dest, screen_size, rescale_cover=True):
try:
from PyQt4.Qt import QImage, Qt
if QApplication.instance() is None:
QApplication([])
im = QImage()
im.load(src)
if im.isNull():
raise ValueError
if screen_size is not None:
raise ValueError('Invalid cover image')
if rescale_cover and screen_size is not None:
width, height = im.width(), im.height()
dw, dh = (screen_size[0]-width)/float(width), (screen_size[1]-height)/float(height)
delta = min(dw, dh)
@ -189,7 +232,7 @@ def create_cover_image(src, dest, screen_size):
nwidth = int(width + delta*(width))
nheight = int(height + delta*(height))
im = im.scaled(int(nwidth), int(nheight), Qt.IgnoreAspectRatio, Qt.SmoothTransformation)
im.save(dest)
im.save(dest)
except:
import traceback
traceback.print_exc()
@ -202,7 +245,6 @@ def process_title_page(mi, filelist, htmlfilemap, opts, tdir):
if mi.cover:
if f(filelist[0].path) == f(mi.cover):
old_title_page = htmlfilemap[filelist[0].path]
#logger = logging.getLogger('html2epub')
metadata_cover = mi.cover
if metadata_cover and not os.path.exists(metadata_cover):
@ -211,14 +253,15 @@ def process_title_page(mi, filelist, htmlfilemap, opts, tdir):
cpath = '/'.join(('resources', '_cover_.jpg'))
cover_dest = os.path.join(tdir, 'content', *cpath.split('/'))
if metadata_cover is not None:
if not create_cover_image(metadata_cover, cover_dest, opts.profile.screen_size):
if not create_cover_image(metadata_cover, cover_dest,
opts.profile.screen_size):
metadata_cover = None
specified_cover = opts.cover
if specified_cover and not os.path.exists(specified_cover):
specified_cover = None
if specified_cover is not None:
if not create_cover_image(specified_cover, cover_dest, opts.profile.screen_size):
if not create_cover_image(specified_cover, cover_dest,
opts.profile.screen_size):
specified_cover = None
cover = metadata_cover if specified_cover is None or (opts.prefer_metadata_cover and metadata_cover is not None) else specified_cover
@ -233,9 +276,26 @@ def process_title_page(mi, filelist, htmlfilemap, opts, tdir):
elif os.path.exists(cover_dest):
os.remove(cover_dest)
return None, old_title_page is not None
def convert(htmlfile, opts, notification=None):
def find_oeb_cover(htmlfile):
if os.stat(htmlfile).st_size > 2048:
return None
match = re.search(r'(?i)<img[^<>]+src\s*=\s*[\'"](.+?)[\'"]', open(htmlfile, 'rb').read())
if match:
return match.group(1)
def condense_ncx(ncx_path):
tree = etree.parse(ncx_path)
for tag in tree.getroot().iter(tag=etree.Element):
if tag.text:
tag.text = tag.text.strip()
if tag.tail:
tag.tail = tag.tail.strip()
compressed = etree.tostring(tree.getroot(), encoding='utf-8')
open(ncx_path, 'wb').write(compressed)
def convert(htmlfile, opts, notification=None, create_epub=True,
oeb_cover=False, extract_to=None):
htmlfile = os.path.abspath(htmlfile)
if opts.output is None:
opts.output = os.path.splitext(os.path.basename(htmlfile))[0] + '.epub'
@ -287,7 +347,7 @@ def convert(htmlfile, opts, notification=None):
title_page, has_title_page = process_title_page(mi, filelist, htmlfile_map, opts, tdir)
spine = [htmlfile_map[f.path] for f in filelist]
if title_page is not None:
if not oeb_cover and title_page is not None:
spine = [title_page] + spine
mi.cover = None
mi.cover_data = (None, None)
@ -316,26 +376,55 @@ def convert(htmlfile, opts, notification=None):
if opts.show_ncx:
print toc
split(opf_path, opts, stylesheet_map)
check_links(opf_path, opts.pretty_print)
opf = OPF(opf_path, tdir)
opf.remove_guide()
if has_title_page:
oeb_cover_file = None
if oeb_cover and title_page is not None:
oeb_cover_file = find_oeb_cover(os.path.join(tdir, 'content', title_page))
if has_title_page or (oeb_cover and oeb_cover_file):
opf.create_guide_element()
opf.add_guide_item('cover', 'Cover', 'content/'+spine[0])
if has_title_page and not oeb_cover:
opf.add_guide_item('cover', 'Cover', 'content/'+spine[0])
if oeb_cover and oeb_cover_file:
opf.add_guide_item('cover', 'Cover', 'content/'+oeb_cover_file)
opf.add_path_to_manifest(os.path.join(tdir, 'content', 'resources', '_cover_.jpg'), 'image/jpeg')
cpath = os.path.join(tdir, 'content', 'resources', '_cover_.jpg')
if os.path.exists(cpath):
opf.add_path_to_manifest(cpath, 'image/jpeg')
with open(opf_path, 'wb') as f:
raw = opf.render()
if not raw.startswith('<?xml '):
raw = '<?xml version="1.0" encoding="UTF-8"?>\n'+raw
f.write(raw)
epub = initialize_container(opts.output)
epub.add_dir(tdir)
ncx_path = os.path.join(os.path.dirname(opf_path), 'toc.ncx')
if os.path.exists(ncx_path) and os.stat(ncx_path).st_size > opts.profile.flow_size:
logger.info('Condensing NCX from %d bytes...'%os.stat(ncx_path).st_size)
condense_ncx(ncx_path)
if os.stat(ncx_path).st_size > opts.profile.flow_size:
logger.warn('NCX still larger than allowed size at %d bytes. Menu based Table of Contents may not work on device.'%os.stat(ncx_path).st_size)
if create_epub:
epub = initialize_container(opts.output)
epub.add_dir(tdir)
epub.close()
run_plugins_on_postprocess(opts.output, 'epub')
logger.info(_('Output written to ')+opts.output)
if opts.show_opf:
print open(os.path.join(tdir, 'metadata.opf')).read()
logger.info('Output written to %s'%opts.output)
if opts.extract_to is not None:
epub.extractall(opts.extract_to)
epub.close()
if os.path.exists(opts.extract_to):
shutil.rmtree(opts.extract_to)
shutil.copytree(tdir, opts.extract_to)
if extract_to is not None:
if os.path.exists(extract_to):
shutil.rmtree(extract_to)
shutil.copytree(tdir, extract_to)
def main(args=sys.argv):

View File

@ -37,14 +37,16 @@ class UnsupportedFormatError(Exception):
class SpineItem(unicode):
def __init__(self, path):
unicode.__init__(self, path)
def __new__(cls, *args):
obj = super(SpineItem, cls).__new__(cls, *args)
path = args[0]
raw = open(path, 'rb').read()
raw, self.encoding = xml_to_unicode(raw)
self.character_count = character_count(raw)
self.start_page = -1
self.pages = -1
self.max_page = -1
raw, obj.encoding = xml_to_unicode(raw)
obj.character_count = character_count(raw)
obj.start_page = -1
obj.pages = -1
obj.max_page = -1
return obj
def html2opf(path, tdir, opts):
opts = copy.copy(opts)
@ -166,8 +168,12 @@ class EbookIterator(object):
if bookmarks is None:
bookmarks = self.bookmarks
dat = self.serialize_bookmarks(bookmarks)
if os.path.splitext(self.pathtoebook)[1].lower() == '.epub':
zf = open(self.pathtoebook, 'r+b')
if os.path.splitext(self.pathtoebook)[1].lower() == '.epub' and \
os.access(self.pathtoebook, os.R_OK):
try:
zf = open(self.pathtoebook, 'r+b')
except IOError:
return
zipf = ZipFile(zf, mode='a')
for name in zipf.namelist():
if name == 'META-INF/calibre_bookmarks.txt':

View File

@ -7,7 +7,7 @@ __docformat__ = 'restructuredtext en'
Split the flows in an epub file to conform to size limitations.
'''
import os, math, logging, functools, collections, re, copy
import os, math, logging, functools, collections, re, copy, sys
from lxml.etree import XPath as _XPath
from lxml import etree, html
@ -34,29 +34,61 @@ class SplitError(ValueError):
class Splitter(LoggingInterface):
def __init__(self, path, opts, stylesheet_map, always_remove=False):
def __init__(self, path, opts, stylesheet_map, opf):
LoggingInterface.__init__(self, logging.getLogger('htmlsplit'))
self.setup_cli_handler(opts.verbose)
self.path = path
self.always_remove = always_remove
self.always_remove = not opts.preserve_tag_structure or \
os.stat(content(path)).st_size > 5*opts.profile.flow_size
self.base = (os.path.splitext(path)[0].replace('%', '%%') + '_split_%d.html')
self.opts = opts
self.orig_size = os.stat(content(path)).st_size
self.log_info('\tSplitting %s (%d KB)', path, self.orig_size/1024.)
root = html.fromstring(open(content(path)).read())
self.page_breaks = []
self.find_page_breaks(stylesheet_map[self.path], root)
self.trees = []
self.page_breaks, self.trees = [], []
self.split_size = 0
self.split(root.getroottree())
self.commit()
self.log_info('\t\tSplit into %d parts.', len(self.trees))
if self.opts.verbose:
for f in self.files:
self.log_info('\t\t\t%s - %d KB', f, os.stat(content(f)).st_size/1024.)
# Split on page breaks
self.log_info('\tSplitting on page breaks...')
if self.path in stylesheet_map:
self.find_page_breaks(stylesheet_map[self.path], root)
self.split_on_page_breaks(root.getroottree())
trees = list(self.trees)
# Split any remaining over-sized trees
if self.opts.profile.flow_size < sys.maxint:
lt_found = False
self.log_info('\tLooking for large trees...')
for i, tree in enumerate(list(trees)):
self.trees = []
size = len(tostring(tree.getroot()))
if size > self.opts.profile.flow_size:
lt_found = True
try:
self.split_to_size(tree)
except (SplitError, RuntimeError): # Splitting fails
if not self.always_remove:
self.always_remove = True
self.split_to_size(tree)
else:
raise
trees[i:i+1] = list(self.trees)
if not lt_found:
self.log_info('\tNo large trees found')
self.trees = trees
self.was_split = len(self.trees) > 1
if self.was_split:
self.commit()
self.log_info('\t\tSplit into %d parts.', len(self.trees))
if self.opts.verbose:
for f in self.files:
self.log_info('\t\t\t%s - %d KB', f, os.stat(content(f)).st_size/1024.)
self.fix_opf(opf)
self.trees = None
def split_text(self, text, root, size):
self.log_debug('\t\t\tSplitting text of length: %d'%len(text))
@ -76,12 +108,7 @@ class Splitter(LoggingInterface):
return ans
def split(self, tree):
'''
Split ``tree`` into a *before* and *after* tree, preserving tag structure,
but not duplicating any text. All tags that have had their text and tail
removed have the attribute ``calibre_split`` set to 1.
'''
def split_to_size(self, tree):
self.log_debug('\t\tSplitting...')
root = tree.getroot()
# Split large <pre> tags
@ -108,10 +135,50 @@ class Splitter(LoggingInterface):
if not self.always_remove:
self.log_warn(_('\t\tToo much markup. Re-splitting without structure preservation. This may cause incorrect rendering.'))
raise SplitError(self.path, root)
tree2 = copy.deepcopy(tree)
root2 = tree2.getroot()
body, body2 = root.body, root2.body
path = tree.getpath(split_point)
for t in self.do_split(tree, split_point, before):
r = t.getroot()
if self.is_page_empty(r):
continue
size = len(tostring(r))
if size <= self.opts.profile.flow_size:
self.trees.append(t)
#print tostring(t.getroot(), pretty_print=True)
self.log_debug('\t\t\tCommitted sub-tree #%d (%d KB)', len(self.trees), size/1024.)
self.split_size += size
else:
self.split_to_size(t)
def is_page_empty(self, root):
body = root.find('body')
if body is None:
return False
txt = re.sub(r'\s+', '', html.tostring(body, method='text', encoding=unicode))
if len(txt) > 4:
#if len(txt) < 100:
# print 1111111, html.tostring(body, method='html', encoding=unicode)
return False
for img in root.xpath('//img'):
if img.get('style', '') != 'display:none':
return False
return True
def do_split(self, tree, split_point, before):
'''
Split ``tree`` into a *before* and *after* tree at ``split_point``,
preserving tag structure, but not duplicating any text.
All tags that have had their text and tail
removed have the attribute ``calibre_split`` set to 1.
:param before: If True tree is split before split_point, otherwise after split_point
:return: before_tree, after_tree
'''
path = tree.getpath(split_point)
tree, tree2 = copy.deepcopy(tree), copy.deepcopy(tree)
root = tree.getroot()
root2 = tree2.getroot()
body, body2 = root.body, root2.body
split_point = root.xpath(path)[0]
split_point2 = root2.xpath(path)[0]
def nix_element(elem, top=True):
@ -129,7 +196,7 @@ class Splitter(LoggingInterface):
elem.tail = u''
elem.set(SPLIT_ATTR, '1')
if elem.tag.lower() in ['ul', 'ol', 'dl', 'table', 'hr', 'img']:
elem.set('style', 'display:none;')
elem.set('style', 'display:none')
def fix_split_point(sp):
sp.set('style', sp.get('style', '')+'page-break-before:avoid;page-break-after:avoid')
@ -163,20 +230,35 @@ class Splitter(LoggingInterface):
if not hit_split_point:
nix_element(elem, top=False)
for t, r in [(tree, root), (tree2, root2)]:
size = len(tostring(r))
if size <= self.opts.profile.flow_size:
self.trees.append(t)
#print tostring(t.getroot(), pretty_print=True)
self.log_debug('\t\t\tCommitted sub-tree #%d (%d KB)', len(self.trees), size/1024.)
self.split_size += size
else:
self.split(t)
return tree, tree2
def split_on_page_breaks(self, orig_tree):
ordered_ids = []
for elem in orig_tree.xpath('//*[@id]'):
id = elem.get('id')
if id in self.page_break_ids:
ordered_ids.append(self.page_breaks[self.page_break_ids.index(id)])
self.trees = []
tree = orig_tree
for pattern, before in ordered_ids:
self.log_info('\t\tSplitting on page-break')
elem = pattern(tree)
if elem:
before, after = self.do_split(tree, elem[0], before)
self.trees.append(before)
tree = after
self.trees.append(tree)
self.trees = [t for t in self.trees if not self.is_page_empty(t.getroot())]
def find_page_breaks(self, stylesheets, root):
'''
Find all elements that have either page-break-before or page-break-after set.
Populates `self.page_breaks` with id based XPath selectors (for elements that don't
have ids, an id is created).
'''
page_break_selectors = set([])
for rule in rules(stylesheets):
@ -204,16 +286,18 @@ class Splitter(LoggingInterface):
page_breaks = list(page_breaks)
page_breaks.sort(cmp=lambda x,y : cmp(x.pb_order, y.pb_order))
self.page_break_ids = []
for i, x in enumerate(page_breaks):
x.set('id', x.get('id', 'calibre_pb_%d'%i))
self.page_breaks.append((XPath('//*[@id="%s"]'%x.get('id')), x.pb_before))
id = x.get('id')
self.page_breaks.append((XPath('//*[@id="%s"]'%id), x.pb_before))
self.page_break_ids.append(id)
def find_split_point(self, root):
'''
Find the tag at which to split the tree rooted at `root`.
Search order is:
* page breaks
* Heading tags
* <div> tags
* <p> tags
@ -229,19 +313,6 @@ class Splitter(LoggingInterface):
elems[i].set(SPLIT_POINT_ATTR, '1')
return elems[i]
page_breaks = []
for x in self.page_breaks:
pb = x[0](root)
if pb:
page_breaks.append(pb[0])
elem = pick_elem(page_breaks)
if elem is not None:
i = page_breaks.index(elem)
return elem, self.page_breaks[i][1]
for path in (
'//*[re:match(name(), "h[1-6]", "i")]',
'/html/body/div',
@ -315,6 +386,9 @@ class Splitter(LoggingInterface):
frag = None
if len(href) > 1:
frag = href[1]
if frag not in self.anchor_map:
self.log_warning('\t\tUnable to re-map OPF link', href)
continue
new_file = self.anchor_map[frag]
ref.set('href', 'content/'+new_file+('' if frag is None else ('#'+frag)))
@ -341,7 +415,11 @@ def fix_content_links(html_files, changes, opts):
anchor = href[1] if len(href) > 1 else None
href = href[0]
if href in split_files:
newf = anchor_maps[split_files.index(href)][anchor]
try:
newf = anchor_maps[split_files.index(href)][anchor]
except:
print '\t\tUnable to remap HTML link:', href, anchor
continue
frag = ('#'+anchor) if anchor else ''
a.set('href', newf+frag)
changed = True
@ -354,52 +432,54 @@ def fix_ncx(path, changes):
anchor_maps = [f.anchor_map for f in changes]
tree = etree.parse(path)
changed = False
for content in tree.getroot().xpath('//x:content[@src]', namespaces={'x':"http://www.daisy.org/z3986/2005/ncx/"}):
for content in tree.getroot().xpath('//x:content[@src]',
namespaces={'x':"http://www.daisy.org/z3986/2005/ncx/"}):
href = content.get('src')
if not href.startswith('#'):
href = href.split('#')
anchor = href[1] if len(href) > 1 else None
href = href[0].split('/')[-1]
if href in split_files:
newf = anchor_maps[split_files.index(href)][anchor]
try:
newf = anchor_maps[split_files.index(href)][anchor]
except:
print 'Unable to remap NCX link:', href, anchor
frag = ('#'+anchor) if anchor else ''
content.set('src', 'content/'+newf+frag)
changed = True
if changed:
open(path, 'wb').write(etree.tostring(tree.getroot(), encoding='UTF-8', xml_declaration=True))
def find_html_files(opf):
'''
Find all HTML files referenced by `opf`.
'''
html_files = []
for item in opf.itermanifest():
if 'html' in item.get('media-type', '').lower():
f = item.get('href').split('/')[-1]
f2 = f.replace('&', '%26')
if not os.path.exists(content(f)) and os.path.exists(content(f2)):
f = f2
item.set('href', item.get('href').replace('&', '%26'))
if os.path.exists(content(f)):
html_files.append(f)
return html_files
def split(pathtoopf, opts, stylesheet_map):
pathtoopf = os.path.abspath(pathtoopf)
opf = OPF(open(pathtoopf, 'rb'), os.path.dirname(pathtoopf))
with CurrentDir(os.path.dirname(pathtoopf)):
opf = OPF(open(pathtoopf, 'rb'), os.path.dirname(pathtoopf))
html_files = []
for item in opf.itermanifest():
if 'html' in item.get('media-type', '').lower():
f = item.get('href').split('/')[-1]
f2 = f.replace('&', '%26')
if not os.path.exists(content(f)) and os.path.exists(content(f2)):
f = f2
item.set('href', item.get('href').replace('&', '%26'))
html_files.append(f)
changes = []
always_remove = not opts.preserve_tag_structure
for f in html_files:
if os.stat(content(f)).st_size > opts.profile.flow_size:
try:
changes.append(Splitter(f, opts, stylesheet_map,
always_remove=(always_remove or \
os.stat(content(f)).st_size > 5*opts.profile.flow_size)))
except (SplitError, RuntimeError):
if not always_remove:
changes.append(Splitter(f, opts, stylesheet_map, always_remove=True))
else:
raise
changes[-1].fix_opf(opf)
html_files = find_html_files(opf)
changes = [Splitter(f, opts, stylesheet_map, opf) for f in html_files]
changes = [c for c in changes if c.was_split]
open(pathtoopf, 'wb').write(opf.render())
fix_content_links(html_files, changes, opts)
for item in opf.itermanifest():
if item.get('media-type', '') == 'application/x-dtbncx+xml':
fix_ncx(item.get('href'), changes)
break
open(pathtoopf, 'wb').write(opf.render())

View File

@ -314,10 +314,22 @@ def opf_traverse(opf_reader, verbose=0, encoding=None):
convert_entities = functools.partial(entity_to_unicode, exceptions=['quot', 'apos', 'lt', 'gt', 'amp'])
_span_pat = re.compile('<span.*?</span>', re.DOTALL|re.IGNORECASE)
def sanitize_head(match):
x = match.group(1)
x = _span_pat.sub('', x)
return '<head>\n'+x+'\n</head>'
class PreProcessor(object):
PREPROCESS = [
# Some idiotic HTML generators (Frontpage I'm looking at you)
# Put all sorts of crap into <head>. This messes up lxml
(re.compile(r'<head[^>]*>(.*?)</head>', re.IGNORECASE|re.DOTALL),
sanitize_head),
# Convert all entities, since lxml doesn't handle them well
(re.compile(r'&(\S+?);'), convert_entities),
]
# Fix pdftohtml markup
@ -446,7 +458,10 @@ class Parser(PreProcessor, LoggingInterface):
def parse_html(self):
''' Create lxml ElementTree from HTML '''
self.log_info('\tParsing '+os.sep.join(self.htmlfile.path.split(os.sep)[-3:]))
if self.htmlfile.is_binary:
raise ValueError('Not a valid HTML file: '+self.htmlfile.path)
src = open(self.htmlfile.path, 'rb').read().decode(self.htmlfile.encoding, 'replace').strip()
src = src.replace('\x00', '')
src = self.preprocess(src)
# lxml chokes on unicode input when it contains encoding declarations
for pat in ENCODING_PATS:
@ -527,6 +542,7 @@ class Processor(Parser):
LINKS_PATH = XPath('//a[@href]')
PIXEL_PAT = re.compile(r'([-]?\d+|[-]?\d*\.\d+)px')
PAGE_PAT = re.compile(r'@page[^{]*?{[^}]*?}')
def __init__(self, *args, **kwargs):
Parser.__init__(self, *args, **kwargs)
@ -696,7 +712,9 @@ class Processor(Parser):
return ''
return '%fpt'%(72 * val/dpi)
return cls.PIXEL_PAT.sub(rescale, css)
css = cls.PIXEL_PAT.sub(rescale, css)
css = cls.PAGE_PAT.sub('', css)
return css
def extract_css(self, parsed_sheets):
'''
@ -732,7 +750,12 @@ class Processor(Parser):
self.log_error('Failed to open stylesheet: %s'%file)
else:
try:
parsed_sheets[file] = self.css_parser.parseString(css)
try:
parsed_sheets[file] = self.css_parser.parseString(css)
except ValueError:
parsed_sheets[file] = \
self.css_parser.parseString(\
css.decode('utf8', 'replace'))
except:
parsed_sheets[file] = css.decode('utf8', 'replace')
self.log_warning('Failed to parse stylesheet: %s'%file)
@ -762,10 +785,15 @@ class Processor(Parser):
class_counter = 0
for font in self.root.xpath('//font'):
try:
size = int(font.attrib.pop('size', '3'))
size = font.attrib.pop('size', '3')
except:
size = 3
setting = 'font-size: %d%%;'%int((float(size)/3) * 100)
size = '3'
if size and size.strip() and size.strip()[0] in ('+', '-'):
size = 3 + float(size) # Hack assumes basefont=3
try:
setting = 'font-size: %d%%;'%int((float(size)/3) * 100)
except ValueError:
setting = ''
face = font.attrib.pop('face', None)
if face is not None:
setting += 'font-face:%s;'%face
@ -798,16 +826,17 @@ class Processor(Parser):
css = '\n'.join(['.%s {%s;}'%(cn, setting) for \
setting, cn in cache.items()])
sheet = self.css_parser.parseString(self.preprocess_css(css))
sheet = self.css_parser.parseString(self.preprocess_css(css.replace(';;}', ';}')))
for rule in sheet:
self.stylesheet.add(rule)
css = ''
css += '\n\n' + 'body {margin-top: 0pt; margin-bottom: 0pt; margin-left: 0pt; margin-right: 0pt;}'
css += '\n\n@page {margin-top: %fpt; margin-bottom: %fpt; margin-left: %fpt; margin-right: %fpt}'%(self.opts.margin_top, self.opts.margin_bottom, self.opts.margin_left, self.opts.margin_right)
css += '\n\n@page {margin-top: %fpt; margin-bottom: %fpt; }'%(self.opts.margin_top, self.opts.margin_bottom)
css += '\n\nbody {margin-left: %fpt; margin-right: %fpt}'%(self.opts.margin_left, self.opts.margin_right)
# Workaround for anchor rendering bug in ADE
css += '\n\na { color: inherit; text-decoration: inherit; cursor: default; }\na[href] { color: blue; text-decoration: underline; cursor:pointer; }'
if self.opts.remove_paragraph_spacing:
css += '\n\np {text-indent: 2em; margin-top:1pt; margin-bottom:1pt; padding:0pt; border:0pt;}'
css += '\n\np {text-indent: 2em; margin-top:0pt; margin-bottom:0pt; padding:0pt; border:0pt;}'
if self.opts.override_css:
css += '\n\n' + self.opts.override_css
self.override_css = self.css_parser.parseString(self.preprocess_css(css))
@ -863,7 +892,7 @@ def option_parser():
%prog [options] file.html|opf
Follow all links in an HTML file and collect them into the specified directory.
Also collects any references resources like images, stylesheets, scripts, etc.
Also collects any resources like images, stylesheets, scripts, etc.
If an OPF file is specified instead, the list of files in its <spine> element
is used.
'''))
@ -1042,11 +1071,12 @@ def main(args=sys.argv):
return 0
def gui_main(htmlfile):
def gui_main(htmlfile, pt=None):
'''
Convenience wrapper for use in recursively importing HTML files.
'''
pt = PersistentTemporaryFile('_html2oeb_gui.oeb.zip')
if pt is None:
pt = PersistentTemporaryFile('_html2oeb_gui.oeb.zip')
pt.close()
opts = '''
pretty_print = True

View File

@ -0,0 +1,59 @@
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
__docformat__ = 'restructuredtext en'
'''
Convert any ebook format to LIT.
'''
import sys, os, glob, logging
from calibre.ebooks.epub.from_any import any2epub, formats, USAGE
from calibre.ebooks.epub import config as common_config
from calibre.ptempfile import TemporaryDirectory
from calibre.ebooks.lit.writer import oeb2lit
def config(defaults=None):
c = common_config(defaults=defaults, name='lit')
return c
def option_parser(usage=USAGE):
return config().option_parser(usage=usage%('LIT', formats()))
def any2lit(opts, path):
ext = os.path.splitext(path)[1]
if not ext:
raise ValueError('Unknown file type: '+path)
ext = ext.lower()[1:]
if opts.output is None:
opts.output = os.path.splitext(os.path.basename(path))[0]+'.lit'
opts.output = os.path.abspath(opts.output)
orig_output = opts.output
with TemporaryDirectory('_any2lit') as tdir:
oebdir = os.path.join(tdir, 'oeb')
os.mkdir(oebdir)
opts.output = os.path.join(tdir, 'dummy.epub')
opts.profile = 'None'
any2epub(opts, path, create_epub=False, oeb_cover=True, extract_to=oebdir)
opf = glob.glob(os.path.join(oebdir, '*.opf'))[0]
opts.output = orig_output
logging.getLogger('html2epub').info(_('Creating LIT file from EPUB...'))
oeb2lit(opts, opf)
def main(args=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(args)
if len(args) < 2:
parser.print_help()
print 'No input file specified.'
return 1
any2lit(opts, args[1])
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@ -0,0 +1,426 @@
/* ***** BEGIN LICENSE BLOCK *****
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
*
* The contents of this file are subject to the Mozilla Public License Version
* 1.1 (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
* http://www.mozilla.org/MPL/
*
* Software distributed under the License is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
* for the specific language governing rights and limitations under the
* License.
*
* The Original Code is mozilla.org code.
*
* The Initial Developer of the Original Code is
* Netscape Communications Corporation.
* Portions created by the Initial Developer are Copyright (C) 1998
* the Initial Developer. All Rights Reserved.
*
* Contributor(s):
* Blake Ross <BlakeR1234@aol.com>
*
* Alternatively, the contents of this file may be used under the terms of
* either of the GNU General Public License Version 2 or later (the "GPL"),
* or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
* in which case the provisions of the GPL or the LGPL are applicable instead
* of those above. If you wish to allow use of your version of this file only
* under the terms of either the GPL or the LGPL, and not to allow others to
* use your version of this file under the terms of the MPL, indicate your
* decision by deleting the provisions above and replace them with the notice
* and other provisions required by the GPL or the LGPL. If you do not delete
* the provisions above, a recipient may use your version of this file under
* the terms of any one of the MPL, the GPL or the LGPL.
*
* ***** END LICENSE BLOCK ***** */
@namespace url(http://www.w3.org/1999/xhtml); /* set default namespace to HTML */
/* blocks */
html, div, map, dt, isindex, form {
display: block;
}
body {
display: block;
margin: 8px;
}
p, dl, multicol {
display: block;
margin: 1em 0;
}
dd {
display: block;
}
blockquote {
display: block;
margin: 1em 40px;
}
address {
display: block;
font-style: italic;
}
center {
display: block;
text-align: center;
}
blockquote[type=cite] {
display: block;
margin: 1em 0px;
border-color: blue;
border-width: thin;
}
span[_moz_quote=true] {
color: blue;
}
pre[_moz_quote=true] {
color: blue;
}
h1 {
display: block;
font-size: 2em;
font-weight: bold;
margin: .67em 0;
}
h2 {
display: block;
font-size: 1.5em;
font-weight: bold;
margin: .83em 0;
}
h3 {
display: block;
font-size: 1.17em;
font-weight: bold;
margin: 1em 0;
}
h4 {
display: block;
font-weight: bold;
margin: 1.33em 0;
}
h5 {
display: block;
font-size: 0.83em;
font-weight: bold;
margin: 1.67em 0;
}
h6 {
display: block;
font-size: 0.67em;
font-weight: bold;
margin: 2.33em 0;
}
listing {
display: block;
font-family: monospace;
font-size: medium;
white-space: pre;
margin: 1em 0;
}
xmp, pre, plaintext {
display: block;
font-family: monospace;
white-space: pre;
margin: 1em 0;
}
/* tables */
table {
display: table;
border-spacing: 2px;
border-collapse: separate;
margin-top: 0;
margin-bottom: 0;
text-indent: 0;
}
table[align="left"] {
float: left;
}
table[align="right"] {
float: right;
}
table[rules]:not([rules="none"]) {
border-collapse: collapse;
}
/* caption inherits from table not table-outer */
caption {
display: table-caption;
text-align: center;
}
table[align="center"] > caption {
margin-left: auto;
margin-right: auto;
}
table[align="center"] > caption[align="left"] {
margin-right: 0;
}
table[align="center"] > caption[align="right"] {
margin-left: 0;
}
tr {
display: table-row;
vertical-align: inherit;
}
col {
display: table-column;
}
colgroup {
display: table-column-group;
}
tbody {
display: table-row-group;
vertical-align: middle;
}
thead {
display: table-header-group;
vertical-align: middle;
}
tfoot {
display: table-footer-group;
vertical-align: middle;
}
/* for XHTML tables without tbody */
table > tr {
vertical-align: middle;
}
td {
display: table-cell;
vertical-align: inherit;
text-align: inherit;
padding: 1px;
}
th {
display: table-cell;
vertical-align: inherit;
font-weight: bold;
padding: 1px;
}
/* inlines */
q:before {
content: open-quote;
}
q:after {
content: close-quote;
}
b, strong {
font-weight: bolder;
}
i, cite, em, var, dfn {
font-style: italic;
}
tt, code, kbd, samp {
font-family: monospace;
}
u, ins {
text-decoration: underline;
}
s, strike, del {
text-decoration: line-through;
}
blink {
text-decoration: blink;
}
big {
font-size: larger;
}
small {
font-size: smaller;
}
sub {
vertical-align: sub;
font-size: smaller;
line-height: normal;
}
sup {
vertical-align: super;
font-size: smaller;
line-height: normal;
}
nobr {
white-space: nowrap;
}
/* titles */
abbr[title], acronym[title] {
border-bottom: dotted 1px;
}
/* lists */
ul, menu, dir {
display: block;
list-style-type: disc;
margin: 1em 0;
}
ol {
display: block;
list-style-type: decimal;
margin: 1em 0;
}
li {
display: list-item;
}
/* nested lists have no top/bottom margins */
ul ul, ul ol, ul dir, ul menu, ul dl,
ol ul, ol ol, ol dir, ol menu, ol dl,
dir ul, dir ol, dir dir, dir menu, dir dl,
menu ul, menu ol, menu dir, menu menu, menu dl,
dl ul, dl ol, dl dir, dl menu, dl dl {
margin-top: 0;
margin-bottom: 0;
}
/* 2 deep unordered lists use a circle */
ol ul, ul ul, menu ul, dir ul,
ol menu, ul menu, menu menu, dir menu,
ol dir, ul dir, menu dir, dir dir {
list-style-type: circle;
}
/* 3 deep (or more) unordered lists use a square */
ol ol ul, ol ul ul, ol menu ul, ol dir ul,
ol ol menu, ol ul menu, ol menu menu, ol dir menu,
ol ol dir, ol ul dir, ol menu dir, ol dir dir,
ul ol ul, ul ul ul, ul menu ul, ul dir ul,
ul ol menu, ul ul menu, ul menu menu, ul dir menu,
ul ol dir, ul ul dir, ul menu dir, ul dir dir,
menu ol ul, menu ul ul, menu menu ul, menu dir ul,
menu ol menu, menu ul menu, menu menu menu, menu dir menu,
menu ol dir, menu ul dir, menu menu dir, menu dir dir,
dir ol ul, dir ul ul, dir menu ul, dir dir ul,
dir ol menu, dir ul menu, dir menu menu, dir dir menu,
dir ol dir, dir ul dir, dir menu dir, dir dir dir {
list-style-type: square;
}
/* leafs */
/* <hr> noshade and color attributes are handled completely by
* the nsHTMLHRElement attribute mapping code
*/
hr {
display: block;
height: 2px;
border: 1px inset;
margin: 0.5em auto 0.5em auto;
color: gray;
}
hr[size="1"] {
border-style: solid none none none;
}
img[usemap], object[usemap] {
color: blue;
}
frameset {
display: block ! important;
position: static ! important;
float: none ! important;
border: none ! important;
}
frame {
border: none ! important;
}
iframe {
border: 2px inset;
}
noframes {
display: none;
}
spacer {
position: static ! important;
float: none ! important;
}
/* focusable content: anything w/ tabindex >=0 is focusable */
abbr:focus, acronym:focus, address:focus, applet:focus, b:focus,
base:focus, big:focus, blockquote:focus, br:focus, canvas:focus, caption:focus,
center:focus, cite:focus, code:focus, col:focus, colgroup:focus, dd:focus,
del:focus, dfn:focus, dir:focus, div:focus, dl:focus, dt:focus, em:focus,
fieldset:focus, font:focus, form:focus, h1:focus, h2:focus, h3:focus, h4:focus,
h5:focus, h6:focus, hr:focus, i:focus, img:focus, ins:focus,
kbd:focus, label:focus, legend:focus, li:focus, link:focus, menu:focus,
object:focus, ol:focus, p:focus, pre:focus, q:focus, s:focus, samp:focus,
small:focus, span:focus, strike:focus, strong:focus, sub:focus, sup:focus,
table:focus, tbody:focus, td:focus, tfoot:focus, th:focus, thead:focus,
tr:focus, tt:focus, u:focus, ul:focus, var:focus {
/* Don't specify the outline-color, we should always use initial value. */
outline: 1px dotted;
}
/* hidden elements */
area, base, basefont, head, meta, script, style, title,
noembed, param, link {
display: none;
}
/* Page breaks at body tags, to help out with LIT-generation */
body {
page-break-before: always;
}
/* Explicit line-breaks are blocks, sure... */
br {
display: block;
}

View File

@ -0,0 +1,31 @@
'''
LZX compression/decompression wrapper.
'''
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Marshall T. Vandegrift <llasram@gmail.com>'
import sys
from calibre import plugins
_lzx, _error = plugins['lzx']
if _lzx is None:
raise RuntimeError('Failed to load the lzx plugin: %s' % _error)
__all__ = ['Compressor', 'Decompressor', 'LZXError']
LZXError = _lzx.LZXError
Compressor = _lzx.Compressor
class Decompressor(object):
def __init__(self, wbits):
self.wbits = wbits
self.blocksize = 1 << wbits
_lzx.init(wbits)
def decompress(self, data, outlen):
return _lzx.decompress(data, outlen)
def reset(self):
return _lzx.reset()

View File

@ -4,6 +4,9 @@ Modified version of SHA-1 used in Microsoft LIT files.
Adapted from the PyPy pure-Python SHA-1 implementation.
"""
__license__ = 'GPL v3'
__copyright__ = '2008, Marshall T. Vandegrift <llasram@gmail.com>'
import struct, copy
# ======================================================================

View File

@ -0,0 +1,770 @@
'''
Basic support for manipulating OEB 1.x/2.0 content and metadata.
'''
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Marshall T. Vandegrift <llasram@gmail.com>'
import os
import sys
from collections import defaultdict
from types import StringTypes
from itertools import izip, count
from urlparse import urldefrag, urlparse, urlunparse
from urllib import unquote as urlunquote
import logging
from lxml import etree
from calibre import LoggingInterface
XML_PARSER = etree.XMLParser(recover=True, resolve_entities=False)
XML_NS = 'http://www.w3.org/XML/1998/namespace'
XHTML_NS = 'http://www.w3.org/1999/xhtml'
OPF1_NS = 'http://openebook.org/namespaces/oeb-package/1.0/'
OPF2_NS = 'http://www.idpf.org/2007/opf'
DC09_NS = 'http://purl.org/metadata/dublin_core'
DC10_NS = 'http://purl.org/dc/elements/1.0/'
DC11_NS = 'http://purl.org/dc/elements/1.1/'
XSI_NS = 'http://www.w3.org/2001/XMLSchema-instance'
DCTERMS_NS = 'http://purl.org/dc/terms/'
NCX_NS = 'http://www.daisy.org/z3986/2005/ncx/'
XPNSMAP = {'h': XHTML_NS, 'o1': OPF1_NS, 'o2': OPF2_NS,
'd09': DC09_NS, 'd10': DC10_NS, 'd11': DC11_NS,
'xsi': XSI_NS, 'dt': DCTERMS_NS, 'ncx': NCX_NS}
def XML(name): return '{%s}%s' % (XML_NS, name)
def XHTML(name): return '{%s}%s' % (XHTML_NS, name)
def OPF(name): return '{%s}%s' % (OPF2_NS, name)
def DC(name): return '{%s}%s' % (DC11_NS, name)
def NCX(name): return '{%s}%s' % (NCX_NS, name)
XHTML_MIME = 'application/xhtml+xml'
CSS_MIME = 'text/css'
NCX_MIME = 'application/x-dtbncx+xml'
OPF_MIME = 'application/oebps-package+xml'
OEB_DOC_MIME = 'text/x-oeb1-document'
OEB_CSS_MIME = 'text/x-oeb1-css'
OEB_STYLES = set([CSS_MIME, OEB_CSS_MIME, 'text/x-oeb-css'])
OEB_DOCS = set([XHTML_MIME, 'text/html', OEB_DOC_MIME, 'text/x-oeb-document'])
def element(parent, *args, **kwargs):
if parent is not None:
return etree.SubElement(parent, *args, **kwargs)
return etree.Element(*args, **kwargs)
def namespace(name):
if '}' in name:
return name.split('}', 1)[0][1:]
return ''
def barename(name):
if '}' in name:
return name.split('}', 1)[1]
return name
def xpath(elem, expr):
return elem.xpath(expr, namespaces=XPNSMAP)
URL_UNSAFE = r"""`!@#$%^&*[](){}?+=;:'",<>\| """
def urlquote(href):
result = []
for char in href:
if char in URL_UNSAFE:
char = "%%%02x" % ord(char)
result.append(char)
return ''.join(result)
def urlnormalize(href):
parts = urlparse(href)
parts = (part.replace('\\', '/') for part in parts)
parts = (urlunquote(part) for part in parts)
parts = (urlquote(part) for part in parts)
return urlunparse(parts)
class FauxLogger(object):
def __getattr__(self, name):
return self
def __call__(self, message):
print message
class AbstractContainer(object):
def read_xml(self, path):
return etree.fromstring(
self.read(path), parser=XML_PARSER,
base_url=os.path.dirname(path))
class DirContainer(AbstractContainer):
def __init__(self, rootdir):
self.rootdir = rootdir
def read(self, path):
path = os.path.join(self.rootdir, path)
with open(urlunquote(path), 'rb') as f:
return f.read()
def write(self, path, data):
path = os.path.join(self.rootdir, path)
with open(urlunquote(path), 'wb') as f:
return f.write(data)
def exists(self, path):
path = os.path.join(self.rootdir, path)
return os.path.isfile(path)
class Metadata(object):
TERMS = set(['contributor', 'coverage', 'creator', 'date', 'description',
'format', 'identifier', 'language', 'publisher', 'relation',
'rights', 'source', 'subject', 'title', 'type'])
OPF1_NSMAP = {'dc': DC11_NS, 'oebpackage': OPF1_NS}
OPF2_NSMAP = {'opf': OPF2_NS, 'dc': DC11_NS, 'dcterms': DCTERMS_NS,
'xsi': XSI_NS}
class Item(object):
def __init__(self, term, value, fq_attrib={}):
self.fq_attrib = dict(fq_attrib)
if term == OPF('meta') and not value:
term = self.fq_attrib.pop('name')
value = self.fq_attrib.pop('content')
elif term in Metadata.TERMS and not namespace(term):
term = DC(term)
self.term = term
self.value = value
self.attrib = attrib = {}
for fq_attr in fq_attrib:
attr = barename(fq_attr)
attrib[attr] = fq_attrib[fq_attr]
def __getattr__(self, name):
name = name.replace('_', '-')
try:
return self.attrib[name]
except KeyError:
raise AttributeError(
'%r object has no attribute %r' \
% (self.__class__.__name__, name))
def __repr__(self):
return 'Item(term=%r, value=%r, attrib=%r)' \
% (barename(self.term), self.value, self.attrib)
def __str__(self):
return str(self.value)
def __unicode__(self):
return unicode(self.value)
def to_opf1(self, dcmeta=None, xmeta=None):
if namespace(self.term) == DC11_NS:
name = DC(barename(self.term).title())
elem = element(dcmeta, name, attrib=self.attrib)
elem.text = self.value
else:
elem = element(xmeta, 'meta', attrib=self.attrib)
elem.attrib['name'] = self.term
elem.attrib['content'] = self.value
return elem
def to_opf2(self, parent=None):
if namespace(self.term) == DC11_NS:
elem = element(parent, self.term, attrib=self.fq_attrib)
elem.text = self.value
else:
elem = element(parent, OPF('meta'), attrib=self.fq_attrib)
elem.attrib['name'] = self.term
elem.attrib['content'] = self.value
return elem
def __init__(self, oeb):
self.oeb = oeb
self.items = defaultdict(list)
def add(self, term, value, attrib={}):
item = self.Item(term, value, attrib)
items = self.items[barename(item.term)]
items.append(item)
return item
def iterkeys(self):
for key in self.items:
yield key
__iter__ = iterkeys
def __getitem__(self, key):
return self.items[key]
def __contains__(self, key):
return key in self.items
def __getattr__(self, term):
return self.items[term]
def to_opf1(self, parent=None):
elem = element(parent, 'metadata')
dcmeta = element(elem, 'dc-metadata', nsmap=self.OPF1_NSMAP)
xmeta = element(elem, 'x-metadata')
for term in self.items:
for item in self.items[term]:
item.to_opf1(dcmeta, xmeta)
if 'ms-chaptertour' not in self.items:
chaptertour = self.Item('ms-chaptertour', 'chaptertour')
chaptertour.to_opf1(dcmeta, xmeta)
return elem
def to_opf2(self, parent=None):
elem = element(parent, OPF('metadata'), nsmap=self.OPF2_NSMAP)
for term in self.items:
for item in self.items[term]:
item.to_opf2(elem)
return elem
class Manifest(object):
class Item(object):
def __init__(self, id, href, media_type, fallback=None, loader=str):
self.id = id
self.href = self.path = urlnormalize(href)
self.media_type = media_type
self.fallback = fallback
self.spine_position = None
self.linear = True
self._loader = loader
self._data = None
def __repr__(self):
return 'Item(id=%r, href=%r, media_type=%r)' \
% (self.id, self.href, self.media_type)
def data():
def fget(self):
if self._data:
return self._data
data = self._loader(self.href)
if self.media_type == XHTML_MIME:
data = etree.fromstring(data, parser=XML_PARSER)
if namespace(data.tag) != XHTML_NS:
data.attrib['xmlns'] = XHTML_NS
data = etree.tostring(data)
data = etree.fromstring(data, parser=XML_PARSER)
elif self.media_type.startswith('application/') \
and self.media_type.endswith('+xml'):
data = etree.fromstring(data, parser=XML_PARSER)
return data
def fset(self, value):
self._data = value
def fdel(self):
self._data = None
return property(fget, fset, fdel)
data = data()
def __cmp__(self, other):
result = cmp(self.spine_position, other.spine_position)
if result != 0:
return result
return cmp(self.id, other.id)
def __init__(self, oeb):
self.oeb = oeb
self.items = {}
self.hrefs = {}
def add(self, id, href, media_type, fallback=None):
item = self.Item(
id, href, media_type, fallback, self.oeb.container.read)
self.items[item.id] = item
self.hrefs[item.href] = item
return item
def remove(self, id):
href = self.items[id].href
del self.items[id]
del self.hrefs[href]
def __iter__(self):
for id in self.items:
yield id
def __getitem__(self, id):
return self.items[id]
def values(self):
for item in self.items.values():
yield item
def items(self):
for id, item in self.refs.items():
yield id, items
def __contains__(self, key):
return key in self.items
def to_opf1(self, parent=None):
elem = element(parent, 'manifest')
for item in self.items.values():
media_type = item.media_type
if media_type == XHTML_MIME:
media_type = OEB_DOC_MIME
elif media_type == CSS_MIME:
media_type = OEB_CSS_MIME
attrib = {'id': item.id, 'href': item.href,
'media-type': media_type}
if item.fallback:
attrib['fallback'] = item.fallback
element(elem, 'item', attrib=attrib)
return elem
def to_opf2(self, parent=None):
elem = element(parent, OPF('manifest'))
for item in self.items.values():
attrib = {'id': item.id, 'href': item.href,
'media-type': item.media_type}
if item.fallback:
attrib['fallback'] = item.fallback
element(elem, OPF('item'), attrib=attrib)
return elem
class Spine(object):
def __init__(self, oeb):
self.oeb = oeb
self.items = []
def add(self, item, linear):
if isinstance(linear, StringTypes):
linear = linear.lower()
if linear is None or linear in ('yes', 'true'):
linear = True
elif linear in ('no', 'false'):
linear = False
item.linear = linear
item.spine_position = len(self.items)
self.items.append(item)
return item
def __iter__(self):
for item in self.items:
yield item
def __getitem__(self, index):
return self.items[index]
def __len__(self):
return len(self.items)
def __contains__(self, item):
return (item in self.items)
def to_opf1(self, parent=None):
elem = element(parent, 'spine')
for item in self.items:
if item.linear:
element(elem, 'itemref', attrib={'idref': item.id})
return elem
def to_opf2(self, parent=None):
elem = element(parent, OPF('spine'))
for item in self.items:
attrib = {'idref': item.id}
if not item.linear:
attrib['linear'] = 'no'
element(elem, OPF('itemref'), attrib=attrib)
return elem
class Guide(object):
class Reference(object):
def __init__(self, type, title, href):
self.type = type
self.title = title
self.href = urlnormalize(href)
def __repr__(self):
return 'Reference(type=%r, title=%r, href=%r)' \
% (self.type, self.title, self.href)
def __init__(self, oeb):
self.oeb = oeb
self.refs = {}
def add(self, type, title, href):
ref = self.Reference(type, title, href)
self.refs[type] = ref
return ref
def by_type(self, type):
return self.ref_types[type]
def iterkeys(self):
for type in self.refs:
yield type
__iter__ = iterkeys
def values(self):
for ref in self.refs.values():
yield ref
def items(self):
for type, ref in self.refs.items():
yield type, ref
def __getitem__(self, index):
return self.refs[index]
def __contains__(self, key):
return key in self.refs
def to_opf1(self, parent=None):
elem = element(parent, 'guide')
for ref in self.refs.values():
attrib = {'type': ref.type, 'href': ref.href}
if ref.title:
attrib['title'] = ref.title
element(elem, 'reference', attrib=attrib)
return elem
def to_opf2(self, parent=None):
elem = element(parent, OPF('guide'))
for ref in self.refs.values():
attrib = {'type': ref.type, 'href': ref.href}
if ref.title:
attrib['title'] = ref.title
element(elem, OPF('reference'), attrib=attrib)
return elem
class TOC(object):
def __init__(self, title=None, href=None, klass=None, id=None):
self.title = title
self.href = urlnormalize(href) if href else href
self.klass = klass
self.id = id
self.nodes = []
def add(self, title, href, klass=None, id=None):
node = TOC(title, href, klass, id)
self.nodes.append(node)
return node
def __iter__(self):
for node in self.nodes:
yield node
def __getitem__(self, index):
return self.nodes[index]
def depth(self, level=0):
if self.nodes:
return self.nodes[0].depth(level+1)
return level
def to_opf1(self, tour):
for node in self.nodes:
element(tour, 'site', attrib={
'title': node.title, 'href': node.href})
node.to_opf1(tour)
return tour
def to_ncx(self, parent, playorder=None, depth=1):
if not playorder: playorder = [0]
for node in self.nodes:
playorder[0] += 1
point = etree.SubElement(parent,
NCX('navPoint'), attrib={'playOrder': str(playorder[0])})
if self.klass:
point.attrib['class'] = node.klass
if self.id:
point.attrib['id'] = node.id
label = etree.SubElement(point, NCX('navLabel'))
etree.SubElement(label, NCX('text')).text = node.title
href = node.href if depth > 1 else urldefrag(node.href)[0]
child = etree.SubElement(point,
NCX('content'), attrib={'src': href})
node.to_ncx(point, playorder, depth+1)
return parent
class OEBBook(object):
def __init__(self, opfpath, container=None, logger=FauxLogger()):
if not container:
container = DirContainer(os.path.dirname(opfpath))
opfpath = os.path.basename(opfpath)
self.container = container
self.logger = logger
opf = self._read_opf(opfpath)
self._all_from_opf(opf)
def _convert_opf1(self, opf):
nroot = etree.Element(OPF('package'),
nsmap={None: OPF2_NS}, version="2.0", **dict(opf.attrib))
metadata = etree.SubElement(nroot, OPF('metadata'),
nsmap={'opf': OPF2_NS, 'dc': DC11_NS,
'xsi': XSI_NS, 'dcterms': DCTERMS_NS})
for prefix in ('d11', 'd10', 'd09'):
elements = xpath(opf, 'metadata/dc-metadata/%s:*' % prefix)
if elements: break
for element in elements:
if not element.text: continue
tag = barename(element.tag).lower()
element.tag = '{%s}%s' % (DC11_NS, tag)
for name in element.attrib:
if name in ('role', 'file-as', 'scheme'):
nsname = '{%s}%s' % (OPF2_NS, name)
element.attrib[nsname] = element.attrib[name]
del element.attrib[name]
metadata.append(element)
for element in opf.xpath('metadata/x-metadata/meta'):
metadata.append(element)
for item in opf.xpath('manifest/item'):
media_type = item.attrib['media-type'].lower()
if media_type in OEB_DOCS:
media_type = XHTML_MIME
elif media_type in OEB_STYLES:
media_type = CSS_MIME
item.attrib['media-type'] = media_type
for tag in ('manifest', 'spine', 'tours', 'guide'):
for element in opf.xpath(tag):
nroot.append(element)
return etree.fromstring(etree.tostring(nroot), parser=XML_PARSER)
def _read_opf(self, opfpath):
opf = self.container.read_xml(opfpath)
version = float(opf.get('version', 1.0))
if version < 2.0:
opf = self._convert_opf1(opf)
return opf
def _metadata_from_opf(self, opf):
uid = opf.attrib['unique-identifier']
self.metadata = metadata = Metadata(self)
for elem in xpath(opf, '/o2:package/o2:metadata/*'):
if elem.text or elem.attrib:
metadata.add(elem.tag, elem.text, elem.attrib)
for item in metadata.identifier:
if item.id == uid:
self.uid = item
break
else:
self.logger.log_warn(u'Unique-identifier %r not found.' % uid)
self.uid = metadata.identifier[0]
def _manifest_from_opf(self, opf):
self.manifest = manifest = Manifest(self)
for elem in xpath(opf, '/o2:package/o2:manifest/o2:item'):
href = elem.get('href')
if not self.container.exists(href):
self.logger.log_warn(u'Manifest item %r not found.' % href)
continue
manifest.add(elem.get('id'), href, elem.get('media-type'),
elem.get('fallback'))
def _spine_from_opf(self, opf):
self.spine = spine = Spine(self)
for elem in xpath(opf, '/o2:package/o2:spine/o2:itemref'):
idref = elem.get('idref')
if idref not in self.manifest:
self.logger.log_warn(u'Spine item %r not found.' % idref)
continue
item = self.manifest[idref]
spine.add(item, elem.get('linear'))
extras = []
for item in self.manifest.values():
if item.media_type == XHTML_MIME \
and item not in spine:
extras.append(item)
extras.sort()
for item in extras:
spine.add(item, False)
def _guide_from_opf(self, opf):
self.guide = guide = Guide(self)
for elem in xpath(opf, '/o2:package/o2:guide/o2:reference'):
href = elem.get('href')
path, frag = urldefrag(href)
if path not in self.manifest.hrefs:
self.logger.log_warn(u'Guide reference %r not found' % href)
continue
guide.add(elem.get('type'), elem.get('title'), href)
def _toc_from_navpoint(self, toc, navpoint):
children = xpath(navpoint, 'ncx:navPoint')
for child in children:
title = ''.join(xpath(child, 'ncx:navLabel/ncx:text/text()'))
href = xpath(child, 'ncx:content/@src')[0]
id = child.get('id')
klass = child.get('class')
node = toc.add(title, href, id=id, klass=klass)
self._toc_from_navpoint(node, child)
def _toc_from_ncx(self, opf):
result = xpath(opf, '/o2:package/o2:spine/@toc')
if not result:
expr = '/o2:package/o2:manifest/o2:item[@media-type="%s"]/@id'
result = xpath(opf, expr % NCX_MIME)
if len(result) != 1:
return False
id = result[0]
ncx = self.manifest[id].data
self.manifest.remove(id)
title = xpath(ncx, 'ncx:docTitle/ncx:text/text()')[0]
self.toc = toc = TOC(title)
navmaps = xpath(ncx, 'ncx:navMap')
for navmap in navmaps:
self._toc_from_navpoint(toc, navmap)
return True
def _toc_from_tour(self, opf):
result = xpath(opf, '/o2:package/o2:tours/o2:tour')
if not result:
return False
tour = result[0]
self.toc = toc = TOC(tour.get('title'))
sites = xpath(tour, 'o2:site')
for site in sites:
toc.add(site.get('title'), site.get('href'))
return True
def _toc_from_html(self, opf):
if 'toc' not in self.guide:
return False
self.toc = toc = TOC()
itempath, frag = urldefrag(self.guide['toc'].href)
item = self.manifest.hrefs[itempath]
html = item.data
if frag:
elems = xpath(html, './/*[@id="%s"]' % frag)
if not elems:
elems = xpath(html, './/*[@name="%s"]' % frag)
elem = elems[0] if elems else html
while elem != html and not xpath(elem, './/h:a[@href]'):
elem = elem.getparent()
html = elem
titles = defaultdict(list)
order = []
for anchor in xpath(html, './/h:a[@href]'):
href = anchor.attrib['href']
path, frag = urldefrag(href)
if not path:
href = '#'.join((itempath, frag))
title = ' '.join(xpath(anchor, './/text()'))
href = urlnormalize(href)
if href not in titles:
order.append(href)
titles[href].append(title)
for href in order:
toc.add(' '.join(titles[href]), href)
return True
def _toc_from_spine(self, opf):
self.toc = toc = TOC()
titles = []
headers = []
for item in self.spine:
if not item.linear: continue
html = item.data
title = xpath(html, '/h:html/h:head/h:title/text()')
if title: titles.append(title[0])
headers.append('(unlabled)')
for tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'strong'):
expr = '/h:html/h:body//h:%s[position()=1]/text()' % (tag,)
header = xpath(html, expr)
if header:
headers[-1] = header[0]
break
use = titles
if len(titles) > len(set(titles)):
use = headers
for title, item in izip(use, self.spine):
if not item.linear: continue
toc.add(title, item.href)
return True
def _toc_from_opf(self, opf):
if self._toc_from_ncx(opf): return
if self._toc_from_tour(opf): return
if self._toc_from_html(opf): return
self._toc_from_spine(opf)
def _all_from_opf(self, opf):
self._metadata_from_opf(opf)
self._manifest_from_opf(opf)
self._spine_from_opf(opf)
self._guide_from_opf(opf)
self._toc_from_opf(opf)
def to_opf1(self):
package = etree.Element('package',
attrib={'unique-identifier': self.uid.id})
metadata = self.metadata.to_opf1(package)
manifest = self.manifest.to_opf1(package)
spine = self.spine.to_opf1(package)
tours = element(package, 'tours')
tour = element(tours, 'tour',
attrib={'id': 'chaptertour', 'title': 'Chapter Tour'})
self.toc.to_opf1(tour)
guide = self.guide.to_opf1(package)
return {OPF_MIME: ('content.opf', package)}
def _generate_ncx_item(self):
id = 'ncx'
index = 0
while id in self.manifest:
id = 'ncx' + str(index)
index = index + 1
href = 'toc'
index = 0
while (href + '.ncx') in self.manifest.hrefs:
href = 'toc' + str(index)
href += '.ncx'
return (id, href)
def _to_ncx(self):
ncx = etree.Element(NCX('ncx'), attrib={'version': '2005-1'},
nsmap={None: NCX_NS})
head = etree.SubElement(ncx, NCX('head'))
etree.SubElement(head, NCX('meta'),
attrib={'name': 'dtb:uid', 'content': unicode(self.uid)})
etree.SubElement(head, NCX('meta'),
attrib={'name': 'dtb:depth', 'content': str(self.toc.depth())})
etree.SubElement(head, NCX('meta'),
attrib={'name': 'dtb:totalPageCount', 'content': '0'})
etree.SubElement(head, NCX('meta'),
attrib={'name': 'dtb:maxPageNumber', 'content': '0'})
title = etree.SubElement(ncx, NCX('docTitle'))
text = etree.SubElement(title, NCX('text'))
text.text = unicode(self.metadata.title[0])
navmap = etree.SubElement(ncx, NCX('navMap'))
self.toc.to_ncx(navmap)
return ncx
def to_opf2(self):
package = etree.Element(OPF('package'),
attrib={'version': '2.0', 'unique-identifier': self.uid.id},
nsmap={None: OPF2_NS})
metadata = self.metadata.to_opf2(package)
manifest = self.manifest.to_opf2(package)
id, href = self._generate_ncx_item()
etree.SubElement(manifest, OPF('item'),
attrib={'id': id, 'href': href, 'media-type': NCX_MIME})
spine = self.spine.to_opf2(package)
spine.attrib['toc'] = id
guide = self.guide.to_opf2(package)
ncx = self._to_ncx()
return {OPF_MIME: ('content.opf', package),
NCX_MIME: (href, ncx)}
def main(argv=sys.argv):
for arg in argv[1:]:
oeb = OEBBook(arg)
for name, doc in oeb.to_opf1().values():
print etree.tostring(doc, pretty_print=True)
for name, doc in oeb.to_opf2().values():
print etree.tostring(doc, pretty_print=True)
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@ -10,10 +10,12 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net> ' \
import sys, struct, cStringIO, os
import functools
import re
from urlparse import urldefrag
from lxml import etree
from calibre.ebooks.lit import LitError
from calibre.ebooks.lit.maps import OPF_MAP, HTML_MAP
import calibre.ebooks.lit.mssha1 as mssha1
from calibre.ebooks.lit.oeb import urlnormalize
from calibre.ebooks import DRMError
from calibre import plugins
lzx, lxzerror = plugins['lzx']
@ -110,7 +112,7 @@ class UnBinary(object):
AMPERSAND_RE = re.compile(
r'&(?!(?:#[0-9]+|#x[0-9a-fA-F]+|[a-zA-Z_:][a-zA-Z0-9.-_:]+);)')
OPEN_ANGLE_RE = re.compile(r'<<(?![!]--)')
CLOSE_ANGLE_RE = re.compile(r'(?<!--)>>')
CLOSE_ANGLE_RE = re.compile(r'(?<!--)>>(?=>>|[^>])')
DOUBLE_ANGLE_RE = re.compile(r'([<>])\1')
def __init__(self, bin, path, manifest={}, map=HTML_MAP):
@ -322,12 +324,12 @@ class UnBinary(object):
href += c
count -= 1
if count == 0:
doc, m, frag = href[1:].partition('#')
doc, frag = urldefrag(href[1:])
path = self.item_path(doc)
if m and frag:
path += m + frag
self.buf.write((u'"%s"' % path).encode(
'ascii', 'xmlcharrefreplace'))
if frag:
path = '#'.join((path, frag))
path = urlnormalize(path)
self.buf.write((u'"%s"' % path).encode('utf-8'))
state = 'get attr'
return index
@ -385,7 +387,7 @@ def preserve(function):
class LitReader(object):
PIECE_SIZE = 16
XML_PARSER = etree.XMLParser(
remove_blank_text=True, resolve_entities=False)
recover=True, resolve_entities=False)
def magic():
@preserve
@ -781,7 +783,7 @@ class LitReader(object):
try:
result.append(
lzx.decompress(content[base:size], window_bytes))
except lzx.LzxError:
except lzx.LZXError:
self._warn("LZX decompression error; skipping chunk")
bytes_remaining -= window_bytes
base = size
@ -791,7 +793,7 @@ class LitReader(object):
lzx.reset()
try:
result.append(lzx.decompress(content[base:], bytes_remaining))
except lzx.LzxError:
except lzx.LZXError:
self._warn("LZX decompression error; skipping chunk")
bytes_remaining = 0
if bytes_remaining > 0:

View File

@ -0,0 +1,444 @@
# -*- encoding: utf-8 -*-
'''
CSS property propagation class.
'''
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Marshall T. Vandegrift <llasram@gmail.com>'
import sys
import os
import locale
import codecs
import itertools
import types
import re
import copy
import cssutils
from cssutils.css import CSSStyleRule, CSSPageRule, CSSStyleDeclaration, \
CSSValueList, cssproperties
from lxml import etree
from calibre.ebooks.lit.oeb import XHTML_NS, CSS_MIME, OEB_STYLES
from calibre.ebooks.lit.oeb import barename, urlnormalize
from calibre.resources import html_css
HTML_CSS_STYLESHEET = cssutils.parseString(html_css)
XHTML_CSS_NAMESPACE = "@namespace url(http://www.w3.org/1999/xhtml);\n"
INHERITED = set(['azimuth', 'border-collapse', 'border-spacing',
'caption-side', 'color', 'cursor', 'direction', 'elevation',
'empty-cells', 'font-family', 'font-size', 'font-style',
'font-variant', 'font-weight', 'letter-spacing',
'line-height', 'list-style-image', 'list-style-position',
'list-style-type', 'orphans', 'page-break-inside',
'pitch-range', 'pitch', 'quotes', 'richness', 'speak-header',
'speak-numeral', 'speak-punctuation', 'speak', 'speech-rate',
'stress', 'text-align', 'text-indent', 'text-transform',
'visibility', 'voice-family', 'volume', 'white-space',
'widows', 'word-spacing'])
DEFAULTS = {'azimuth': 'center', 'background-attachment': 'scroll',
'background-color': 'transparent', 'background-image': 'none',
'background-position': '0% 0%', 'background-repeat': 'repeat',
'border-bottom-color': ':color', 'border-bottom-style': 'none',
'border-bottom-width': 'medium', 'border-collapse': 'separate',
'border-left-color': ':color', 'border-left-style': 'none',
'border-left-width': 'medium', 'border-right-color': ':color',
'border-right-style': 'none', 'border-right-width': 'medium',
'border-spacing': 0, 'border-top-color': ':color',
'border-top-style': 'none', 'border-top-width': 'medium', 'bottom':
'auto', 'caption-side': 'top', 'clear': 'none', 'clip': 'auto',
'color': 'black', 'content': 'normal', 'counter-increment': 'none',
'counter-reset': 'none', 'cue-after': 'none', 'cue-before': 'none',
'cursor': 'auto', 'direction': 'ltr', 'display': 'inline',
'elevation': 'level', 'empty-cells': 'show', 'float': 'none',
'font-family': 'serif', 'font-size': 'medium', 'font-style':
'normal', 'font-variant': 'normal', 'font-weight': 'normal',
'height': 'auto', 'left': 'auto', 'letter-spacing': 'normal',
'line-height': 'normal', 'list-style-image': 'none',
'list-style-position': 'outside', 'list-style-type': 'disc',
'margin-bottom': 0, 'margin-left': 0, 'margin-right': 0,
'margin-top': 0, 'max-height': 'none', 'max-width': 'none',
'min-height': 0, 'min-width': 0, 'orphans': '2',
'outline-color': 'invert', 'outline-style': 'none',
'outline-width': 'medium', 'overflow': 'visible', 'padding-bottom':
0, 'padding-left': 0, 'padding-right': 0, 'padding-top': 0,
'page-break-after': 'auto', 'page-break-before': 'auto',
'page-break-inside': 'auto', 'pause-after': 0, 'pause-before':
0, 'pitch': 'medium', 'pitch-range': '50', 'play-during': 'auto',
'position': 'static', 'quotes': u"'' '' '' ''", 'richness':
'50', 'right': 'auto', 'speak': 'normal', 'speak-header': 'once',
'speak-numeral': 'continuous', 'speak-punctuation': 'none',
'speech-rate': 'medium', 'stress': '50', 'table-layout': 'auto',
'text-align': 'left', 'text-decoration': 'none', 'text-indent':
0, 'text-transform': 'none', 'top': 'auto', 'unicode-bidi':
'normal', 'vertical-align': 'baseline', 'visibility': 'visible',
'voice-family': 'default', 'volume': 'medium', 'white-space':
'normal', 'widows': '2', 'width': 'auto', 'word-spacing': 'normal',
'z-index': 'auto'}
FONT_SIZE_NAMES = set(['xx-small', 'x-small', 'small', 'medium', 'large',
'x-large', 'xx-large'])
FONT_SIZE_LIST = [('xx-small', 1, 6.),
('x-small', None, 7.),
('small', 2, 8.),
('medium', 3, 9.),
('large', 4, 11.),
('x-large', 5, 13.),
('xx-large', 6, 15.),
(None, 7, 17.)]
FONT_SIZE_BY_NAME = {}
FONT_SIZE_BY_NUM = {}
for name, num, size in FONT_SIZE_LIST:
FONT_SIZE_BY_NAME[name] = size
FONT_SIZE_BY_NUM[num] = size
XPNSMAP = {'h': XHTML_NS,}
def xpath(elem, expr):
return elem.xpath(expr, namespaces=XPNSMAP)
class Page(object):
def __init__(self, width, height, dpi):
self.width = float(width)
self.height = float(height)
self.dpi = float(dpi)
class Profiles(object):
PRS500 = Page(584, 754, 168.451)
PRS505 = PRS500
class Stylizer(object):
STYLESHEETS = {}
def __init__(self, tree, path, oeb, page=Profiles.PRS505):
self.page = page
base = os.path.dirname(path)
basename = os.path.basename(path)
cssname = os.path.splitext(basename)[0] + '.css'
stylesheets = [HTML_CSS_STYLESHEET]
head = xpath(tree, '/h:html/h:head')[0]
parser = cssutils.CSSParser()
parser.setFetcher(lambda path: ('utf-8', oeb.container.read(path)))
for elem in head:
tag = barename(elem.tag)
if tag == 'style':
text = ''.join(elem.text)
stylesheet = parser.parseString(text, href=cssname)
stylesheets.append(stylesheet)
elif tag == 'link' \
and elem.get('rel', 'stylesheet') == 'stylesheet' \
and elem.get('type', CSS_MIME) in OEB_STYLES:
href = urlnormalize(elem.attrib['href'])
path = os.path.join(base, href)
path = os.path.normpath(path).replace('\\', '/')
if path in self.STYLESHEETS:
stylesheet = self.STYLESHEETS[path]
else:
data = XHTML_CSS_NAMESPACE
data += oeb.manifest.hrefs[path].data
stylesheet = parser.parseString(data, href=path)
self.STYLESHEETS[path] = stylesheet
stylesheets.append(stylesheet)
rules = []
index = 0
self.stylesheets = set()
for stylesheet in stylesheets:
href = stylesheet.href
self.stylesheets.add(href)
for rule in stylesheet.cssRules:
rules.extend(self.flatten_rule(rule, href, index))
index = index + 1
rules.sort()
self.rules = rules
self._styles = {}
def flatten_rule(self, rule, href, index):
results = []
if isinstance(rule, CSSStyleRule):
style = self.flatten_style(rule.style)
for selector in rule.selectorList:
specificity = selector.specificity + (index,)
text = selector.selectorText
selector = list(selector.seq)
results.append((specificity, selector, style, text, href))
elif isinstance(rule, CSSPageRule):
style = self.flatten_style(rule.style)
results.append(((0, 0, 0, 0), [], style, '@page', href))
return results
def flatten_style(self, cssstyle):
style = {}
for prop in cssstyle:
name = prop.name
if name in ('margin', 'padding'):
style.update(self._normalize_edge(prop.cssValue, name))
elif name == 'font':
style.update(self._normalize_font(prop.cssValue))
else:
style[name] = prop.value
if 'font-size' in style:
size = style['font-size']
if size == 'normal': size = 'medium'
if size in FONT_SIZE_NAMES:
style['font-size'] = "%dpt" % FONT_SIZE_BY_NAME[size]
return style
def _normalize_edge(self, cssvalue, name):
style = {}
if isinstance(cssvalue, CSSValueList):
primitives = [v.cssText for v in cssvalue]
else:
primitives = [cssvalue.cssText]
if len(primitives) == 1:
value, = primitives
values = [value, value, value, value]
elif len(primitives) == 2:
vert, horiz = primitives
values = [vert, horiz, vert, horiz]
elif len(primitives) == 3:
top, horiz, bottom = primitives
values = [top, horiz, bottom, horiz]
else:
values = primitives[:4]
edges = ('top', 'right', 'bottom', 'left')
for edge, value in itertools.izip(edges, values):
style["%s-%s" % (name, edge)] = value
return style
def _normalize_font(self, cssvalue):
composition = ('font-style', 'font-variant', 'font-weight',
'font-size', 'line-height', 'font-family')
style = {}
if cssvalue.cssText == 'inherit':
for key in composition:
style[key] = 'inherit'
else:
primitives = [v.cssText for v in cssvalue]
primitites.reverse()
value = primitives.pop()
for key in composition:
if cssproperties.cssvalues[key](value):
style[key] = value
if not primitives: break
value = primitives.pop()
for key in composition:
if key not in style:
style[key] = DEFAULTS[key]
return style
def style(self, element):
try: return self._styles[element]
except: pass
return Style(element, self)
def stylesheet(self, name, font_scale=None):
rules = []
for _, _, style, selector, href in self.rules:
if href != name: continue
if font_scale and 'font-size' in style and \
style['font-size'].endswith('pt'):
style = copy.copy(style)
size = float(style['font-size'][:-2])
style['font-size'] = "%.2fpt" % (size * font_scale)
style = ';\n '.join(': '.join(item) for item in style.items())
rules.append('%s {\n %s;\n}' % (selector, style))
return '\n'.join(rules)
class Style(object):
def __init__(self, element, stylizer):
self._element = element
self._page = stylizer.page
self._stylizer = stylizer
self._style = self._assemble_style(element, stylizer)
stylizer._styles[element] = self
def _assemble_style(self, element, stylizer):
result = {}
rules = stylizer.rules
for _, selector, style, _, _ in rules:
if self._selects_element(element, selector):
result.update(style)
try:
style = CSSStyleDeclaration(element.attrib['style'])
result.update(stylizer.flatten_style(style))
except KeyError:
pass
return result
def _selects_element(self, element, selector):
def _selects_element(element, items, index):
if index == -1:
return True
item = items[index]
if item.type == 'universal':
pass
elif item.type == 'type-selector':
name1 = ("{%s}%s" % item.value).lower()
name2 = element.tag.lower()
if name1 != name2:
return False
elif item.type == 'id':
name1 = item.value[1:]
name2 = element.get('id', '')
if name1 != name2:
return False
elif item.type == 'class':
name = item.value[1:].lower()
classes = element.get('class', '').lower().split()
if name not in classes:
return False
elif item.type == 'child':
parent = element.getparent()
if parent is None:
return False
element = parent
elif item.type == 'descendant':
element = element.getparent()
while element is not None:
if _selects_element(element, items, index - 1):
return True
element = element.getparent()
return False
elif item.type == 'pseudo-class':
if item.value == ':first-child':
e = element.getprevious()
if e is not None:
return False
else:
return False
elif item.type == 'pseudo-element':
return False
else:
return False
return _selects_element(element, items, index - 1)
return _selects_element(element, selector, len(selector) - 1)
def _has_parent(self):
parent = self._element.getparent()
return (parent is not None) \
and (parent in self._stylizer._styles)
def __getitem__(self, name):
domname = cssproperties._toDOMname(name)
if hasattr(self, domname):
return getattr(self, domname)
return self._unit_convert(self._get(name))
def _get(self, name):
result = None
if name in self._style:
result = self._style[name]
if (result == 'inherit'
or (result is None and name in INHERITED
and self._has_parent())):
styles = self._stylizer._styles
result = styles[self._element.getparent()]._get(name)
if result is None:
result = DEFAULTS[name]
return result
def _unit_convert(self, value, base=None, font=None):
if isinstance(value, (int, long, float)):
return value
try:
if float(value) == 0:
return 0.0
except:
pass
result = value
m = re.search(
r"^(-*[0-9]*\.?[0-9]*)\s*(%|em|px|mm|cm|in|pt|pc)$", value)
if m is not None and m.group(1):
value = float(m.group(1))
unit = m.group(2)
if unit == '%':
base = base or self.width
result = (value/100.0) * base
elif unit == 'px':
result = value * 72.0 / self._page.dpi
elif unit == 'in':
result = value * 72.0
elif unit == 'pt':
result = value
elif unit == 'em':
font = font or self.fontSize
result = value * font
elif unit == 'pc':
result = value * 12.0
elif unit == 'mm':
result = value * 0.04
elif unit == 'cm':
result = value * 0.40
return result
@property
def fontSize(self):
def normalize_fontsize(value, base=None):
result = None
factor = None
if value == 'inherit':
value = 'medium'
if value in FONT_SIZE_NAMES:
result = FONT_SIZE_BY_NAME[value]
elif value == 'smaller':
factor = 1.0/1.2
for _, _, size in FONT_SIZE_LIST:
if base <= size: break
factor = None
result = size
elif value == 'larger':
factor = 1.2
for _, _, size in reversed(FONT_SIZE_LIST):
if base >= size: break
factor = None
result = size
else:
result = self._unit_convert(value, base=base, font=base)
if result < 0:
result = normalize_fontsize("smaller", base)
if factor:
result = factor * base
return result
result = None
if self._has_parent():
styles = self._stylizer._styles
base = styles[self._element.getparent()].fontSize
else:
base = normalize_fontsize(DEFAULTS['font-size'])
if 'font-size' in self._style:
size = self._style['font-size']
result = normalize_fontsize(size, base)
else:
result = base
self.__dict__['fontSize'] = result
return result
@property
def width(self):
result = None
base = None
if self._has_parent():
styles = self._stylizer._styles
base = styles[self._element.getparent()].width
else:
base = self._page.width
if 'width' in self._style:
width = self._style['width']
if width == 'auto':
result = base
else:
result = self._unit_convert(width, base=base)
else:
result = base
self.__dict__['width'] = result
return result
def __str__(self):
items = self._style.items()
return '; '.join("%s: %s" % (key, val) for key, val in items)

View File

@ -0,0 +1,753 @@
'''
Basic support for writing LIT files.
'''
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2008, Marshall T. Vandegrift <llasram@gmail.com>'
import sys
import os
from cStringIO import StringIO
from struct import pack
from itertools import izip, count, chain
import time
import random
import re
import copy
import uuid
import functools
import logging
from urlparse import urldefrag
from urllib import unquote as urlunquote
from lxml import etree
from calibre.ebooks.lit.reader import DirectoryEntry
import calibre.ebooks.lit.maps as maps
from calibre.ebooks.lit.oeb import OEB_DOCS, OEB_STYLES, OEB_CSS_MIME, \
CSS_MIME, OPF_MIME, XML_NS, XML
from calibre.ebooks.lit.oeb import namespace, barename, urlnormalize, xpath
from calibre.ebooks.lit.oeb import FauxLogger, OEBBook
from calibre.ebooks.lit.stylizer import Stylizer
from calibre.ebooks.lit.lzx import Compressor
import calibre
from calibre import LoggingInterface
from calibre import plugins
msdes, msdeserror = plugins['msdes']
import calibre.ebooks.lit.mssha1 as mssha1
from calibre.customize.ui import run_plugins_on_postprocess
__all__ = ['LitWriter']
LIT_IMAGES = set(['image/png', 'image/jpeg', 'image/gif'])
LIT_MIMES = OEB_DOCS | OEB_STYLES | LIT_IMAGES
MS_COVER_TYPE = 'other.ms-coverimage-standard'
ALL_MS_COVER_TYPES = [
(MS_COVER_TYPE, 'Standard cover image'),
('other.ms-thumbimage-standard', 'Standard thumbnail image'),
('other.ms-coverimage', 'PocketPC cover image'),
('other.ms-thumbimage', 'PocketPC thumbnail image'),
]
def invert_tag_map(tag_map):
tags, dattrs, tattrs = tag_map
tags = dict((tags[i], i) for i in xrange(len(tags)))
dattrs = dict((v, k) for k, v in dattrs.items())
tattrs = [dict((v, k) for k, v in (map or {}).items()) for map in tattrs]
for map in tattrs:
if map: map.update(dattrs)
tattrs[0] = dattrs
return tags, tattrs
OPF_MAP = invert_tag_map(maps.OPF_MAP)
HTML_MAP = invert_tag_map(maps.HTML_MAP)
LIT_MAGIC = 'ITOLITLS'
LITFILE_GUID = "{0A9007C1-4076-11D3-8789-0000F8105754}"
PIECE3_GUID = "{0A9007C3-4076-11D3-8789-0000F8105754}"
PIECE4_GUID = "{0A9007C4-4076-11D3-8789-0000F8105754}"
DESENCRYPT_GUID = "{67F6E4A2-60BF-11D3-8540-00C04F58C3CF}"
LZXCOMPRESS_GUID = "{0A9007C6-4076-11D3-8789-0000F8105754}"
def packguid(guid):
values = guid[1:9], guid[10:14], guid[15:19], \
guid[20:22], guid[22:24], guid[25:27], guid[27:29], \
guid[29:31], guid[31:33], guid[33:35], guid[35:37]
values = [int(value, 16) for value in values]
return pack("<LHHBBBBBBBB", *values)
FLAG_OPENING = (1 << 0)
FLAG_CLOSING = (1 << 1)
FLAG_BLOCK = (1 << 2)
FLAG_HEAD = (1 << 3)
FLAG_ATOM = (1 << 4)
FLAG_CUSTOM = (1 << 15)
ATTR_NUMBER = 0xffff
PIECE_SIZE = 16
PRIMARY_SIZE = 40
SECONDARY_SIZE = 232
DCHUNK_SIZE = 0x2000
CCHUNK_SIZE = 0x0200
ULL_NEG1 = 0xffffffffffffffff
ROOT_OFFSET = 1284508585713721976
ROOT_SIZE = 4165955342166943123
BLOCK_CAOL = \
"\x43\x41\x4f\x4c\x02\x00\x00\x00" \
"\x50\x00\x00\x00\x37\x13\x03\x00" \
"\x00\x00\x00\x00\x00\x20\x00\x00" \
"\x00\x02\x00\x00\x00\x00\x10\x00" \
"\x00\x00\x02\x00\x00\x00\x00\x00" \
"\x00\x00\x00\x00\x00\x00\x00\x00"
BLOCK_ITSF = \
"\x49\x54\x53\x46\x04\x00\x00\x00" \
"\x20\x00\x00\x00\x01\x00\x00\x00"
MSDES_CONTROL = \
"\x03\x00\x00\x00\x29\x17\x00\x00" \
"\x01\x00\x00\x00\xa5\xa5\x00\x00"
LZXC_CONTROL = \
"\x07\x00\x00\x00\x4c\x5a\x58\x43" \
"\x03\x00\x00\x00\x04\x00\x00\x00" \
"\x04\x00\x00\x00\x02\x00\x00\x00" \
"\x00\x00\x00\x00\x00\x00\x00\x00"
COLLAPSE = re.compile(r'[ \t\r\n\v]+')
def prefixname(name, nsrmap):
prefix = nsrmap[namespace(name)]
if not prefix:
return barename(name)
return ':'.join((prefix, barename(name)))
def decint(value):
bytes = []
while True:
b = value & 0x7f
value >>= 7
if bytes:
b |= 0x80
bytes.append(chr(b))
if value == 0:
break
return ''.join(reversed(bytes))
def randbytes(n):
return ''.join(chr(random.randint(0, 255)) for x in xrange(n))
def warn(x):
print x
class ReBinary(object):
NSRMAP = {'': None, XML_NS: 'xml'}
def __init__(self, root, path, oeb, map=HTML_MAP, logger=FauxLogger()):
self.path = path
self.logger = logger
self.dir = os.path.dirname(path)
self.manifest = oeb.manifest
self.tags, self.tattrs = map
self.buf = StringIO()
self.anchors = []
self.page_breaks = []
self.is_html = is_html = map is HTML_MAP
self.stylizer = Stylizer(root, path, oeb) if is_html else None
self.tree_to_binary(root)
self.content = self.buf.getvalue()
self.ahc = self.build_ahc() if is_html else None
self.aht = self.build_aht() if is_html else None
def write(self, *values):
for value in values:
if isinstance(value, (int, long)):
value = unichr(value)
self.buf.write(value.encode('utf-8'))
def is_block(self, style):
return style['display'] not in ('inline', 'inline-block')
def tree_to_binary(self, elem, nsrmap=NSRMAP, parents=[],
inhead=False, preserve=False):
if not isinstance(elem.tag, basestring):
self.write(etree.tostring(elem))
return
nsrmap = copy.copy(nsrmap)
attrib = dict(elem.attrib)
style = self.stylizer.style(elem) if self.stylizer else None
for key, value in elem.nsmap.items():
if value not in nsrmap or nsrmap[value] != key:
xmlns = ('xmlns:' + key) if key else 'xmlns'
attrib[xmlns] = value
nsrmap[value] = key
tag = prefixname(elem.tag, nsrmap)
tag_offset = self.buf.tell()
if tag == 'head':
inhead = True
flags = FLAG_OPENING
if not elem.text and len(elem) == 0:
flags |= FLAG_CLOSING
if inhead:
flags |= FLAG_HEAD
if style and self.is_block(style):
flags |= FLAG_BLOCK
self.write(0, flags)
tattrs = self.tattrs[0]
if tag in self.tags:
index = self.tags[tag]
self.write(index)
if self.tattrs[index]:
tattrs = self.tattrs[index]
else:
self.write(FLAG_CUSTOM, len(tag)+1, tag)
last_break = self.page_breaks[-1][0] if self.page_breaks else None
if style and last_break != tag_offset \
and style['page-break-before'] not in ('avoid', 'auto'):
self.page_breaks.append((tag_offset, list(parents)))
for attr, value in attrib.items():
attr = prefixname(attr, nsrmap)
if attr in ('href', 'src'):
value = urlnormalize(value)
path, frag = urldefrag(value)
prefix = unichr(3)
if path in self.manifest.hrefs:
prefix = unichr(2)
value = self.manifest.hrefs[path].id
if frag:
value = '#'.join((value, frag))
value = prefix + value
elif attr in ('id', 'name'):
self.anchors.append((value, tag_offset))
elif attr.startswith('ms--'):
attr = '%' + attr[4:]
elif tag == 'link' and attr == 'type' and value in OEB_STYLES:
value = OEB_CSS_MIME
if attr in tattrs:
self.write(tattrs[attr])
else:
self.write(FLAG_CUSTOM, len(attr)+1, attr)
try:
self.write(ATTR_NUMBER, int(value)+1)
except ValueError:
self.write(len(value)+1, value)
self.write(0)
old_preserve = preserve
if style:
preserve = (style['white-space'] in ('pre', 'pre-wrap'))
xml_space = elem.get(XML('space'))
if xml_space == 'preserve':
preserve = True
elif xml_space == 'normal':
preserve = False
if elem.text:
if preserve:
self.write(elem.text)
elif len(elem) == 0 or not elem.text.isspace():
self.write(COLLAPSE.sub(' ', elem.text))
# else: de nada
parents.append(tag_offset)
child = cstyle = nstyle = None
for next in chain(elem, [None]):
if self.stylizer:
nstyle = None if next is None else self.stylizer.style(next)
if child is not None:
if not preserve \
and (inhead or not nstyle
or self.is_block(cstyle)
or self.is_block(nstyle)) \
and child.tail and child.tail.isspace():
child.tail = None
self.tree_to_binary(child, nsrmap, parents, inhead, preserve)
child, cstyle = next, nstyle
parents.pop()
preserve = old_preserve
if not flags & FLAG_CLOSING:
self.write(0, (flags & ~FLAG_OPENING) | FLAG_CLOSING, 0)
if elem.tail and tag != 'html':
tail = elem.tail
if not preserve:
tail = COLLAPSE.sub(' ', tail)
self.write(tail)
if style and style['page-break-after'] not in ('avoid', 'auto'):
self.page_breaks.append((self.buf.tell(), list(parents)))
def build_ahc(self):
if len(self.anchors) > 6:
self.logger.log_warn("More than six anchors in file %r. " \
"Some links may not work properly." % self.path)
data = StringIO()
data.write(unichr(len(self.anchors)).encode('utf-8'))
for anchor, offset in self.anchors:
data.write(unichr(len(anchor)).encode('utf-8'))
data.write(anchor)
data.write(pack('<I', offset))
return data.getvalue()
def build_aht(self):
return pack('<I', 0)
def preserve(function):
def wrapper(self, *args, **kwargs):
opos = self._stream.tell()
try:
return function(self, *args, **kwargs)
finally:
self._stream.seek(opos)
functools.update_wrapper(wrapper, function)
return wrapper
class LitWriter(object):
def __init__(self, oeb, logger=FauxLogger()):
self._oeb = oeb
self._logger = logger
self._litize_oeb()
def _litize_oeb(self):
oeb = self._oeb
oeb.metadata.add('calibre-oeb2lit-version', calibre.__version__)
cover = None
if oeb.metadata.cover:
id = str(oeb.metadata.cover[0])
cover = oeb.manifest[id]
elif MS_COVER_TYPE in oeb.guide:
href = oeb.guide[MS_COVER_TYPE].href
cover = oeb.manifest.hrefs[href]
elif 'cover' in oeb.guide:
href = oeb.guide['cover'].href
cover = oeb.manifest.hrefs[href]
else:
html = oeb.spine[0].data
imgs = xpath(html, '//img[position()=1]')
href = imgs[0].get('src') if imgs else None
cover = oeb.manifest.hrefs[href] if href else None
if cover:
if not oeb.metadata.cover:
oeb.metadata.add('cover', cover.id)
for type, title in ALL_MS_COVER_TYPES:
if type not in oeb.guide:
oeb.guide.add(type, title, cover.href)
else:
self._logger.log_warn('No suitable cover image found.')
def dump(self, stream):
self._stream = stream
self._sections = [StringIO() for i in xrange(4)]
self._directory = []
self._meta = None
self._dump()
def _write(self, *data):
for datum in data:
self._stream.write(datum)
@preserve
def _writeat(self, pos, *data):
self._stream.seek(pos)
self._write(*data)
def _tell(self):
return self._stream.tell()
def _dump(self):
# Build content sections
self._build_sections()
# Build directory chunks
dcounts, dchunks, ichunk = self._build_dchunks()
# Write headers
self._write(LIT_MAGIC)
self._write(pack('<IIII',
1, PRIMARY_SIZE, 5, SECONDARY_SIZE))
self._write(packguid(LITFILE_GUID))
offset = self._tell()
pieces = list(xrange(offset, offset + (PIECE_SIZE * 5), PIECE_SIZE))
self._write((5 * PIECE_SIZE) * '\0')
aoli1 = len(dchunks) if ichunk else ULL_NEG1
last = len(dchunks) - 1
ddepth = 2 if ichunk else 1
self._write(pack('<IIQQQQIIIIQIIQQQQIIIIQIIIIQ',
2, 0x98, aoli1, 0, last, 0, DCHUNK_SIZE, 2, 0, ddepth, 0,
len(self._directory), 0, ULL_NEG1, 0, 0, 0, CCHUNK_SIZE, 2,
0, 1, 0, len(dcounts), 0, 0x100000, 0x20000, 0))
self._write(BLOCK_CAOL)
self._write(BLOCK_ITSF)
conoff_offset = self._tell()
timestamp = int(time.time())
self._write(pack('<QII', 0, timestamp, 0x409))
# Piece #0
piece0_offset = self._tell()
self._write(pack('<II', 0x1fe, 0))
filesz_offset = self._tell()
self._write(pack('<QQ', 0, 0))
self._writeat(pieces[0], pack('<QQ',
piece0_offset, self._tell() - piece0_offset))
# Piece #1: Directory chunks
piece1_offset = self._tell()
number = len(dchunks) + ((ichunk and 1) or 0)
self._write('IFCM', pack('<IIIQQ',
1, DCHUNK_SIZE, 0x100000, ULL_NEG1, number))
for dchunk in dchunks:
self._write(dchunk)
if ichunk:
self._write(ichunk)
self._writeat(pieces[1], pack('<QQ',
piece1_offset, self._tell() - piece1_offset))
# Piece #2: Count chunks
piece2_offset = self._tell()
self._write('IFCM', pack('<IIIQQ',
1, CCHUNK_SIZE, 0x20000, ULL_NEG1, 1))
cchunk = StringIO()
last = 0
for i, dcount in izip(count(), dcounts):
cchunk.write(decint(last))
cchunk.write(decint(dcount))
cchunk.write(decint(i))
last = dcount
cchunk = cchunk.getvalue()
rem = CCHUNK_SIZE - (len(cchunk) + 50)
self._write('AOLL', pack('<IQQQQQ',
rem, 0, ULL_NEG1, ULL_NEG1, 0, 1))
filler = '\0' * rem
self._write(cchunk, filler, pack('<H', len(dcounts)))
self._writeat(pieces[2], pack('<QQ',
piece2_offset, self._tell() - piece2_offset))
# Piece #3: GUID3
piece3_offset = self._tell()
self._write(packguid(PIECE3_GUID))
self._writeat(pieces[3], pack('<QQ',
piece3_offset, self._tell() - piece3_offset))
# Piece #4: GUID4
piece4_offset = self._tell()
self._write(packguid(PIECE4_GUID))
self._writeat(pieces[4], pack('<QQ',
piece4_offset, self._tell() - piece4_offset))
# The actual section content
content_offset = self._tell()
self._writeat(conoff_offset, pack('<Q', content_offset))
self._write(self._sections[0].getvalue())
self._writeat(filesz_offset, pack('<Q', self._tell()))
def _add_file(self, name, data, secnum=0):
if len(data) > 0:
section = self._sections[secnum]
offset = section.tell()
section.write(data)
else:
offset = 0
self._directory.append(
DirectoryEntry(name, secnum, offset, len(data)))
def _add_folder(self, name, offset=0, size=0):
if not name.endswith('/'):
name += '/'
self._directory.append(
DirectoryEntry(name, 0, offset, size))
def _djoin(self, *names):
return '/'.join(names)
def _build_sections(self):
self._add_folder('/', ROOT_OFFSET, ROOT_SIZE)
self._build_data()
self._build_manifest()
self._build_page_breaks()
self._build_meta()
self._build_drm_storage()
self._build_version()
self._build_namelist()
self._build_storage()
self._build_transforms()
def _build_data(self):
self._add_folder('/data')
for item in self._oeb.manifest.values():
if item.media_type not in LIT_MIMES:
self._logger.log_warn("File %r of unknown media-type %r " \
"excluded from output." % (item.href, item.media_type))
continue
name = '/data/' + item.id
data = item.data
secnum = 0
if not isinstance(data, basestring):
self._add_folder(name)
rebin = ReBinary(data, item.href, self._oeb, map=HTML_MAP,
logger=self._logger)
self._add_file(name + '/ahc', rebin.ahc, 0)
self._add_file(name + '/aht', rebin.aht, 0)
item.page_breaks = rebin.page_breaks
data = rebin.content
name = name + '/content'
secnum = 1
self._add_file(name, data, secnum)
item.size = len(data)
def _build_manifest(self):
states = ['linear', 'nonlinear', 'css', 'images']
manifest = dict((state, []) for state in states)
for item in self._oeb.manifest.values():
if item.spine_position is not None:
key = 'linear' if item.linear else 'nonlinear'
manifest[key].append(item)
elif item.media_type == CSS_MIME:
manifest['css'].append(item)
elif item.media_type in LIT_IMAGES:
manifest['images'].append(item)
data = StringIO()
data.write(pack('<Bc', 1, '\\'))
offset = 0
for state in states:
items = manifest[state]
items.sort()
data.write(pack('<I', len(items)))
for item in items:
id, media_type = item.id, item.media_type
href = urlunquote(item.href)
item.offset = offset \
if state in ('linear', 'nonlinear') else 0
data.write(pack('<I', item.offset))
entry = [unichr(len(id)), unicode(id),
unichr(len(href)), unicode(href),
unichr(len(media_type)), unicode(media_type)]
for value in entry:
data.write(value.encode('utf-8'))
data.write('\0')
offset += item.size
self._add_file('/manifest', data.getvalue())
def _build_page_breaks(self):
pb1 = StringIO()
pb2 = StringIO()
pb3 = StringIO()
pb3cur = 0
bits = 0
for item in self._oeb.spine:
page_breaks = copy.copy(item.page_breaks)
if not item.linear:
page_breaks.insert(0, (0, []))
for pbreak, parents in page_breaks:
pb3cur = (pb3cur << 2) | 1
if len(parents) > 1:
pb3cur |= 0x2
bits += 2
if bits >= 8:
pb3.write(pack('<B', pb3cur))
pb3cur = 0
bits = 0
pbreak += item.offset
pb1.write(pack('<II', pbreak, pb2.tell()))
pb2.write(pack('<I', len(parents)))
for parent in parents:
pb2.write(pack('<I', parent))
if bits != 0:
pb3cur <<= (8 - bits)
pb3.write(pack('<B', pb3cur))
self._add_file('/pb1', pb1.getvalue(), 0)
self._add_file('/pb2', pb2.getvalue(), 0)
self._add_file('/pb3', pb3.getvalue(), 0)
def _build_meta(self):
_, meta = self._oeb.to_opf1()[OPF_MIME]
meta.attrib['ms--minimum_level'] = '0'
meta.attrib['ms--attr5'] = '1'
meta.attrib['ms--guid'] = '{%s}' % str(uuid.uuid4()).upper()
rebin = ReBinary(meta, 'content.opf', self._oeb, map=OPF_MAP,
logger=self._logger)
meta = rebin.content
self._meta = meta
self._add_file('/meta', meta)
def _build_drm_storage(self):
drmsource = u'Free as in freedom\0'.encode('utf-16-le')
self._add_file('/DRMStorage/DRMSource', drmsource)
tempkey = self._calculate_deskey([self._meta, drmsource])
msdes.deskey(tempkey, msdes.EN0)
self._add_file('/DRMStorage/DRMSealed', msdes.des("\0" * 16))
self._bookkey = '\0' * 8
self._add_file('/DRMStorage/ValidationStream', 'MSReader', 3)
def _build_version(self):
self._add_file('/Version', pack('<HH', 8, 1))
def _build_namelist(self):
data = StringIO()
data.write(pack('<HH', 0x3c, len(self._sections)))
names = ['Uncompressed', 'MSCompressed', 'EbEncryptDS',
'EbEncryptOnlyDS']
for name in names:
data.write(pack('<H', len(name)))
data.write(name.encode('utf-16-le'))
data.write('\0\0')
self._add_file('::DataSpace/NameList', data.getvalue())
def _build_storage(self):
mapping = [(1, 'MSCompressed', (LZXCOMPRESS_GUID,)),
(2, 'EbEncryptDS', (LZXCOMPRESS_GUID, DESENCRYPT_GUID)),
(3, 'EbEncryptOnlyDS', (DESENCRYPT_GUID,)),]
for secnum, name, transforms in mapping:
root = '::DataSpace/Storage/' + name
data = self._sections[secnum].getvalue()
cdata, sdata, tdata, rdata = '', '', '', ''
for guid in transforms:
tdata = packguid(guid) + tdata
sdata = sdata + pack('<Q', len(data))
if guid == DESENCRYPT_GUID:
cdata = MSDES_CONTROL + cdata
if not data: continue
msdes.deskey(self._bookkey, msdes.EN0)
pad = 8 - (len(data) & 0x7)
if pad != 8:
data = data + ('\0' * pad)
data = msdes.des(data)
elif guid == LZXCOMPRESS_GUID:
cdata = LZXC_CONTROL + cdata
if not data: continue
unlen = len(data)
lzx = Compressor(17)
data, rtable = lzx.compress(data, flush=True)
rdata = StringIO()
rdata.write(pack('<IIIIQQQQ',
3, len(rtable), 8, 0x28, unlen, len(data), 0x8000, 0))
for uncomp, comp in rtable[:-1]:
rdata.write(pack('<Q', comp))
rdata = rdata.getvalue()
self._add_file(root + '/Content', data)
self._add_file(root + '/ControlData', cdata)
self._add_file(root + '/SpanInfo', sdata)
self._add_file(root + '/Transform/List', tdata)
troot = root + '/Transform'
for guid in transforms:
dname = self._djoin(troot, guid, 'InstanceData')
self._add_folder(dname)
if guid == LZXCOMPRESS_GUID:
dname += '/ResetTable'
self._add_file(dname, rdata)
def _build_transforms(self):
for guid in (LZXCOMPRESS_GUID, DESENCRYPT_GUID):
self._add_folder('::Transform/'+ guid)
def _calculate_deskey(self, hashdata):
prepad = 2
hash = mssha1.new()
for data in hashdata:
if prepad > 0:
data = ("\000" * prepad) + data
prepad = 0
postpad = 64 - (len(data) % 64)
if postpad < 64:
data = data + ("\000" * postpad)
hash.update(data)
digest = hash.digest()
key = [0] * 8
for i in xrange(0, len(digest)):
key[i % 8] ^= ord(digest[i])
return ''.join(chr(x) for x in key)
def _build_dchunks(self):
ddata = []
directory = list(self._directory)
directory.sort(cmp=lambda x, y: \
cmp(x.name.lower(), y.name.lower()))
qrn = 1 + (1 << 2)
dchunk = StringIO()
dcount = 0
quickref = []
name = directory[0].name
for entry in directory:
next = ''.join([decint(len(entry.name)), entry.name,
decint(entry.section), decint(entry.offset),
decint(entry.size)])
usedlen = dchunk.tell() + len(next) + (len(quickref) * 2) + 52
if usedlen >= DCHUNK_SIZE:
ddata.append((dchunk.getvalue(), quickref, dcount, name))
dchunk = StringIO()
dcount = 0
quickref = []
name = entry.name
if (dcount % qrn) == 0:
quickref.append(dchunk.tell())
dchunk.write(next)
dcount = dcount + 1
ddata.append((dchunk.getvalue(), quickref, dcount, name))
cidmax = len(ddata) - 1
rdcount = 0
dchunks = []
dcounts = []
ichunk = None
if len(ddata) > 1:
ichunk = StringIO()
for cid, (content, quickref, dcount, name) in izip(count(), ddata):
dchunk = StringIO()
prev = cid - 1 if cid > 0 else ULL_NEG1
next = cid + 1 if cid < cidmax else ULL_NEG1
rem = DCHUNK_SIZE - (len(content) + 50)
pad = rem - (len(quickref) * 2)
dchunk.write('AOLL')
dchunk.write(pack('<IQQQQQ', rem, cid, prev, next, rdcount, 1))
dchunk.write(content)
dchunk.write('\0' * pad)
for ref in reversed(quickref):
dchunk.write(pack('<H', ref))
dchunk.write(pack('<H', dcount))
rdcount = rdcount + dcount
dchunks.append(dchunk.getvalue())
dcounts.append(dcount)
if ichunk:
ichunk.write(decint(len(name)))
ichunk.write(name)
ichunk.write(decint(cid))
if ichunk:
rem = DCHUNK_SIZE - (ichunk.tell() + 16)
pad = rem - 2
ichunk = ''.join(['AOLI', pack('<IQ', rem, len(dchunks)),
ichunk.getvalue(), ('\0' * pad), pack('<H', len(dchunks))])
return dcounts, dchunks, ichunk
def option_parser():
from calibre.utils.config import OptionParser
parser = OptionParser(usage=_('%prog [options] OPFFILE'))
parser.add_option(
'-o', '--output', default=None,
help=_('Output file. Default is derived from input filename.'))
parser.add_option(
'--verbose', default=False, action='store_true',
help=_('Useful for debugging.'))
return parser
def oeb2lit(opts, opfpath):
logger = LoggingInterface(logging.getLogger('oeb2lit'))
logger.setup_cli_handler(opts.verbose)
litpath = opts.output
if litpath is None:
litpath = os.path.basename(opfpath)
litpath = os.path.splitext(litpath)[0] + '.lit'
litpath = os.path.abspath(litpath)
lit = LitWriter(OEBBook(opfpath, logger=logger), logger=logger)
with open(litpath, 'wb') as f:
lit.dump(f)
run_plugins_on_postprocess(litpath, 'lit')
logger.log_info(_('Output written to ')+litpath)
def main(argv=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(argv[1:])
if len(args) != 1:
parser.print_help()
return 1
opfpath = args[0]
oeb2lit(opts, opfpath)
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@ -108,6 +108,8 @@ def option_parser(usage, gui_mode=False):
help=_('Add a header to all the pages with title and author.'))
laf.add_option('--headerformat', default="%t by %a", dest='headerformat', type='string',
help=_('Set the format of the header. %a is replaced by the author and %t by the title. Default is %default'))
laf.add_option('--header-separation', default=0, type='int',
help=_('Add extra spacing below the header. Default is %default px.'))
laf.add_option('--override-css', default=None, dest='_override_css', type='string',
help=_('Override the CSS. Can be either a path to a CSS stylesheet or a string. If it is a string it is interpreted as CSS.'))
laf.add_option('--use-spine', default=False, dest='use_spine', action='store_true',
@ -260,10 +262,11 @@ def Book(options, logger, font_delta=0, header=None,
hb.append(header)
hdr.PutObj(hb)
ps['headheight'] = profile.header_height
ps['headsep'] = options.header_separation
ps['header'] = hdr
ps['topmargin'] = 0
ps['textheight'] = profile.screen_height - (options.bottom_margin + ps['topmargin']) \
- ps['headheight'] - profile.fudge
- ps['headheight'] - ps['headsep'] - profile.fudge
fontsize = int(10*profile.font_size+font_delta*20)
baselineskip = fontsize + 20

View File

@ -18,6 +18,8 @@ from calibre.ebooks.lrf.epub.convert_from import process_file as epub2lrf
from calibre.ebooks.lrf.mobi.convert_from import process_file as mobi2lrf
from calibre.ebooks.lrf.fb2.convert_from import process_file as fb22lrf
from calibre.customize.ui import run_plugins_on_postprocess, run_plugins_on_preprocess
def largest_file(files):
maxsize, file = 0, None
for f in files:
@ -108,6 +110,7 @@ def odt2lrf(path, options, logger):
def process_file(path, options, logger=None):
path = os.path.abspath(os.path.expanduser(path))
path = run_plugins_on_preprocess(path)
tdir = None
if logger is None:
level = logging.DEBUG if options.verbose else logging.INFO
@ -160,6 +163,7 @@ def process_file(path, options, logger=None):
if not convertor:
raise UnknownFormatError(_('Converting from %s to LRF is not supported.')%ext)
convertor(path, options, logger)
finally:
os.chdir(cwd)
if tdir and os.path.exists(tdir):

View File

@ -10,6 +10,14 @@ Based on ideas from comiclrf created by FangornUK.
import os, sys, shutil, traceback, textwrap
from uuid import uuid4
try:
from reportlab.pdfgen import canvas
_reportlab = True
except:
_reportlab = False
from calibre import extract, terminal_controller, __appname__, __version__
from calibre.utils.config import Config, StringConfig
from calibre.ptempfile import PersistentTemporaryDirectory
@ -19,6 +27,7 @@ from calibre.ebooks.lrf.pylrs.pylrs import Book, BookSetting, ImageStream, Image
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.opf import OPFCreator
from calibre.ebooks.epub.from_html import config as html2epub_config, convert as html2epub
from calibre.customize.ui import run_plugins_on_preprocess
try:
from calibre.utils.PythonMagickWand import \
NewMagickWand, NewPixelWand, \
@ -41,6 +50,8 @@ except:
PROFILES = {
# Name : (width, height) in pixels
'prs500':(584, 754),
# The SONY's LRF renderer (on the PRS500) only uses the first 800x600 block of the image
'prs500-landscape': (784, 1200-92)
}
def extract_comic(path_to_comic_file):
@ -276,7 +287,7 @@ def process_pages(pages, opts, update):
failures += failures_
return ans, failures, tdir
def config(defaults=None):
def config(defaults=None,output_format='lrf'):
desc = _('Options to control the conversion of comics (CBR, CBZ) files into ebooks')
if defaults is None:
c = Config('comic', desc)
@ -313,10 +324,13 @@ def config(defaults=None):
help=_('Be verbose, useful for debugging. Can be specified multiple times for greater verbosity.'))
c.add_opt('no_progress_bar', ['--no-progress-bar'], default=False,
help=_("Don't show progress bar."))
if output_format == 'pdf':
c.add_opt('no_process',['--no_process'], default=False,
help=_("Apply no processing to the image"))
return c
def option_parser():
c = config()
def option_parser(output_format='lrf'):
c = config(output_format=output_format)
return c.option_parser(usage=_('''\
%prog [options] comic.cb[z|r]
@ -379,38 +393,60 @@ def create_lrf(pages, profile, opts, thumbnail=None):
book.renderLrf(open(opts.output, 'wb'))
print _('Output written to'), opts.output
def create_pdf(pages, profile, opts, thumbnail=None):
width, height = PROFILES[profile]
if not _reportlab:
raise RuntimeError('Failed to load reportlab')
pdf = canvas.Canvas(filename=opts.output, pagesize=(width,height+15))
for page in pages:
pdf.drawImage(page, x=0,y=0,width=width, height=height)
pdf.showPage()
# Write the document to disk
pdf.save()
def do_convert(path_to_file, opts, notification=lambda m, p: p, output_format='lrf'):
path_to_file = run_plugins_on_preprocess(path_to_file)
source = path_to_file
if not opts.title:
opts.title = os.path.splitext(os.path.basename(source))[0]
if not opts.output:
opts.output = os.path.abspath(os.path.splitext(os.path.basename(source))[0]+'.'+output_format)
tdir = extract_comic(source)
pages = find_pages(tdir, sort_on_mtime=opts.no_sort, verbose=opts.verbose)
thumbnail = None
if not pages:
raise ValueError('Could not find any pages in the comic: %s'%source)
pages, failures, tdir2 = process_pages(pages, opts, notification)
if not pages:
raise ValueError('Could not find any valid pages in the comic: %s'%source)
if failures:
print 'Could not process the following pages (run with --verbose to see why):'
for f in failures:
print '\t', f
thumbnail = os.path.join(tdir2, 'thumbnail.png')
if not os.access(thumbnail, os.R_OK):
thumbnail = None
if not opts.no_process:
pages, failures, tdir2 = process_pages(pages, opts, notification)
if not pages:
raise ValueError('Could not find any valid pages in the comic: %s'%source)
if failures:
print 'Could not process the following pages (run with --verbose to see why):'
for f in failures:
print '\t', f
thumbnail = os.path.join(tdir2, 'thumbnail.png')
if not os.access(thumbnail, os.R_OK):
thumbnail = None
if output_format == 'lrf':
create_lrf(pages, opts.profile, opts, thumbnail=thumbnail)
else:
if output_format == 'epub':
create_epub(pages, opts.profile, opts, thumbnail=thumbnail)
if output_format == 'pdf':
create_pdf(pages, opts.profile, opts, thumbnail=thumbnail)
shutil.rmtree(tdir)
shutil.rmtree(tdir2)
if not opts.no_process:
shutil.rmtree(tdir2)
def main(args=sys.argv, notification=None, output_format='lrf'):
parser = option_parser()
parser = option_parser(output_format=output_format)
opts, args = parser.parse_args(args)
if len(args) < 2:
parser.print_help()
@ -424,7 +460,6 @@ def main(args=sys.argv, notification=None, output_format='lrf'):
source = os.path.abspath(args[1])
do_convert(source, opts, notification, output_format=output_format)
return 0
if __name__ == '__main__':

View File

@ -2,14 +2,14 @@ __license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
import os, sys, shutil, logging
from tempfile import mkdtemp
from calibre.ebooks.lrf import option_parser as lrf_option_parser
from calibre.ebooks import ConversionError, DRMError
from calibre.ebooks.lrf.html.convert_from import process_file as html_process_file
from calibre.ebooks.metadata.opf import OPF
from calibre.ebooks.metadata.epub import OCFDirReader
from calibre.utils.zipfile import ZipFile
from calibre import __appname__, setup_cli_handlers
from calibre import setup_cli_handlers
from calibre.ptempfile import PersistentTemporaryDirectory
def option_parser():
@ -22,17 +22,16 @@ _('''Usage: %prog [options] mybook.epub
def generate_html(pathtoepub, logger):
if not os.access(pathtoepub, os.R_OK):
raise ConversionError, 'Cannot read from ' + pathtoepub
tdir = mkdtemp(prefix=__appname__+'_')
os.rmdir(tdir)
raise ConversionError('Cannot read from ' + pathtoepub)
tdir = PersistentTemporaryDirectory('_epub2lrf')
#os.rmdir(tdir)
try:
ZipFile(pathtoepub).extractall(tdir)
if os.path.exists(os.path.join(tdir, 'META-INF', 'encryption.xml')):
raise DRMError(os.path.basename(pathtoepub))
except:
if os.path.exists(tdir) and os.path.isdir(tdir):
shutil.rmtree(tdir)
raise ConversionError, '.epub extraction failed'
if os.path.exists(os.path.join(tdir, 'META-INF', 'encryption.xml')):
raise DRMError(os.path.basename(pathtoepub))
return tdir
def process_file(path, options, logger=None):

View File

@ -12,24 +12,61 @@ except ImportError:
'''
Default fonts used in the PRS500
'''
from calibre.ebooks.lrf.fonts.prs500 import tt0003m_, tt0011m_, tt0419m_
SYSTEM_FONT_PATH = '/usr/share/fonts/truetype/ttf-liberation/'
FONT_MAP = {
'Swis721 BT Roman' : tt0003m_,
'Dutch801 Rm BT Roman' : tt0011m_,
'Courier10 BT Roman' : tt0419m_,
'Swis721 BT Roman' : 'tt0003m_',
'Dutch801 Rm BT Roman' : 'tt0011m_',
'Courier10 BT Roman' : 'tt0419m_',
}
LIBERATION_FONT_MAP = {
'Swis721 BT Roman' : 'LiberationSans_Regular',
'Dutch801 Rm BT Roman' : 'LiberationSerif_Regular',
'Courier10 BT Roman' : 'LiberationMono_Regular',
}
SYSTEM_FONT_MAP = {}
for key, val in LIBERATION_FONT_MAP.items():
SYSTEM_FONT_MAP[key] = SYSTEM_FONT_PATH + val.replace('_', '-') + '.ttf'
FONT_FILE_MAP = {}
def get_font_path(name):
if FONT_FILE_MAP.has_key(name) and os.access(FONT_FILE_MAP[name].name, os.R_OK):
return FONT_FILE_MAP[name].name
p = PersistentTemporaryFile('.ttf', 'font_')
p.write(FONT_MAP[name].font_data)
p.close()
FONT_FILE_MAP[name] = p
return p.name
# translate font into file name
fname = FONT_MAP[name]
# first, check configuration in /etc/
etc_file = os.path.join(os.path.sep, 'etc', 'calibre', 'fonts', fname + '.ttf')
if os.access(etc_file, os.R_OK):
return etc_file
# then, try calibre shipped ones
try:
try:
font_mod = __import__('calibre.ebooks.lrf.fonts.prs500', {}, {},
[fname], -1)
except ImportError:
font_mod = __import__('calibre.ebooks.lrf.fonts.liberation', {}, {},
[LIBERATION_FONT_MAP[name]], -1)
p = PersistentTemporaryFile('.ttf', 'font_')
p.write(getattr(font_mod, fname).font_data)
p.close()
FONT_FILE_MAP[name] = p
return p.name
except ImportError:
pass
# finally, try system default ones
if SYSTEM_FONT_MAP.has_key(name) and os.access(SYSTEM_FONT_MAP[name], os.R_OK):
return SYSTEM_FONT_MAP[name]
# not found
raise SystemError, 'font %s (in file %s) not installed' % (name, fname)
def get_font(name, size, encoding='unic'):
'''

View File

@ -12,6 +12,7 @@ from urllib import unquote
from urlparse import urlparse
from math import ceil, floor
from functools import partial
from calibre.customize.ui import run_plugins_on_postprocess
try:
from PIL import Image as PILImage
@ -32,7 +33,6 @@ from calibre.ebooks.lrf.html.table import Table
from calibre import filename_to_utf8, setup_cli_handlers, __appname__, \
fit_image, LoggingInterface, preferred_encoding
from calibre.ptempfile import PersistentTemporaryFile
from calibre.ebooks.metadata.opf import OPFReader
from calibre.devices.interface import Device
from calibre.ebooks.lrf.html.color_map import lrs_color
from calibre.ebooks.chardet import xml_to_unicode
@ -106,6 +106,8 @@ class HTMLConverter(object, LoggingInterface):
(re.compile(r'(<style.*?</style>)', re.IGNORECASE|re.DOTALL),
strip_style_comments),
# Remove self closing script tags as they also mess up BeautifulSoup
(re.compile(r'(?i)<script[^<>]+?/>'), lambda match: ''),
]
# Fix Baen markup
@ -243,7 +245,6 @@ class HTMLConverter(object, LoggingInterface):
self.override_css = {}
self.override_pcss = {}
self.table_render_job_server = None
if self._override_css is not None:
if os.access(self._override_css, os.R_OK):
@ -264,41 +265,37 @@ class HTMLConverter(object, LoggingInterface):
paths = [os.path.abspath(path) for path in paths]
paths = [path.decode(sys.getfilesystemencoding()) if not isinstance(path, unicode) else path for path in paths]
try:
while len(paths) > 0 and self.link_level <= self.link_levels:
for path in paths:
if path in self.processed_files:
continue
try:
self.add_file(path)
except KeyboardInterrupt:
while len(paths) > 0 and self.link_level <= self.link_levels:
for path in paths:
if path in self.processed_files:
continue
try:
self.add_file(path)
except KeyboardInterrupt:
raise
except:
if self.link_level == 0: # Die on errors in the first level
raise
except:
if self.link_level == 0: # Die on errors in the first level
raise
for link in self.links:
if link['path'] == path:
self.links.remove(link)
break
self.log_warn('Could not process '+path)
if self.verbose:
self.log_exception(' ')
self.links = self.process_links()
self.link_level += 1
paths = [link['path'] for link in self.links]
if self.current_page is not None and self.current_page.has_text():
self.book.append(self.current_page)
for text, tb in self.extra_toc_entries:
self.book.addTocEntry(text, tb)
if self.base_font_size > 0:
self.log_info('\tRationalizing font sizes...')
self.book.rationalize_font_sizes(self.base_font_size)
finally:
if self.table_render_job_server is not None:
self.table_render_job_server.killall()
for link in self.links:
if link['path'] == path:
self.links.remove(link)
break
self.log_warn('Could not process '+path)
if self.verbose:
self.log_exception(' ')
self.links = self.process_links()
self.link_level += 1
paths = [link['path'] for link in self.links]
if self.current_page is not None and self.current_page.has_text():
self.book.append(self.current_page)
for text, tb in self.extra_toc_entries:
self.book.addTocEntry(text, tb)
if self.base_font_size > 0:
self.log_info('\tRationalizing font sizes...')
self.book.rationalize_font_sizes(self.base_font_size)
def is_baen(self, soup):
return bool(soup.find('meta', attrs={'name':'Publisher',
@ -334,7 +331,8 @@ class HTMLConverter(object, LoggingInterface):
soup = BeautifulSoup(raw,
convertEntities=BeautifulSoup.XHTML_ENTITIES,
markupMassage=nmassage)
else:
raise
if not self.baen and self.is_baen(soup):
self.baen = True
self.log_info(_('\tBaen file detected. Re-parsing...'))
@ -520,6 +518,8 @@ class HTMLConverter(object, LoggingInterface):
self.book.append(self.current_page)
self.current_page = None
if top not in top.parent.contents: # May have been removed for a cover image
top = top.parent.contents[0]
if not top.has_text() and top.parent.contents.index(top) == len(top.parent.contents)-1:
# Empty block at the bottom of a page
opage = top.parent
@ -809,7 +809,7 @@ class HTMLConverter(object, LoggingInterface):
def append_text(src):
fp, key, variant = self.font_properties(css)
for x, y in [(u'\xa0', ' '), (u'\ufb00', 'ff'), (u'\ufb01', 'fi'), (u'\ufb02', 'fl'), (u'\ufb03', 'ffi'), (u'\ufb04', 'ffl')]:
for x, y in [(u'\xad', ''), (u'\xa0', ' '), (u'\ufb00', 'ff'), (u'\ufb01', 'fi'), (u'\ufb02', 'fl'), (u'\ufb03', 'ffi'), (u'\ufb04', 'ffl')]:
src = src.replace(x, y)
valigner = lambda x: x
@ -1028,6 +1028,8 @@ class HTMLConverter(object, LoggingInterface):
self.current_para = Paragraph()
else:
self.end_page()
if len(self.current_page.contents) == 1 and not self.current_page.has_text():
self.current_page.contents[0:1] = []
self.current_page.append(Canvas(width=pwidth,
height=height))
left = int(floor((pwidth - width)/2.))
@ -1725,15 +1727,11 @@ class HTMLConverter(object, LoggingInterface):
self.process_children(tag, tag_css, tag_pseudo_css)
elif tagname == 'table' and not self.ignore_tables and not self.in_table:
if self.render_tables_as_images:
if self.table_render_job_server is None:
from calibre.parallel import Server
self.table_render_job_server = Server(number_of_workers=1)
print 'Rendering table...'
from calibre.ebooks.lrf.html.table_as_image import render_table
pheight = int(self.current_page.pageStyle.attrs['textheight'])
pwidth = int(self.current_page.pageStyle.attrs['textwidth'])
images = render_table(self.table_render_job_server,
self.soup, tag, tag_css,
images = render_table(self.soup, tag, tag_css,
os.path.dirname(self.target_prefix),
pwidth, pheight, self.profile.dpi,
self.text_size_multiplier_for_rendered_tables)
@ -1846,7 +1844,7 @@ def process_file(path, options, logger=None):
scaled else im
cf = PersistentTemporaryFile(prefix=__appname__+"_", suffix=".jpg")
cf.close()
cim.save(cf.name)
cim.convert('RGB').save(cf.name)
options.cover = cf.name
tim = im.resize((int(0.75*th), th), PILImage.ANTIALIAS).convert('RGB')
@ -1899,6 +1897,8 @@ def process_file(path, options, logger=None):
fpb = re.compile(options.force_page_break, re.IGNORECASE) if options.force_page_break else \
re.compile('$')
cq = options.chapter_attr.split(',')
if len(cq) < 3:
raise ValueError('The --chapter-attr setting must have 2 commas.')
options.chapter_attr = [re.compile(cq[0], re.IGNORECASE), cq[1],
re.compile(cq[2], re.IGNORECASE)]
options.force_page_break = fpb
@ -1916,7 +1916,7 @@ def process_file(path, options, logger=None):
options.anchor_ids = True
files = options.spine if (options.use_spine and hasattr(options, 'spine')) else [path]
conv = HTMLConverter(book, fonts, options, logger, files)
if options.use_spine and hasattr(options, 'toc'):
if options.use_spine and hasattr(options, 'toc') and options.toc is not None:
conv.create_toc(options.toc)
oname = options.output
if not oname:
@ -1925,7 +1925,8 @@ def process_file(path, options, logger=None):
oname = os.path.join(os.getcwd(), name)
oname = os.path.abspath(os.path.expanduser(oname))
conv.writeto(oname, lrs=options.lrs)
logger.info('Output written to %s', oname)
run_plugins_on_postprocess(oname, 'lrf')
conv.log_info('Output written to %s', oname)
conv.cleanup()
return oname
@ -1944,7 +1945,8 @@ def try_opf(path, options, logger):
return
dirpath = os.path.dirname(os.path.abspath(opf))
opf = OPFReader(open(opf, 'rb'), dirpath)
from calibre.ebooks.metadata.opf2 import OPF as OPF2
opf = OPF2(open(opf, 'rb'), dirpath)
try:
title = opf.title
if title and not getattr(options, 'title', None):
@ -1958,10 +1960,6 @@ def try_opf(path, options, logger):
publisher = opf.publisher
if publisher:
options.publisher = publisher
if not getattr(options, 'category', None):
category = opf.category
if category:
options.category = category
if not getattr(options, 'cover', None) or options.use_metadata_cover:
orig_cover = getattr(options, 'cover', None)
options.cover = None
@ -1975,17 +1973,7 @@ def try_opf(path, options, logger):
PILImage.open(cover)
options.cover = cover
except:
for prefix in opf.possible_cover_prefixes():
if options.cover:
break
for suffix in ['.jpg', '.jpeg', '.gif', '.png', '.bmp']:
cpath = os.path.join(os.path.dirname(path), prefix+suffix)
try:
PILImage.open(cpath)
options.cover = cpath
break
except:
continue
pass
if not getattr(options, 'cover', None) and orig_cover is not None:
options.cover = orig_cover
if getattr(opf, 'spine', False):

View File

@ -6,14 +6,11 @@ __docformat__ = 'restructuredtext en'
'''
Render HTML tables as images.
'''
import os, tempfile, atexit, shutil, time
from PyQt4.Qt import QUrl, QApplication, QSize, \
import os, tempfile, atexit, shutil
from PyQt4.Qt import QUrl, QApplication, QSize, QEventLoop, \
SIGNAL, QPainter, QImage, QObject, Qt
from PyQt4.QtWebKit import QWebPage
from calibre.parallel import ParallelJob
__app = None
class HTMLTableRenderer(QObject):
@ -27,13 +24,15 @@ class HTMLTableRenderer(QObject):
self.app = None
self.width, self.height, self.dpi = width, height, dpi
self.base_dir = base_dir
self.images = []
self.tdir = tempfile.mkdtemp(prefix='calibre_render_table')
self.loop = QEventLoop()
self.page = QWebPage()
self.connect(self.page, SIGNAL('loadFinished(bool)'), self.render_html)
self.page.mainFrame().setTextSizeMultiplier(factor)
self.page.mainFrame().setHtml(html,
QUrl('file:'+os.path.abspath(self.base_dir)))
self.images = []
self.tdir = tempfile.mkdtemp(prefix='calibre_render_table')
def render_html(self, ok):
try:
@ -63,7 +62,7 @@ class HTMLTableRenderer(QObject):
finally:
QApplication.quit()
def render_table(server, soup, table, css, base_dir, width, height, dpi, factor=1.0):
def render_table(soup, table, css, base_dir, width, height, dpi, factor=1.0):
head = ''
for e in soup.findAll(['link', 'style']):
head += unicode(e)+'\n\n'
@ -83,24 +82,13 @@ def render_table(server, soup, table, css, base_dir, width, height, dpi, factor=
</body>
</html>
'''%(head, width-10, style, unicode(table))
job = ParallelJob('render_table', lambda j : j, None,
args=[html, base_dir, width, height, dpi, factor])
server.add_job(job)
while not job.has_run:
time.sleep(2)
if job.exception is not None:
print 'Failed to render table'
print job.exception
print job.traceback
images, tdir = job.result
images, tdir = do_render(html, base_dir, width, height, dpi, factor)
atexit.register(shutil.rmtree, tdir)
return images
def do_render(html, base_dir, width, height, dpi, factor):
app = QApplication.instance()
if app is None:
app = QApplication([])
if QApplication.instance() is None:
QApplication([])
tr = HTMLTableRenderer(html, base_dir, width, height, dpi, factor)
app.exec_()
tr.loop.exec_()
return tr.images, tr.tdir

View File

@ -77,7 +77,7 @@ class LRFDocument(LRFMetaFile):
for obj in self.image_map.values() + self.font_map.values():
open(obj.file, 'wb').write(obj.stream)
def to_xml(self):
def to_xml(self, write_files=True):
bookinfo = u'<BookInformation>\n<Info version="1.1">\n<BookInfo>\n'
bookinfo += u'<Title reading="%s">%s</Title>\n'%(self.metadata.title_reading, self.metadata.title)
bookinfo += u'<Author reading="%s">%s</Author>\n'%(self.metadata.author_reading, self.metadata.author)
@ -89,9 +89,10 @@ class LRFDocument(LRFMetaFile):
bookinfo += u'<FreeText reading="">%s</FreeText>\n</BookInfo>\n<DocInfo>\n'%(self.metadata.free_text,)
th = self.doc_info.thumbnail
if th:
prefix = sanitize_file_name(self.metadata.title)
prefix = sanitize_file_name(self.metadata.title, as_unicode=True)
bookinfo += u'<CThumbnail file="%s" />\n'%(prefix+'_thumbnail.'+self.doc_info.thumbnail_extension,)
open(prefix+'_thumbnail.'+self.doc_info.thumbnail_extension, 'wb').write(th)
if write_files:
open(prefix+'_thumbnail.'+self.doc_info.thumbnail_extension, 'wb').write(th)
bookinfo += u'<Language reading="">%s</Language>\n'%(self.doc_info.language,)
bookinfo += u'<Creator reading="">%s</Creator>\n'%(self.doc_info.creator,)
bookinfo += u'<Producer reading="">%s</Producer>\n'%(self.doc_info.producer,)
@ -127,12 +128,16 @@ class LRFDocument(LRFMetaFile):
objects += unicode(obj)
styles += '</Style>\n'
objects += '</Objects>\n'
self.write_files()
if write_files:
self.write_files()
return '<BBeBXylog version="1.0">\n' + bookinfo + pages + styles + objects + '</BBeBXylog>'
def option_parser():
parser = OptionParser(usage=_('%prog book.lrf\nConvert an LRF file into an LRS (XML UTF-8 encoded) file'))
parser.add_option('--output', '-o', default=None, help=_('Output LRS file'), dest='out')
parser.add_option('--dont-output-resources', default=True, action='store_false',
help=_('Do not save embedded image and font files to disk'),
dest='output_resources')
parser.add_option('--verbose', default=False, action='store_true', dest='verbose')
return parser
@ -154,7 +159,7 @@ def main(args=sys.argv, logger=None):
d = LRFDocument(open(args[1], 'rb'))
d.parse()
logger.info(_('Creating XML...'))
o.write(d.to_xml())
o.write(d.to_xml(write_files=opts.output_resources))
logger.info(_('LRS written to ')+opts.out)
return 0

View File

@ -603,12 +603,18 @@ Show/edit the metadata in an LRF file.\n\n'''),
parser.add_option("--get-thumbnail", action="store_true", \
dest="get_thumbnail", default=False, \
help=_("Extract thumbnail from LRF file"))
parser.add_option('--publisher', default=None, help=_('Set the publisher'))
parser.add_option('--classification', default=None, help=_('Set the book classification'))
parser.add_option('--creator', default=None, help=_('Set the book creator'))
parser.add_option('--producer', default=None, help=_('Set the book producer'))
parser.add_option('--get-cover', action='store_true', default=False,
help=_('Extract cover from LRF file. Note that the LRF format has no defined cover, so we use some heuristics to guess the cover.'))
parser.add_option('--bookid', action='store', type='string', default=None,
dest='book_id', help=_('Set book ID'))
parser.add_option("-p", "--page", action="store", type="string", \
dest="page", help=_("Don't know what this is for"))
# The SumPage element specifies the number of "View"s (visible pages for the BookSetting element conditions) of the content.
# Basically, the total pages per the page size, font size, etc. when the LRF is first created. Since this will change as the book is reflowed, it is probably not worth using.
#parser.add_option("-p", "--page", action="store", type="string", \
# dest="page", help=_("Don't know what this is for"))
return parser
@ -624,6 +630,8 @@ def set_metadata(stream, mi):
lrf.free_text = mi.comments
if mi.author_sort:
lrf.author_reading = mi.author_sort
if mi.publisher:
lrf.publisher = mi.publisher
def main(args=sys.argv):
@ -644,10 +652,16 @@ def main(args=sys.argv):
lrf.author_reading = options.author_reading
if options.author:
lrf.author = options.author
if options.publisher:
lrf.publisher = options.publisher
if options.classification:
lrf.classification = options.classification
if options.category:
lrf.category = options.category
if options.page:
lrf.page = options.page
if options.creator:
lrf.creator = options.creator
if options.producer:
lrf.producer = options.producer
if options.thumbnail:
path = os.path.expanduser(os.path.expandvars(options.thumbnail))
f = open(path, "rb")

View File

@ -2338,6 +2338,9 @@ class Canvas(LrsObject, LrsContainer, LrsAttributes):
container.addLrfObject(c.objId)
lrfWriter.append(c)
def has_text(self):
return bool(self.contents)
class PutObj(LrsContainer):

View File

@ -1,37 +1,6 @@
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
from calibre.ebooks.lrf.web.profiles.nytimes import NYTimes
from calibre.ebooks.lrf.web.profiles.bbc import BBC
from calibre.ebooks.lrf.web.profiles.newsweek import Newsweek
from calibre.ebooks.lrf.web.profiles.economist import Economist
from calibre.ebooks.lrf.web.profiles.newyorkreview import NewYorkReviewOfBooks
from calibre.ebooks.lrf.web.profiles.spiegelde import SpiegelOnline
from calibre.ebooks.lrf.web.profiles.zeitde import ZeitNachrichten
from calibre.ebooks.lrf.web.profiles.faznet import FazNet
from calibre.ebooks.lrf.web.profiles.wsj import WallStreetJournal
from calibre.ebooks.lrf.web.profiles.barrons import Barrons
from calibre.ebooks.lrf.web.profiles.portfolio import Portfolio
from calibre.ebooks.lrf.web.profiles.cnn import CNN
from calibre.ebooks.lrf.web.profiles.chr_mon import ChristianScienceMonitor
from calibre.ebooks.lrf.web.profiles.jpost import JerusalemPost
from calibre.ebooks.lrf.web.profiles.reuters import Reuters
from calibre.ebooks.lrf.web.profiles.atlantic import Atlantic
from calibre.ebooks.lrf.web.profiles.ap import AssociatedPress
from calibre.ebooks.lrf.web.profiles.newyorker import NewYorker
from calibre.ebooks.lrf.web.profiles.jutarnji import Jutarnji
from calibre.ebooks.lrf.web.profiles.usatoday import USAToday
from calibre.ebooks.lrf.web.profiles.upi import UnitedPressInternational
from calibre.ebooks.lrf.web.profiles.wash_post import WashingtonPost
from calibre.ebooks.lrf.web.profiles.nasa import NASA
builtin_profiles = [Atlantic, AssociatedPress, Barrons, BBC,
ChristianScienceMonitor, CNN, Economist, FazNet,
JerusalemPost, Jutarnji, NASA, Newsweek, NewYorker,
NewYorkReviewOfBooks, NYTimes, UnitedPressInternational, USAToday,
Portfolio, Reuters, SpiegelOnline, WallStreetJournal,
WashingtonPost, ZeitNachrichten,
]
available_profiles = [i.__module__.rpartition('.')[2] for i in builtin_profiles]
builtin_profiles = []
available_profiles = [i.__module__.rpartition('.')[2] for i in builtin_profiles]

View File

@ -61,7 +61,7 @@ class NASA(DefaultProfile):
(re.compile(r'<!-- Top Header starts -->.*?<!---->', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
## This removes the "download image" of various sizes from the Image of the day.
(re.compile(r'<div id="download_image_box_print">.*?<div id="caption_region_print">', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
(re.compile(r'(?is)<div id="download_image_box_print">.*?<div id="caption_region_print">'), lambda match : '<New Stuff>'),
]

View File

@ -103,6 +103,8 @@ class OCFDirReader(OCFReader):
return open(os.path.join(self.root, path), *args, **kwargs)
class CoverRenderer(QObject):
WIDTH = 1280
HEIGHT = 1024
def __init__(self, url, size, loop):
QObject.__init__(self)
@ -111,6 +113,9 @@ class CoverRenderer(QObject):
pal = self.page.palette()
pal.setBrush(QPalette.Background, Qt.white)
self.page.setPalette(pal)
self.page.setViewportSize(QSize(600, 800))
self.page.mainFrame().setScrollBarPolicy(Qt.Vertical, Qt.ScrollBarAlwaysOff)
self.page.mainFrame().setScrollBarPolicy(Qt.Horizontal, Qt.ScrollBarAlwaysOff)
QObject.connect(self.page, SIGNAL('loadFinished(bool)'), self.render_html)
self.image_data = None
self.rendered = False
@ -122,7 +127,7 @@ class CoverRenderer(QObject):
if not ok:
return
size = self.page.mainFrame().contentsSize()
width, height = fit_image(size.width(), size.height(), 1280, 1024)[1:]
width, height = fit_image(size.width(), size.height(), self.WIDTH, self.HEIGHT)[1:]
self.page.setViewportSize(QSize(width, height))
image = QImage(self.page.viewportSize(), QImage.Format_ARGB32)
image.setDotsPerMeterX(96*(100/2.54))

View File

@ -10,13 +10,14 @@ Try to read metadata from an HTML file.
import re
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.chardet import xml_to_unicode
def get_metadata(stream):
src = stream.read()
src = xml_to_unicode(stream.read())[0]
# Title
title = None
pat = re.compile(r'<!--.*?TITLE=(?P<q>[\'"])(.+)(?P=q).*?-->', re.DOTALL)
pat = re.compile(r'<!--.*?TITLE=(?P<q>[\'"])(.+?)(?P=q).*?-->', re.DOTALL)
match = pat.search(src)
if match:
title = match.group(2)
@ -28,7 +29,7 @@ def get_metadata(stream):
# Author
author = None
pat = re.compile(r'<!--.*?AUTHOR=(?P<q>[\'"])(.+)(?P=q).*?-->', re.DOTALL)
pat = re.compile(r'<!--.*?AUTHOR=(?P<q>[\'"])(.+?)(?P=q).*?-->', re.DOTALL)
match = pat.search(src)
if match:
author = match.group(2).replace(',', ';')
@ -36,7 +37,7 @@ def get_metadata(stream):
mi = MetaInformation(title, [author] if author else None)
# Publisher
pat = re.compile(r'<!--.*?PUBLISHER=(?P<q>[\'"])(.+)(?P=q).*?-->', re.DOTALL)
pat = re.compile(r'<!--.*?PUBLISHER=(?P<q>[\'"])(.+?)(?P=q).*?-->', re.DOTALL)
match = pat.search(src)
if match:
mi.publisher = match.group(2)

View File

@ -0,0 +1,89 @@
#!/usr/bin/env python
__license__ = 'GPL v3'
__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
__docformat__ = 'restructuredtext en'
'''
Read metadata from LRX files
'''
import sys, struct
from zlib import decompress
from lxml import etree
from calibre.ebooks.metadata import MetaInformation, string_to_authors
def _read(f, at, amount):
f.seek(at)
return f.read(amount)
def word_be(buf):
return struct.unpack('>L', buf)[0]
def word_le(buf):
return struct.unpack('<L', buf)[0]
def short_le(buf):
return struct.unpack('<H', buf)[0]
def short_be(buf):
return struct.unpack('>H', buf)[0]
def get_metadata(f):
read = lambda at, amount: _read(f, at, amount)
f.seek(0)
buf = f.read(12)
if buf[4:] == 'ftypLRX2':
offset = 0
while True:
offset += word_be(buf[:4])
try:
buf = read(offset, 8)
except:
raise ValueError('Not a valid LRX file')
if buf[4:] == 'bbeb':
break
offset += 8
buf = read(offset, 16)
if buf[:8].decode('utf-16-le') != 'LRF\x00':
raise ValueError('Not a valid LRX file')
lrf_version = word_le(buf[8:12])
offset += 0x4c
compressed_size = short_le(read(offset, 2))
offset += 2
if lrf_version >= 800:
offset += 6
compressed_size -= 4
uncompressed_size = word_le(read(offset, 4))
info = decompress(f.read(compressed_size))
if len(info) != uncompressed_size:
raise ValueError('LRX file has malformed metadata section')
root = etree.fromstring(info)
bi = root.find('BookInfo')
title = bi.find('Title')
title_sort = title.get('reading', None)
title = title.text
author = bi.find('Author')
author_sort = author.get('reading', None)
mi = MetaInformation(title, string_to_authors(author.text))
mi.title_sort, mi.author_sort = title_sort, author_sort
author = author.text
publisher = bi.find('Publisher')
mi.publisher = getattr(publisher, 'text', None)
mi.tags = [x.text for x in bi.findall('Category')]
mi.language = root.find('DocInfo').find('Language').text
return mi
elif buf[4:8] == 'LRX':
raise ValueError('Librie LRX format not supported')
else:
raise ValueError('Not a LRX file')
def main(args=sys.argv):
print get_metadata(open(args[1], 'rb'))
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@ -5,36 +5,17 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
import os, re, collections
from calibre.utils.config import prefs
from calibre.ebooks.metadata.rtf import get_metadata as rtf_metadata
from calibre.ebooks.metadata.fb2 import get_metadata as fb2_metadata
from calibre.ebooks.lrf.meta import get_metadata as lrf_metadata
from calibre.ebooks.metadata.pdf import get_metadata as pdf_metadata
from calibre.ebooks.metadata.lit import get_metadata as lit_metadata
from calibre.ebooks.metadata.imp import get_metadata as imp_metadata
from calibre.ebooks.metadata.rb import get_metadata as rb_metadata
from calibre.ebooks.metadata.epub import get_metadata as epub_metadata
from calibre.ebooks.metadata.html import get_metadata as html_metadata
from calibre.ebooks.mobi.reader import get_metadata as mobi_metadata
from calibre.ebooks.metadata.odt import get_metadata as odt_metadata
from calibre.ebooks.metadata.opf2 import OPF
from calibre.ebooks.metadata.rtf import set_metadata as set_rtf_metadata
from calibre.ebooks.lrf.meta import set_metadata as set_lrf_metadata
from calibre.ebooks.metadata.epub import set_metadata as set_epub_metadata
from calibre.ebooks.metadata.pdf import set_metadata as set_pdf_metadata
try:
from calibre.libunrar import extract_member as rar_extract_first
except OSError:
rar_extract_first = None
from calibre.libunzip import extract_member as zip_extract_first
from calibre.customize.ui import get_file_type_metadata, set_file_type_metadata
from calibre.ebooks.metadata import MetaInformation
from calibre.ptempfile import TemporaryDirectory
_METADATA_PRIORITIES = [
'html', 'htm', 'xhtml', 'xhtm',
'rtf', 'fb2', 'pdf', 'prc', 'odt',
'epub', 'lit', 'lrf', 'mobi', 'rb', 'imp'
'epub', 'lit', 'lrx', 'lrf', 'mobi',
'rb', 'imp'
]
# The priorities for loading metadata from different file types
@ -87,11 +68,7 @@ def get_metadata(stream, stream_type='lrf', use_libprs_metadata=False):
mi = MetaInformation(None, None)
if prefs['read_file_metadata']:
try:
func = eval(stream_type + '_metadata')
mi = func(stream)
except NameError:
pass
mi = get_file_type_metadata(stream, stream_type)
name = os.path.basename(getattr(stream, 'name', ''))
base = metadata_from_filename(name)
@ -103,37 +80,14 @@ def get_metadata(stream, stream_type='lrf', use_libprs_metadata=False):
if opf is not None:
base.smart_update(opf)
if stream_type in ('cbr', 'cbz'):
try:
cdata = get_comic_cover(stream, stream_type)
if cdata is not None:
base.cover_data = cdata
except:
import traceback
traceback.print_exc()
pass
return base
def get_comic_cover(stream, type):
extract_first = zip_extract_first if type.lower() == 'cbz' else rar_extract_first
ret = extract_first(stream)
if ret is not None:
path, data = ret
ext = os.path.splitext(path)[1][1:]
return (ext.lower(), data)
def set_metadata(stream, mi, stream_type='lrf'):
if stream_type: stream_type = stream_type.lower()
if stream_type == 'lrf':
set_lrf_metadata(stream, mi)
elif stream_type == 'epub':
set_epub_metadata(stream, mi)
elif stream_type == 'rtf':
set_rtf_metadata(stream, mi)
#elif stream_type == 'pdf':
# set_pdf_metadata(stream, mi)
if stream_type:
stream_type = stream_type.lower()
set_file_type_metadata(stream, mi, stream_type)
def metadata_from_filename(name, pat=None):
name = os.path.splitext(name)[0]
mi = MetaInformation(None, None)

View File

@ -70,7 +70,7 @@ class Manifest(ResourceCollection):
@staticmethod
def from_opf_manifest_element(manifest, dir):
m = Manifest()
for item in manifest.findAll('item'):
for item in manifest.findAll(re.compile('item')):
try:
m.append(ManifestItem.from_opf_manifest_item(item, dir))
id = item.get('id', '')
@ -130,7 +130,7 @@ class Spine(ResourceCollection):
@staticmethod
def from_opf_spine_element(spine, manifest):
s = Spine(manifest)
for itemref in spine.findAll('itemref'):
for itemref in spine.findAll(re.compile('itemref')):
if itemref.has_key('idref'):
r = Spine.Item(s.manifest.id_for_path,
s.manifest.path_for_id(itemref['idref']), is_path=True)
@ -216,7 +216,7 @@ class standard_field(object):
def __get__(self, obj, typ=None):
return getattr(obj, 'get_'+self.name)()
class OPF(MetaInformation):
MIMETYPE = 'application/oebps-package+xml'
@ -242,14 +242,27 @@ class OPF(MetaInformation):
def __init__(self):
raise NotImplementedError('Abstract base class')
@apply
def package():
def fget(self):
return self.soup.find(re.compile('package'))
return property(fget=fget)
@apply
def metadata():
def fget(self):
return self.package.find(re.compile('metadata'))
return property(fget=fget)
def get_title(self):
title = self.soup.package.metadata.find('dc:title')
title = self.metadata.find('dc:title')
if title and title.string:
return self.ENTITY_PATTERN.sub(entity_to_unicode, title.string).strip()
return self.default_title.strip()
def get_authors(self):
creators = self.soup.package.metadata.findAll('dc:creator')
creators = self.metadata.findAll('dc:creator')
for elem in creators:
role = elem.get('role')
if not role:
@ -266,7 +279,7 @@ class OPF(MetaInformation):
return []
def get_author_sort(self):
creators = self.soup.package.metadata.findAll('dc:creator')
creators = self.metadata.findAll('dc:creator')
for elem in creators:
role = elem.get('role')
if not role:
@ -277,7 +290,7 @@ class OPF(MetaInformation):
return None
def get_title_sort(self):
title = self.soup.package.find('dc:title')
title = self.package.find('dc:title')
if title:
if title.has_key('file-as'):
return title['file-as'].strip()
@ -290,7 +303,7 @@ class OPF(MetaInformation):
return None
def get_uid(self):
package = self.soup.find('package')
package = self.package
if package.has_key('unique-identifier'):
return package['unique-identifier']
@ -307,7 +320,7 @@ class OPF(MetaInformation):
return None
def get_isbn(self):
for item in self.soup.package.metadata.findAll('dc:identifier'):
for item in self.metadata.findAll('dc:identifier'):
scheme = item.get('scheme')
if not scheme:
scheme = item.get('opf:scheme')
@ -316,13 +329,13 @@ class OPF(MetaInformation):
return None
def get_language(self):
item = self.soup.package.metadata.find('dc:language')
item = self.metadata.find('dc:language')
if not item:
return _('Unknown')
return ''.join(item.findAll(text=True)).strip()
def get_application_id(self):
for item in self.soup.package.metadata.findAll('dc:identifier'):
for item in self.metadata.findAll('dc:identifier'):
scheme = item.get('scheme', None)
if scheme is None:
scheme = item.get('opf:scheme', None)
@ -342,7 +355,7 @@ class OPF(MetaInformation):
def possible_cover_prefixes(self):
isbn, ans = [], []
for item in self.soup.package.metadata.findAll('dc:identifier'):
for item in self.metadata.findAll('dc:identifier'):
scheme = item.get('scheme')
if not scheme:
scheme = item.get('opf:scheme')
@ -352,13 +365,13 @@ class OPF(MetaInformation):
return ans
def get_series(self):
s = self.soup.package.metadata.find('series')
s = self.metadata.find('series')
if s is not None:
return str(s.string).strip()
return None
def get_series_index(self):
s = self.soup.package.metadata.find('series-index')
s = self.metadata.find('series-index')
if s and s.string:
try:
return int(str(s.string).strip())
@ -367,7 +380,7 @@ class OPF(MetaInformation):
return None
def get_rating(self):
s = self.soup.package.metadata.find('rating')
s = self.metadata.find('rating')
if s and s.string:
try:
return int(str(s.string).strip())
@ -400,17 +413,17 @@ class OPFReader(OPF):
if manage:
stream.close()
self.manifest = Manifest()
m = self.soup.find('manifest')
m = self.soup.find(re.compile('manifest'))
if m is not None:
self.manifest = Manifest.from_opf_manifest_element(m, dir)
self.spine = None
spine = self.soup.find('spine')
spine = self.soup.find(re.compile('spine'))
if spine is not None:
self.spine = Spine.from_opf_spine_element(spine, self.manifest)
self.toc = TOC(base_path=dir)
self.toc.read_from_opf(self)
guide = self.soup.find('guide')
guide = self.soup.find(re.compile('guide'))
if guide is not None:
self.guide = Guide.from_opf_guide(guide, dir)
self.base_dir = dir

View File

@ -30,7 +30,6 @@ class Resource(object):
:member:`path`
:member:`mime_type`
:method:`href`
'''
def __init__(self, href_or_path, basedir=os.getcwd(), is_path=True):
@ -419,7 +418,8 @@ class OPF(object):
tags_path = XPath('descendant::*[re:match(name(), "subject", "i")]')
isbn_path = XPath('descendant::*[re:match(name(), "identifier", "i") and '+
'(re:match(@scheme, "isbn", "i") or re:match(@opf:scheme, "isbn", "i"))]')
application_id_path= XPath('descendant::*[re:match(name(), "identifier", "i") and '+
identifier_path = XPath('descendant::*[re:match(name(), "identifier", "i")]')
application_id_path = XPath('descendant::*[re:match(name(), "identifier", "i") and '+
'(re:match(@opf:scheme, "calibre|libprs500", "i") or re:match(@scheme, "calibre|libprs500", "i"))]')
manifest_path = XPath('descendant::*[re:match(name(), "manifest", "i")]/*[re:match(name(), "item", "i")]')
manifest_ppath = XPath('descendant::*[re:match(name(), "manifest", "i")]')
@ -720,6 +720,27 @@ class OPF(object):
return property(fget=fget, fset=fset)
def guess_cover(self):
'''
Try to guess a cover. Needed for some old/badly formed OPF files.
'''
if self.base_dir and os.path.exists(self.base_dir):
for item in self.identifier_path(self.metadata):
scheme = None
for key in item.attrib.keys():
if key.endswith('scheme'):
scheme = item.get(key)
break
if scheme is None:
continue
if item.text:
prefix = item.text.replace('-', '')
for suffix in ['.jpg', '.jpeg', '.gif', '.png', '.bmp']:
cpath = os.access(os.path.join(self.base_dir, prefix+suffix), os.R_OK)
if os.access(os.path.join(self.base_dir, prefix+suffix), os.R_OK):
return cpath
@apply
def cover():
@ -729,6 +750,10 @@ class OPF(object):
for item in self.guide:
if item.type.lower() == t:
return item.path
try:
return self.guess_cover()
except:
pass
def fset(self, path):
if self.guide is not None:

Some files were not shown because too many files have changed in this diff Show More