Added rar,zip support to html2lrf and squashed a few more html processing bugs. Version bump.

Makefile

@@ -1,14 +0,0 @@
-APIDOCS=/var/www/libprs500.kovidgoyal.net/htdocs/apidocs
-targets:
-	@echo Targets are: installer doc
-
-installer:
-	@scp dist/libprs500-*.exe castalia:/var/www/vhosts/kovidgoyal.net/subdomains/libprs500/httpdocs/downloads/
-	@ssh castalia chmod a+r /var/www/vhosts/kovidgoyal.net/subdomains/libprs500/httpdocs/downloads/\*
-	@echo Update link on the libprs500 wiki
-
-doc:
-	epydoc --config epydoc.conf
-	cp -r docs/html ${APIDOCS}/
-	epydoc -v --config epydoc-pdf.conf
-	cp docs/pdf/api.pdf ${APIDOCS}/

installer.nsi

@@ -15,10 +15,11 @@ Var MUI_TEMP
 
 !define PRODUCT_NAME "libprs500"
 !define XPUI_BRANDINGTEXT "${PRODUCT_NAME} created by Kovid Goyal"
-!define PRODUCT_VERSION "0.3.12"
+!define PRODUCT_VERSION "0.3.13"
 !define WEBSITE "https://libprs500.kovidgoyal.net"
 !define PY2EXE_DIR "C:\libprs500"
 !define LIBUSB_DIR "C:\libusb-prs500"
+!define LIBUNRAR_DIR "C:\Program Files\UnrarDLL"
 !define QT_DIR     "C:\Qt\4.2.3\bin"
 
 ;------------------------------------------------------------------------------------------------------
@@ -98,6 +99,7 @@ Section "libprs500" Seclibprs500
   
   SetOutPath "$SYSDIR"
   File "${LIBUSB_DIR}\libusb0.dll"
+  File "${LIBUNRAR_DIR}\unrar.dll"
   DetailPrint " "
   
   DetailPrint "Installing USB driver (this may take a few seconds) ..."

setup.py

@@ -21,6 +21,8 @@ import ez_setup
 ez_setup.use_setuptools()
 from setuptools import setup, find_packages
 
+
+################################# py2exe #######################################
 py2exe_options = {}
 if sys.argv[1] == 'py2exe':
     py2exe_dir = 'C:\libprs500'
@@ -33,10 +35,13 @@ if sys.argv[1] == 'py2exe':
     f.close()
     try:
         import py2exe
-        console = [{
-                    'script' : 'src/libprs500/cli/main.py', 'dest_base':'prs500',
-                    'script' : 'src/libprs500/lrf/html/convert_from.py', 'dest_base':'html2lrf'
-                  }]
+        console = [
+                    {'script' : 'src/libprs500/cli/main.py', 'dest_base':'prs500'},
+                    {'script' : 'src/libprs500/lrf/html/convert_from.py', 'dest_base':'html2lrf'},
+                    {'script' : 'src/libprs500/lrf/txt/convert_from.py', 'dest_base':'txt2lrf'},
+                    {'script' : 'src/libprs500/lrf/meta.py', 'dest_base':'lrf-meta'},
+                    {'script' : 'src/libprs500/metadata/rtf.py', 'dest_base':'rtf-meta'},
+                  ]
         windows = [{'script' : 'src/libprs500/gui/main.py', 'dest_base':'prs500-gui',
                     'icon_resources':[(1,'icons/library.ico')]}]
         excludes = ["Tkconstants", "Tkinter", "tcl", "_imagingtk", 
@@ -50,29 +55,9 @@ if sys.argv[1] == 'py2exe':
     except ImportError:
         print >>sys.stderr, 'Must be in Windows to run py2exe'
         sys.exit(1)
-
+################################################################################
     
 
-# Try to install the Python imaging library as the package name (PIL) doesn't 
-# match the distribution file name, thus declaring itas a dependency is useless
-from setuptools.command.easy_install import main as easy_install
-try:
-  try:
-    import Image
-  except ImportError: 
-    if sys.platform.lower()[:5] not in ['win32', 'darwin']:
-        print "Trying to install the Python Imaging Library"
-        easy_install(["-f", "http://www.pythonware.com/products/pil/", "Imaging"])
-    else:
-        raise Exception('Please install the Python Imaging library manually from '\
-                        'http://www.pythonware.com/products/pil/')
-except Exception, e:
-    print >> sys.stderr, e 
-    print >> sys.stderr, \
-          "WARNING: Could not install the Python Imaging Library.", \
-          "Some functionality will be unavailable"
-
-
 if sys.hexversion < 0x2050000:
     print >> sys.stderr, "You must use python >= 2.5 Try invoking this script as python2.5 setup.py."
     print >> sys.stderr, "If you are using easy_install, try easy_install-2.5"
@@ -110,20 +95,17 @@ setup(
                   """, 
       long_description = 
       """
-      libprs500 is library to interface with the 
-      `SONY Portable Reader`_ over USB_. 
-      It provides methods to list the contents of the file system on the device,
-      as well as copy files from and to the device. 
-      It also provides a command line and a graphical user interface via 
-      the commands prs500 and
-      prs500-gui. The graphical user interface is designed to 
-      manage an ebook library and allows for easy 
-      syncing between the library and the ebook reader. 
-      In addition libprs500 has a utility to read/write the metadata 
-      from LRF files (unencrypted books in the SONY BBeB format). A command line
-      interface to this is provided via the command lrf-meta.
+      libprs500 is a ebook management application. It maintains an ebook library
+      and allows for easy transfer of books from the library to an ebook reader.
+      At the moment, it supports the `SONY Portable Reader`_.
       
-      A windows installer is available from https://libprs500.kovidgoyal.net 
+      It can also convert various popular ebook formats into LRF, the native
+      ebook format of the SONY Reader.
+      
+      For screenshots: https://libprs500.kovidgoyal.net/wiki/Screenshots
+      
+      For installation/usage instructions please see 
+      https://libprs500.kovidgoyal.net/wiki/WikiStart#Installation
       
       For SVN access: svn co https://svn.kovidgoyal.net/code/libprs500
       

src/libprs500/__init__.py

@@ -33,10 +33,24 @@ You may have to adjust the GROUP and the location of the rules file to
 suit your distribution.
 """
 
-__version__   = "0.3.12"
+__version__   = "0.3.13"
 __docformat__ = "epytext"
 __author__    = "Kovid Goyal <kovid@kovidgoyal.net>"
 
 import sys
 iswindows = 'win32' in sys.platform.lower()
 isosx     = 'darwin' in sys.platform.lower()
+
+def extract(path, dir):
+    import os
+    ext = os.path.splitext(path)[1][1:].lower()
+    extractor = None
+    if ext == 'zip':
+        from libprs500.libunzip import extract
+        extractor = extract
+    elif ext == 'rar':
+        from libprs500.libunrar import extract
+        extractor = extract
+    if not extract:
+        raise Exception('Unknown archive type')
+    extractor(path, dir)

src/libprs500/libunrar.py

@@ -17,6 +17,7 @@ This module provides a thin ctypes based wrapper around libunrar.
 
 See  ftp://ftp.rarlabs.com/rar/unrarsrc-3.7.5.tar.gz
 """
+import os
 from ctypes import Structure, c_char_p, c_uint, c_void_p, POINTER, \
                     byref, c_wchar_p, CFUNCTYPE, c_int, c_long, c_char, c_wchar
 from StringIO import StringIO
@@ -164,14 +165,21 @@ def get_archive_info(flags):
     print >>ios, 'First Volume:\t', 'yes' if (flags & 256) else 'no or older than 3.0'
     return ios.getvalue()
 
-def extract(path):    
+def extract(path, dir):
+    """
+    Extract archive C{filename} into directory C{dir}
+    """    
     open_archive_data = RAROpenArchiveDataEx(ArcName=path, OpenMode=RAR_OM_EXTRACT, CmtBuf=None)
     arc_data = _libunrar.RAROpenArchiveEx(byref(open_archive_data))
+    cwd = os.getcwd()
+    if not os.path.isdir( dir ):
+        os.mkdir( dir )
+    os.chdir( dir )
     try:
         if open_archive_data.OpenResult != 0:
             raise UnRARException(_interpret_open_error(open_archive_data.OpenResult, path))
         print 'Archive:', path
-        print get_archive_info(open_archive_data.Flags)        
+        #print get_archive_info(open_archive_data.Flags)        
         header_data = RARHeaderDataEx(CmtBuf=None)
         #_libunrar.RARSetCallback(arc_data, callback_func, mode)
         while True:
@@ -184,7 +192,5 @@ def extract(path):
         if RHCode == ERAR_BAD_DATA:
             raise UnRARException('File header broken')
     finally:
+        os.chdir(cwd)
         _libunrar.RARCloseArchive(arc_data)
-    
-extract(r'z:\home\test.rar')
-#extract('/home/kovid/ero/Fansadox Collections/C21 Ponygirl Inferno.rar')

src/libprs500/libunzip.py

@@ -0,0 +1,52 @@
+##    Copyright (C) 2006 Kovid Goyal kovid@kovidgoyal.net
+##    This program is free software; you can redistribute it and/or modify
+##    it under the terms of the GNU General Public License as published by
+##    the Free Software Foundation; either version 2 of the License, or
+##    (at your option) any later version.
+##
+##    This program is distributed in the hope that it will be useful,
+##    but WITHOUT ANY WARRANTY; without even the implied warranty of
+##    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+##    GNU General Public License for more details.
+##
+##    You should have received a copy of the GNU General Public License along
+##    with this program; if not, write to the Free Software Foundation, Inc.,
+##    51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+import os, zipfile
+from cStringIO import StringIO
+
+def extract(filename, dir):
+    """
+    Extract archive C{filename} into directory C{dir}
+    """
+    zf = zipfile.ZipFile( filename )
+    namelist = zf.namelist()
+    dirlist = filter( lambda x: x.endswith( '/' ), namelist )
+    filelist = filter( lambda x: not x.endswith( '/' ), namelist )
+    # make base
+    pushd = os.getcwd()
+    if not os.path.isdir( dir ):
+        os.mkdir( dir )
+    os.chdir( dir )
+    # create directory structure
+    dirlist.sort()
+    for dirs in dirlist:
+        dirs = dirs.split( '/' )
+        prefix = ''
+        for dir in dirs:
+            dirname = os.path.join( prefix, dir )
+            if dir and not os.path.isdir( dirname ):
+                os.mkdir( dirname )
+            prefix = dirname
+    # extract files
+    for fn in filelist:
+        out = open( fn, 'wb' )
+        buffer = StringIO( zf.read( fn ))
+        buflen = 2 ** 20
+        datum = buffer.read( buflen )
+        while datum:
+            out.write( datum )
+            datum = buffer.read( buflen )
+        out.close()
+    os.chdir( pushd )

src/libprs500/lrf/__init__.py

@@ -20,6 +20,7 @@ At the time fo writing, this package only supports reading and writing LRF meat
 from optparse import OptionParser
 
 from libprs500.lrf.pylrs.pylrs import Book as _Book
+from libprs500 import __version__ as VERSION
 
 __docformat__ = "epytext"
 __author__    = "Kovid Goyal <kovid@kovidgoyal.net>"
@@ -28,7 +29,7 @@ class ConversionError(Exception):
     pass
 
 def option_parser(usage):
-    parser = OptionParser(usage=usage)
+    parser = OptionParser(usage=usage, version='libprs500 '+VERSION)
     parser.add_option("-t", "--title", action="store", type="string", \
                     dest="title", help="Set the title")
     parser.add_option("-a", "--author", action="store", type="string", \

src/libprs500/lrf/html/convert_from.py

@@ -21,16 +21,20 @@ Code to convert HTML ebooks into LRF ebooks.
 I am indebted to esperanc for the CSS->Xylog Style conversion routines
 and to Falstaff for pylrs.
 """
-import os, re, sys
+import os, re, sys, shutil
 from htmlentitydefs import name2codepoint
 from urllib import urlopen
 from urlparse import urlparse
+from tempfile import mkdtemp
+from operator import itemgetter
 
-from libprs500.lrf.html.BeautifulSoup import BeautifulSoup, Comment, Tag, NavigableString
+from libprs500.lrf.html.BeautifulSoup import BeautifulSoup, Comment, Tag, \
+                                             NavigableString, Declaration
 from libprs500.lrf.pylrs.pylrs import Paragraph, CR, Italic, ImageStream, TextBlock, \
                                       ImageBlock, JumpButton, CharButton, Page
 from libprs500.lrf.pylrs.pylrs import Span as _Span
 from libprs500.lrf import ConversionError, option_parser, Book
+from libprs500 import extract
 
 def ImagePage():
     return Page(evensidemargin=0, oddsidemargin=0, topmargin=0, \
@@ -150,7 +154,7 @@ class Span(_Span):
             elif key == 'font-weight':
                 ans = font_weight(val)
                 if ans:
-                    t['fontweight'] = val            
+                    t['fontweight'] = ans
             elif key.startswith("margin"):
                 if key == "margin":
                     u = []
@@ -186,8 +190,6 @@ class Span(_Span):
                     t["align"] = "head"
             else:
                 print >>sys.stderr, 'Unhandled/malformed CSS key:', key, d[key]
-        if 'small' in t.values():
-            print d, 'font-size' in d.keys()
         return t        
     
     def __init__(self, ns, css, font_delta=0):
@@ -252,7 +254,7 @@ class HTMLConverter(object):
         self.current_page = None
         self.current_para = None
         self.current_style = {}        
-        self.parse_file(self.soup.html)
+        self.parse_file()
         HTMLConverter.processed_files[path] = self
         print 'done'
         
@@ -318,14 +320,15 @@ class HTMLConverter(object):
             prop.update(self.parse_style_properties(tag["style"]))    
         return prop
         
-    def parse_file(self, html):        
+    def parse_file(self):        
         self.current_page = Page()
         self.current_block = TextBlock()
         self.current_para = Paragraph()
         if self.cover:
             self.add_image_block(self.cover)
-        self.top = self.current_block        
-        self.parse_tag(html, {})
+        self.top = self.current_block
+        
+        self.process_children(self.soup, {})
         if self.current_para:
             self.current_block.append(self.current_para)
         if self.current_block:
@@ -409,12 +412,12 @@ class HTMLConverter(object):
         """
         if self.current_para.contents:
             self.current_block.append(self.current_para)
+            self.current_para = Paragraph()
         if self.current_block.contents:
             self.current_page.append(self.current_block)
+            self.current_block = TextBlock()
         if self.current_page.contents: 
             self.book.append(self.current_page)
-            self.current_para = Paragraph()            
-            self.current_block = TextBlock()
             self.current_page = Page()
         
         
@@ -423,40 +426,42 @@ class HTMLConverter(object):
             self.end_page()
             page = ImagePage()
             if not self.images.has_key(path):
-                self.images[path] = ImageBlock(ImageStream(path))
-            page.append(self.images[path])
+                self.images[path] = ImageStream(path)
+            page.append(ImageBlock(self.images[path]))
             self.book.append(page)
     
-    def parse_tag(self, tag, parent_css):
-        def sanctify_css(css):
-            """ Make css safe for use in a SPAM Xylog tag """
-            for key in css.keys():
-                test = key.lower()
-                if test.startswith('margin') or 'indent' in test or \
-                   'padding' in test or 'border' in test or 'page-break' in test \
-                   or test.startswith('mso') \
-                   or test in ['color', 'display', 'text-decoration', \
-                               'letter-spacing', 'text-autospace', 'text-transform']:
-                    css.pop(key)
-            return css
+    def process_children(self, ptag, pcss):
+        """ Process the children of ptag """
+        for c in ptag.contents:
+            if isinstance(c, (Comment, Declaration)):
+                continue
+            elif isinstance(c, Tag):
+                self.parse_tag(c, pcss)
+            elif isinstance(c, NavigableString):
+                self.add_text(c, pcss)
                     
-        def add_text(tag, css):
-            try:
-                self.current_para.append(Span(tag, sanctify_css(css), \
-                                              font_delta=self.font_delta))
-            except ConversionError, err:
-                if self.verbose:
-                    print >>sys.stderr, err
+    def add_text(self, tag, css):
+        try:
+            self.current_para.append(Span(tag, self.sanctify_css(css), \
+                                          font_delta=self.font_delta))
+        except ConversionError, err:
+            if self.verbose:
+                print >>sys.stderr, err
         
-        def process_children(ptag, pcss):
-            """ Process the children of ptag """
-            for c in ptag.contents:
-                if isinstance(c, Comment):
-                    continue
-                elif isinstance(c, Tag):
-                    self.parse_tag(c, pcss)
-                elif isinstance(c, NavigableString):
-                    add_text(c, pcss)
+    def sanctify_css(self, css):
+        """ Make css safe for use in a SPAM Xylog tag """
+        for key in css.keys():
+            test = key.lower()
+            if test.startswith('margin') or 'indent' in test or \
+               'padding' in test or 'border' in test or 'page-break' in test \
+               or test.startswith('mso') \
+               or test in ['color', 'display', 'text-decoration', \
+                           'letter-spacing', 'text-autospace', 'text-transform']:
+                css.pop(key)
+        return css
+    
+    
+    def parse_tag(self, tag, parent_css):
                     
         def process_text_tag(tag, tag_css):
             if 'page-break-before' in tag_css.keys():
@@ -467,16 +472,14 @@ class HTMLConverter(object):
             if 'page-break-after' in tag_css.keys():
                 end_page = True
                 tag_css.pop('page-break-after')
-            process_children(tag, tag_css)
+            self.process_children(tag, tag_css)
             if end_page:
                 self.end_page()
-                
-        
             
         try:
             tagname = tag.name.lower()
         except AttributeError:
-            add_text(tag, parent_css)
+            self.add_text(tag, parent_css)
             return
         tag_css = self.tag_css(tag, parent_css=parent_css)
         try: # Skip element if its display attribute is set to none
@@ -494,7 +497,7 @@ class HTMLConverter(object):
                 self.current_block = tb
                 self.current_para = Paragraph()
                 self.targets[tag['name']] = tb
-                process_children(tag, tag_css)
+                self.process_children(tag, tag_css)
             elif tag.has_key('href'):
                 purl = urlparse(tag['href'])
                 path = purl[2]
@@ -516,12 +519,15 @@ class HTMLConverter(object):
             elif tag.has_key('type') and tag['type'] == "text/css" \
                     and tag.has_key('href'):
                 url = tag['href']
-                if url.startswith('http://'):
-                    f = urlopen(url)
-                else:
-                    f = open(url)
-                self.parse_css(f.read())
-                f.close()
+                try:
+                    if url.startswith('http://'):
+                        f = urlopen(url)
+                    else:
+                        f = open(url)
+                    self.parse_css(f.read())
+                    f.close()
+                except IOError:
+                    pass
         elif tagname in ['p', 'div', 'ul', 'ol', 'tr', 'li', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6']:
             # TODO: Implement ol
             indent = tag_css.pop('text-indent', '')
@@ -536,34 +542,33 @@ class HTMLConverter(object):
         elif tagname in ['b', 'strong', 'i', 'em', 'span']:
             process_text_tag(tag, tag_css)
         elif tagname == 'font':
-            pass
-        elif tagname == 'link':
-            pass
-        elif tagname == 'style':
-            pass
+            if tag.has_key('face'):
+                tag_css['font-family'] = tag['face']
+            process_text_tag(tag, tag_css)
         elif tagname == 'br':
             self.current_para.append(CR())
         elif tagname == 'hr':
             self.current_para.append(CR())
             # TODO: Horizontal line?
         else:            
-            process_children(tag, tag_css)            
+            self.process_children(tag, tag_css)            
                     
     def writeto(self, path, lrs=False):
         self.book.renderLrs(path) if lrs else self.book.renderLrf(path)
         
 
 def process_file(path, options):
-    cwd = os.getcwd()    
+    cwd = os.getcwd()
+    dirpath = None
     try:
-        path = os.path.abspath(path)
+        dirpath, path = get_path(path)
         cpath, tpath = options.cover, ''
         if options.cover and os.access(options.cover, os.R_OK):            
             try:
                 from PIL import Image
                 from libprs500.prs500 import PRS500
                 from libprs500.ptempfile import PersistentTemporaryFile
-                im = Image.open(cpath)
+                im = Image.open(os.path.join(cwd, cpath))
                 cim = im.resize((600, 800), Image.BICUBIC)
                 cf = PersistentTemporaryFile(prefix="html2lrf_", suffix=".jpg")
                 cf.close()                
@@ -596,6 +601,8 @@ def process_file(path, options):
         print 'Output written to', oname
     finally:
         os.chdir(cwd)
+        if dirpath:
+            shutil.rmtree(dirpath, True)
         
 def main():
     """ CLI for html -> lrf conversions """
@@ -618,6 +625,67 @@ def main():
         options.title = os.path.splitext(os.path.basename(src))[0]
     process_file(src, options)
 
+def console_query(dirpath, candidate, docs):
+    if len(docs) == 1:
+        return 0
+    try:
+        import readline
+    except ImportError:
+        pass
+    i = 0
+    for doc in docs:
+        prefix = '>' if i == candidate else ''
+        print prefix+str(i)+'.\t', doc[0]
+        i += 1
+    print
+    while True:
+        try:
+            choice = raw_input('Choose file to convert (0-'+str(i-1) + \
+                               '). Current choice is ['+ str(candidate) + ']:')
+            if not choice:
+                return candidate
+            choice = int(choice)
+            if choice < 0 or choice >= i:
+                continue
+            candidate = choice
+        except EOFError, KeyboardInterrupt:
+            sys.exit()
+        except:
+            continue
+        break
+    return candidate
+        
+
+def get_path(path, query=console_query):
+    path = os.path.abspath(path)
+    ext = os.path.splitext(path)[1][1:].lower()
+    if ext in ['htm', 'html', 'xhtml']:
+        return None, path
+    dirpath = mkdtemp('','html2lrf')
+    extract(path, dirpath)
+    candidate, docs = None, []
+    for root, dirs, files in os.walk(dirpath):
+        for name in files:
+            ext = os.path.splitext(name)[1][1:].lower()
+            if ext not in ['html', 'xhtml', 'htm', 'xhtm']:
+                continue
+            docs.append((name, root, os.stat(os.path.join(root, name)).st_size))
+            if 'toc' in name.lower():
+                candidate = name
+    docs.sort(key=itemgetter(2))
+    if candidate:
+        for i in range(len(docs)):
+            if docs[i][0] == candidate:
+                candidate = i
+                break
+    else:
+        candidate = len(docs) - 1
+    if len(docs) == 0:
+        raise ConversionError('No suitable files found in archive')
+    if len(docs) > 0:
+        candidate = query(dirpath, candidate, docs)
+    return dirpath, os.path.join(docs[candidate][1], docs[candidate][0])
+
 
 if __name__ == '__main__':
     main()

src/libprs500/lrf/makelrf.py

@@ -1,123 +0,0 @@
-##    Copyright (C) 2006 Kovid Goyal kovid@kovidgoyal.net
-##    This program is free software; you can redistribute it and/or modify
-##    it under the terms of the GNU General Public License as published by
-##    the Free Software Foundation; either version 2 of the License, or
-##    (at your option) any later version.
-##
-##    This program is distributed in the hope that it will be useful,
-##    but WITHOUT ANY WARRANTY; without even the implied warranty of
-##    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-##    GNU General Public License for more details.
-##
-##    You should have received a copy of the GNU General Public License along
-##    with this program; if not, write to the Free Software Foundation, Inc.,
-##    51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
-import os
-import shutil
-import sys
-import hashlib
-import re
-import pkg_resources
-import subprocess
-from tempfile import mkdtemp
-from optparse import OptionParser
-from libprs500.lrf import ConversionError
-from libprs500.lrf.meta import LRFException, LRFMetaFile
-from libprs500.ptempfile import PersistentTemporaryFile
-
-def generate_thumbnail(path):
-    """ Generate a JPEG thumbnail of size ~ 128x128 (aspect ratio preserved)"""
-    try:
-        from PIL import Image
-    except ImportError:
-        raise LRFException("Unable to initialize Python Imaging Library." \
-                "Thumbnail generation is disabled")
-    im = Image.open(path)    
-    im.thumbnail((128, 128), Image.ANTIALIAS)
-    thumb = PersistentTemporaryFile(prefix="makelrf_", suffix=".jpeg")
-    thumb.close()
-    im = im.convert()
-    im.save(thumb.name)
-    return thumb
-    
-
-def makelrf(author=None, title=None, \
-            thumbnail=None, src=None, odir=".",\
-            rasterize=True, cover=None):
-    src = os.path.normpath(os.path.abspath(src))
-    bbebook = pkg_resources.resource_filename(__name__, _bbebook)
-    if not os.access(src, os.R_OK):
-        raise LRFException("Unable to read from file: " + src)
-    if thumbnail:
-            thumb = os.path.abspath(thumbnail)
-            if not os.access(thumb, os.R_OK):
-                raise LRFException("Unable to read from " + thumb)
-    else:
-        thumb = pkg_resources.resource_filename(__name__, 'cover.jpg')
-            
-    if not author:
-        author = "Unknown"    
-    if not title:
-        title = os.path.basename(src)
-    label = os.path.basename(src)
-    id = 'FB' + hashlib.md5(os.path.basename(label)).hexdigest()[:14]
-    name, ext = os.path.splitext(label)
-    cwd = os.path.dirname(src)    
-    dirpath = None
-    try:
-        if ext == ".rar":
-            dirpath = mkdtemp('','makelrf')
-            cwd = dirpath            
-            cmd = " ".join(["unrar", "e", '"'+src+'"'])
-            proc = subprocess.Popen(cmd, cwd=cwd, shell=True, stderr=subprocess.PIPE)
-            if proc.wait():
-                raise LRFException("unrar failed with error:\n\n" + \
-                        proc.stderr.read())
-            path, msize = None, 0
-            for root, dirs, files in os.walk(dirpath):
-                for name in files:
-                    if os.path.splitext(name)[1] == ".html":
-                        size = os.stat(os.path.join(root, name)).st_size                        
-                        if size > msize:
-                            msize, path = size, os.path.join(root, name)
-            if not path:
-                raise LRFException("Could not find .html file in rar archive")
-            src = path
-        
-        name = re.sub("\s", "_", name)
-        name = os.path.abspath(os.path.join(odir, name)) + ".lrf"
-        cfg = { 'File' : src, 'Output' : name, 'Label' : label, 'BookID' : id, \
-                'Author' : author, 'Title' : title, 'Publisher' : 'Unknown' \
-              }        
-        
-        
-        if cover:
-            cover = os.path.normpath(os.path.abspath(cover))
-            try:
-                thumbf = generate_thumbnail(cover)
-                thumb = thumbf.name
-            except Exception, e:
-                print >> sys.stderr, "WARNING: Unable to generate thumbnail:\n", \
-                         str(e)
-                thumb = cover
-            cfg['Cover'] = cover
-        cfg['Icon'] = thumb
-        config = PersistentTemporaryFile(prefix='makelrf_', suffix='.xml')
-        config.write(create_xml(cfg))
-        config.close()
-        jar = '-jar "' + bbebook + '"'
-        cmd = " ".join(["java", jar, "-r" if rasterize else "", '"'+config.name+'"'])
-        proc = subprocess.Popen(cmd, \
-               cwd=cwd, shell=True, stderr=subprocess.PIPE)
-        if proc.wait():
-            raise LRFException("BBeBook failed with error:\n\n" + \
-                    proc.stderr.read())
-        # Needed as BBeBook-0.2 doesn't handle non GIF thumbnails correctly.
-        lrf = open(name, "r+b")
-        LRFMetaFile(lrf).fix_thumbnail_type()
-        lrf.close()
-        return name
-    finally:
-        if dirpath: 
-            shutil.rmtree(dirpath, True)
-

upload

@@ -0,0 +1,21 @@
+#!/bin/bash
+
+# Script to upload libprs500 to pypi, the installer to libprs500.kovidgoyal.net and update the api documentation
+
+PREFIX=/var/www/vhosts/kovidgoyal.net/subdomains/libprs500
+DOWNLOADS=$PREFIX/httpdocs/downloads
+DOCS=$PREFIX/httpdocs/apidocs
+exe=`cd dist && ls -1 libprs500-*.exe | tail -n1 && cd ..`
+
+ssh castalia rm -f $DOWNLOADS/libprs500\*.exe
+scp dist/$exe castalia:$DOWNLOADS/
+ssh castalia chmod a+r $DOWNLOADS/\*
+ssh castalia /root/bin/update-installer-link $exe
+
+epydoc --config epydoc.conf
+scp -r docs/html castalia:$DOCS/
+epydoc -v --config epydoc-pdf.conf
+scp docs/pdf/api.pdf castalia:$DOCS/
+
+python setup.py register sdist bdist_egg upload
+rm -rf dist/* build/*