IGN:Unquote all URLs in OPF files and fix inclusion of detected chapter in EPUB TOC

src/calibre/ebooks/epub/__init__.py

@@ -90,6 +90,8 @@ to auto-generate a Table of Contents.
         help=_('Maximum number of links from each HTML file to insert into the TOC. Set to 0 to disable. Default is: %default.'))
     toc('no_chapters_in_toc', ['--no-chapters-in-toc'], default=False,
         help=_("Don't add auto-detected chapters to the Table of Contents."))
+    toc('use_auto_toc', ['--use-auto-toc'], default=False,
+        help=_('Normally, if the source file already has a Table of Contents, it is used in preference to the autodetected one. With this option, the autodetected one is always used.'))
     
     layout = c.add_group('page layout', _('Control page layout'))
     layout('margin_top', ['--margin-top'], default=5.0, 

src/calibre/ebooks/epub/from_any.py

@@ -15,7 +15,7 @@ from calibre.ebooks.epub import config as common_config
 from calibre.ebooks.epub.from_html import convert as html2epub
 from calibre.ptempfile import TemporaryDirectory
 from calibre.ebooks.metadata import MetaInformation
-from calibre.ebooks.metadata.opf import OPFCreator
+from calibre.ebooks.metadata.opf2 import OPFCreator
 
 def lit2opf(path, tdir, opts):
     from calibre.ebooks.lit.reader import LitReader
@@ -74,7 +74,7 @@ MAP = {
        'txt'  : txt2opf,
        'pdf'  : pdf2opf,
        }
-SOURCE_FORMATS = ['lit', 'mobi', 'prc', 'fb2', 'rtf', 'txt', 'pdf']    
+SOURCE_FORMATS = ['lit', 'mobi', 'prc', 'fb2', 'rtf', 'txt', 'pdf', 'rar', 'zip']    
 
 def unarchive(path, tdir):
     extract(path, tdir)

src/calibre/ebooks/epub/from_html.py

@@ -1,5 +1,5 @@
 from __future__ import with_statement
-from calibre.ebooks.metadata.opf import OPFReader
+
 __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
 __docformat__ = 'restructuredtext en'
@@ -17,8 +17,10 @@ from calibre.ebooks.epub import config as common_config
 from calibre.ptempfile import TemporaryDirectory
 from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.metadata.toc import TOC
+from calibre.ebooks.metadata.opf2 import OPF
 from calibre.ebooks.epub import initialize_container, PROFILES
 from calibre.ebooks.epub.split import split
+from calibre.constants import preferred_encoding
 
 
 class HTMLProcessor(Processor):
@@ -84,11 +86,11 @@ def convert(htmlfile, opts, notification=None):
     opts.output = os.path.abspath(opts.output)
     if opts.override_css is not None:
         try:
-            opts.override_css = open(opts.override_css, 'rb').read().decode('utf-8', 'replace')
+            opts.override_css = open(opts.override_css, 'rb').read().decode(preferred_encoding, 'replace')
         except:
-            opts.override_css = opts.override_css.decode('utf-8', 'replace')
+            opts.override_css = opts.override_css.decode(preferred_encoding, 'replace')
     if htmlfile.lower().endswith('.opf'):
-        opf = OPFReader(htmlfile, os.path.dirname(os.path.abspath(htmlfile)))
+        opf = OPF(htmlfile, os.path.dirname(os.path.abspath(htmlfile)))
         filelist = opf_traverse(opf, verbose=opts.verbose, encoding=opts.encoding)
         mi = MetaInformation(opf)
     else:
@@ -141,7 +143,7 @@ def convert(htmlfile, opts, notification=None):
         buf = cStringIO.StringIO()
         if mi.toc:
             rebase_toc(mi.toc, htmlfile_map, tdir)
-        if mi.toc is None or len(mi.toc) < 2:
+        if opts.use_auto_toc or mi.toc is None or len(mi.toc) < 2:
             mi.toc = generated_toc
         for item in mi.manifest:
             if getattr(item, 'mime_type', None) == 'text/html':

src/calibre/ebooks/epub/split.py

@@ -8,7 +8,6 @@ Split the flows in an epub file to conform to size limitations.
 '''
 
 import os, math, copy, logging, functools
-from urllib import unquote
 
 from lxml.etree import XPath as _XPath
 from lxml import etree, html
@@ -57,7 +56,8 @@ class Splitter(LoggingInterface):
         if stylesheet is not None:
             self.find_page_breaks(stylesheet, root)
         
-        self.trees = self.split(root.getroottree())
+        self.trees = []
+        self.split(root.getroottree())
         self.commit()
         self.log_info('\t\tSplit into %d parts.', len(self.trees))
         if self.opts.verbose:
@@ -81,7 +81,6 @@ class Splitter(LoggingInterface):
         tree2 = copy.deepcopy(tree)
         root2 = tree2.getroot()
         body, body2 = root.body, root2.body
-        trees = []
         path = tree.getpath(split_point)
         split_point2 = root2.xpath(path)[0]
         
@@ -137,12 +136,11 @@ class Splitter(LoggingInterface):
         for t, r in [(tree, root), (tree2, root2)]:
             size = len(tostring(r)) 
             if size <= self.opts.profile.flow_size:
-                trees.append(t)
+                self.trees.append(t)
-                self.log_debug('\t\t\tCommitted sub-tree #%d (%d KB)', len(trees), size/1024.)
+                self.log_debug('\t\t\tCommitted sub-tree #%d (%d KB)', len(self.trees), size/1024.)
             else:
-                trees.extend(self.split(t))
+                self.split(t)
                 
-        return trees
                 
     def find_page_breaks(self, stylesheet, root):
         '''
@@ -334,7 +332,7 @@ def split(pathtoopf, opts):
         html_files = []
         for item in opf.itermanifest():
             if 'html' in item.get('media-type', '').lower():
-                html_files.append(unquote(item.get('href')).split('/')[-1])
+                html_files.append(item.get('href').split('/')[-1])
         changes = []
         for f in html_files:
             if os.stat(content(f)).st_size > opts.profile.flow_size:

src/calibre/ebooks/html.py

@@ -20,10 +20,9 @@ get_text = XPath("//text()")
 from calibre import LoggingInterface, unicode_path
 from calibre.ebooks.chardet import xml_to_unicode, ENCODING_PATS
 from calibre.utils.config import Config, StringConfig
-from calibre.ebooks.metadata.opf import OPFReader, OPFCreator
 from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.metadata.meta import get_metadata
-from calibre.ebooks.metadata.opf2 import OPF
+from calibre.ebooks.metadata.opf2 import OPF, OPFCreator
 from calibre.ptempfile import PersistentTemporaryDirectory, PersistentTemporaryFile
 from calibre.utils.zipfile import ZipFile
 
@@ -429,6 +428,8 @@ class Processor(Parser):
     def detect_chapters(self):
         self.detected_chapters = self.opts.chapter(self.root)
         for elem in self.detected_chapters:
+            text = u' '.join([t.strip() for t in elem.xpath('descendant::text()')])
+            self.log_info('\tDetected chapter: %s', text[:50])
             if self.opts.chapter_mark in ('both', 'pagebreak'):
                 style = elem.get('style', '').strip()
                 if style and not style.endswith(';'):
@@ -503,12 +504,16 @@ class Processor(Parser):
                         
         # Add chapters to TOC
         if not self.opts.no_chapters_in_toc:
+            counter = 0
             for elem in getattr(self, 'detected_chapters', []):
                 text = (u''.join(elem.xpath('string()'))).strip()
                 if text:
                     name = self.htmlfile_map[self.htmlfile.path]
                     href = 'content/'+name
-                    add_item(href, None, text, target)
+                    counter += 1
+                    id = elem.get('id', 'calibre_chapter_%d'%counter)
+                    elem.set('id', id)
+                    add_item(href, id, text, target)
                     
         
     def extract_css(self):
@@ -647,7 +652,7 @@ is used.
 def search_for_opf(dir):
     for f in os.listdir(dir):
         if f.lower().endswith('.opf'):
-            return OPFReader(open(os.path.join(dir, f), 'rb'), dir)
+            return OPF(open(os.path.join(dir, f), 'rb'), dir)
 
 
 def get_filelist(htmlfile, opts):
@@ -749,7 +754,7 @@ def create_dir(htmlfile, opts):
     Create a directory that contains the open ebook
     '''
     if htmlfile.lower().endswith('.opf'):
-        opf = OPFReader(open(htmlfile, 'rb'), os.path.dirname(os.path.abspath(htmlfile)))
+        opf = OPF(open(htmlfile, 'rb'), os.path.dirname(os.path.abspath(htmlfile)))
         filelist = opf_traverse(opf, verbose=opts.verbose, encoding=opts.encoding)
         mi = MetaInformation(opf)
     else:

src/calibre/ebooks/metadata/opf2.py

@@ -7,13 +7,154 @@ __docformat__ = 'restructuredtext en'
 lxml based OPF parser.
 '''
 
-import sys, unittest, functools, os
+import sys, unittest, functools, os, mimetypes, uuid
-from urllib import unquote, quote
+from urllib import unquote
+from urlparse import urlparse
 
 from lxml import etree
 
 from calibre.ebooks.chardet import xml_to_unicode
-from calibre.ebooks.metadata import Resource, ResourceCollection
+from calibre import relpath
+from calibre.constants import __appname__
+from calibre.ebooks.metadata.toc import TOC
+from calibre.ebooks.metadata import MetaInformation
+
+
+class Resource(object):
+    '''
+    Represents a resource (usually a file on the filesystem or a URL pointing 
+    to the web. Such resources are commonly referred to in OPF files.
+    
+    They have the interface:
+    
+    :member:`path`
+    :member:`mime_type`
+    :method:`href`
+    
+    '''
+    
+    def __init__(self, href_or_path, basedir=os.getcwd(), is_path=True):
+        self._href = None
+        self._basedir = basedir
+        self.path = None
+        self.fragment = ''
+        try:
+            self.mime_type = mimetypes.guess_type(href_or_path)[0]
+        except:
+            self.mime_type = None
+        if self.mime_type is None:
+            self.mime_type = 'application/octet-stream'
+        if is_path:
+            path = href_or_path
+            if not os.path.isabs(path):
+                path = os.path.abspath(os.path.join(basedir, path))
+            if isinstance(path, str):
+                path = path.decode(sys.getfilesystemencoding())
+            self.path = path
+        else:
+            href_or_path = href_or_path
+            url = urlparse(href_or_path)
+            if url[0] not in ('', 'file'):
+                self._href = href_or_path
+            else:
+                pc = url[2]
+                if isinstance(pc, unicode):
+                    pc = pc.encode('utf-8')
+                pc = pc.decode('utf-8')
+                self.path = os.path.abspath(os.path.join(basedir, pc.replace('/', os.sep)))
+                self.fragment = url[-1]
+        
+    
+    def href(self, basedir=None):
+        '''
+        Return a URL pointing to this resource. If it is a file on the filesystem
+        the URL is relative to `basedir`.
+        
+        `basedir`: If None, the basedir of this resource is used (see :method:`set_basedir`).
+        If this resource has no basedir, then the current working directory is used as the basedir.
+        '''
+        if basedir is None:
+            if self._basedir:
+                basedir = self._basedir
+            else:
+                basedir = os.getcwd()
+        if self.path is None:
+            return self._href
+        f = self.fragment.encode('utf-8') if isinstance(self.fragment, unicode) else self.fragment
+        frag = '#'+f if self.fragment else ''
+        if self.path == basedir:
+            return ''+frag
+        try:
+            rpath = relpath(self.path, basedir)
+        except OSError: # On windows path and basedir could be on different drives
+            rpath = self.path
+        if isinstance(rpath, unicode):
+            rpath = rpath.encode('utf-8')
+        return rpath.replace(os.sep, '/')+frag
+    
+    def set_basedir(self, path):
+        self._basedir = path
+        
+    def basedir(self):
+        return self._basedir
+    
+    def __repr__(self):
+        return 'Resource(%s, %s)'%(repr(self.path), repr(self.href()))
+        
+        
+class ResourceCollection(object):
+    
+    def __init__(self):
+        self._resources = []
+        
+    def __iter__(self):
+        for r in self._resources:
+            yield r
+            
+    def __len__(self):
+        return len(self._resources)
+    
+    def __getitem__(self, index):
+        return self._resources[index]
+    
+    def __bool__(self):
+        return len(self._resources) > 0
+    
+    def __str__(self):
+        resources = map(repr, self)
+        return '[%s]'%', '.join(resources)
+    
+    def __repr__(self):
+        return str(self)
+    
+    def append(self, resource):
+        if not isinstance(resource, Resource):
+            raise ValueError('Can only append objects of type Resource')
+        self._resources.append(resource)
+        
+    def remove(self, resource):
+        self._resources.remove(resource)
+    
+    def replace(self, start, end, items):
+        'Same as list[start:end] = items'
+        self._resources[start:end] = items
+        
+    @staticmethod
+    def from_directory_contents(top, topdown=True):
+        collection = ResourceCollection()
+        for spec in os.walk(top, topdown=topdown):
+            path = os.path.abspath(os.path.join(spec[0], spec[1]))
+            res = Resource.from_path(path)
+            res.set_basedir(top)
+            collection.append(res)
+        return collection
+    
+    def set_basedir(self, path):
+        for res in self:
+            res.set_basedir(path)
+        
+
+
 
 class ManifestItem(Resource):
     
@@ -21,8 +162,6 @@ class ManifestItem(Resource):
     def from_opf_manifest_item(item, basedir):
         href = item.get('href', None)
         if href:
-            if unquote(href) == href:
-                href = quote(href)
             res = ManifestItem(href, basedir=basedir, is_path=False)
             mt = item.get('media-type', '').strip()
             if mt:
@@ -293,6 +432,7 @@ class OPF(object):
         if not self.metadata:
             raise ValueError('Malformed OPF file: No <metadata> element')
         self.metadata      = self.metadata[0]
+        self.unquote_urls()
         self.manifest = Manifest()
         m = self.manifest_path(self.tree)
         if m:
@@ -307,6 +447,7 @@ class OPF(object):
             self.guide = Guide.from_opf_guide(guide, basedir)
         self.cover_data = (None, None)
         
+        
     def get_text(self, elem):
         return u''.join(self.TEXT(elem))
     
@@ -355,9 +496,11 @@ class OPF(object):
     def iterguide(self):
         return self.guide_path(self.tree)
     
-    def render(self):
+    def unquote_urls(self):
-        return etree.tostring(self.tree, encoding='UTF-8', xml_declaration=True, 
+        for item in self.itermanifest():
-                              pretty_print=True)
+            item.set('href', unquote(item.get('href', '')))
+        for item in self.iterguide():
+            item.set('href', unquote(item.get('href', '')))
     
     @apply
     def authors():
@@ -450,6 +593,116 @@ class OPF(object):
             if val or val == []:
                 setattr(self, attr, val)
 
+class OPFCreator(MetaInformation):
+    
+    def __init__(self, base_path, *args, **kwargs):
+        '''
+        Initialize.
+        @param base_path: An absolute path to the directory in which this OPF file
+        will eventually be. This is used by the L{create_manifest} method
+        to convert paths to files into relative paths.
+        '''
+        MetaInformation.__init__(self, *args, **kwargs)
+        self.base_path = os.path.abspath(base_path)
+        if self.application_id is None:
+            self.application_id = str(uuid.uuid4())
+        if not isinstance(self.toc, TOC):
+            self.toc = None
+        if not self.authors:
+            self.authors = [_('Unknown')]
+        if self.guide is None:
+            self.guide = Guide()
+        if self.cover:
+            self.guide.set_cover(self.cover)
+        
+        
+    def create_manifest(self, entries):
+        '''
+        Create <manifest>
+        
+        `entries`: List of (path, mime-type) If mime-type is None it is autodetected
+        '''
+        entries = map(lambda x: x if os.path.isabs(x[0]) else 
+                      (os.path.abspath(os.path.join(self.base_path, x[0])), x[1]),
+                      entries)
+        self.manifest = Manifest.from_paths(entries)
+        self.manifest.set_basedir(self.base_path)
+        
+    def create_manifest_from_files_in(self, files_and_dirs):
+        entries = []
+        
+        def dodir(dir):
+            for spec in os.walk(dir):
+                root, files = spec[0], spec[-1]
+                for name in files:
+                    path = os.path.join(root, name)
+                    if os.path.isfile(path):
+                        entries.append((path, None)) 
+        
+        for i in files_and_dirs:
+            if os.path.isdir(i):
+                dodir(i)
+            else:
+                entries.append((i, None))
+                
+        self.create_manifest(entries)    
+            
+    def create_spine(self, entries):
+        '''
+        Create the <spine> element. Must first call :method:`create_manifest`.
+        
+        `entries`: List of paths
+        '''
+        entries = map(lambda x: x if os.path.isabs(x) else 
+                      os.path.abspath(os.path.join(self.base_path, x)), entries)
+        self.spine = Spine.from_paths(entries, self.manifest)
+        
+    def set_toc(self, toc):
+        '''
+        Set the toc. You must call :method:`create_spine` before calling this
+        method.
+        
+        :param toc: A :class:`TOC` object
+        '''
+        self.toc = toc
+        
+    def create_guide(self, guide_element):
+        self.guide = Guide.from_opf_guide(guide_element, self.base_path)
+        self.guide.set_basedir(self.base_path)
+            
+    def render(self, opf_stream, ncx_stream=None, ncx_manifest_entry=None):
+        from calibre.resources import opf_template
+        from calibre.utils.genshi.template import MarkupTemplate
+        template = MarkupTemplate(opf_template)
+        if self.manifest:
+            self.manifest.set_basedir(self.base_path)
+            if ncx_manifest_entry is not None:
+                if not os.path.isabs(ncx_manifest_entry):
+                    ncx_manifest_entry = os.path.join(self.base_path, ncx_manifest_entry)
+                remove = [i for i in self.manifest if i.id == 'ncx']
+                for item in remove:
+                    self.manifest.remove(item)
+                self.manifest.append(ManifestItem(ncx_manifest_entry, self.base_path))
+                self.manifest[-1].id = 'ncx'
+                self.manifest[-1].mime_type = 'application/x-dtbncx+xml'
+        if not self.guide:
+            self.guide = Guide()
+        if self.cover:
+            cover = self.cover
+            if not os.path.isabs(cover):
+                cover = os.path.abspath(os.path.join(self.base_path, cover))
+            self.guide.set_cover(cover)
+        self.guide.set_basedir(self.base_path)
+        
+        opf = template.generate(__appname__=__appname__, mi=self).render('xml')
+        opf_stream.write(opf)
+        opf_stream.flush()
+        toc = getattr(self, 'toc', None)
+        if toc is not None and ncx_stream is not None:
+            toc.render(ncx_stream, self.application_id)
+            ncx_stream.flush()
+
+
 class OPFTest(unittest.TestCase):
     
     def setUp(self):

src/calibre/gui2/dialogs/epub.ui

@@ -77,7 +77,7 @@
      <item>
       <widget class="QStackedWidget" name="stack" >
        <property name="currentIndex" >
-        <number>3</number>
+        <number>0</number>
        </property>
        <widget class="QWidget" name="metadata_page" >
         <layout class="QGridLayout" name="gridLayout_4" >
@@ -89,36 +89,6 @@
               <string>Book Cover</string>
              </property>
              <layout class="QGridLayout" name="_2" >
-              <item row="0" column="0" >
-               <layout class="QHBoxLayout" name="_3" >
-                <item>
-                 <widget class="ImageView" name="cover" >
-                  <property name="text" >
-                   <string/>
-                  </property>
-                  <property name="pixmap" >
-                   <pixmap resource="../images.qrc" >:/images/book.svg</pixmap>
-                  </property>
-                  <property name="scaledContents" >
-                   <bool>true</bool>
-                  </property>
-                  <property name="alignment" >
-                   <set>Qt::AlignCenter</set>
-                  </property>
-                 </widget>
-                </item>
-               </layout>
-              </item>
-              <item row="2" column="0" >
-               <widget class="QCheckBox" name="opt_prefer_metadata_cover" >
-                <property name="text" >
-                 <string>Use cover from &amp;source file</string>
-                </property>
-                <property name="checked" >
-                 <bool>true</bool>
-                </property>
-               </widget>
-              </item>
               <item row="1" column="0" >
                <layout class="QVBoxLayout" name="_4" >
                 <property name="spacing" >
@@ -170,6 +140,36 @@
                 </item>
                </layout>
               </item>
+              <item row="2" column="0" >
+               <widget class="QCheckBox" name="opt_prefer_metadata_cover" >
+                <property name="text" >
+                 <string>Use cover from &amp;source file</string>
+                </property>
+                <property name="checked" >
+                 <bool>true</bool>
+                </property>
+               </widget>
+              </item>
+              <item row="0" column="0" >
+               <layout class="QHBoxLayout" name="_3" >
+                <item>
+                 <widget class="ImageView" name="cover" >
+                  <property name="text" >
+                   <string/>
+                  </property>
+                  <property name="pixmap" >
+                   <pixmap resource="../images.qrc" >:/images/book.svg</pixmap>
+                  </property>
+                  <property name="scaledContents" >
+                   <bool>true</bool>
+                  </property>
+                  <property name="alignment" >
+                   <set>Qt::AlignCenter</set>
+                  </property>
+                 </widget>
+                </item>
+               </layout>
+              </item>
              </layout>
              <zorder>opt_prefer_metadata_cover</zorder>
              <zorder></zorder>
@@ -590,12 +590,6 @@ p, li { white-space: pre-wrap; }
              </widget>
             </item>
            </layout>
-           <zorder>label_17</zorder>
-           <zorder>opt_chapter</zorder>
-           <zorder>label_8</zorder>
-           <zorder>opt_chapter_mark</zorder>
-           <zorder>label_9</zorder>
-           <zorder>verticalSpacer</zorder>
           </widget>
          </item>
          <item>
@@ -604,10 +598,10 @@ p, li { white-space: pre-wrap; }
             <string>Automatic &amp;Table of Contents</string>
            </property>
            <layout class="QGridLayout" name="gridLayout_5" >
-            <item row="1" column="1" >
+            <item row="2" column="1" >
              <widget class="QSpinBox" name="opt_max_toc_links" />
             </item>
-            <item row="1" column="0" >
+            <item row="2" column="0" >
              <widget class="QLabel" name="label_10" >
               <property name="text" >
                <string>Number of &amp;links to add to Table of Contents</string>
@@ -617,17 +611,17 @@ p, li { white-space: pre-wrap; }
               </property>
              </widget>
             </item>
-            <item row="0" column="0" >
+            <item row="1" column="0" >
              <widget class="QCheckBox" name="opt_no_chapters_in_toc" >
               <property name="text" >
-               <string>Do not add &amp;detected chapters ot the Table of Contents</string>
+               <string>Do not add &amp;detected chapters to the Table of Contents</string>
               </property>
              </widget>
             </item>
-            <item row="2" column="1" >
+            <item row="3" column="1" >
              <widget class="QSpinBox" name="opt_max_toc_recursion" />
             </item>
-            <item row="2" column="0" >
+            <item row="3" column="0" >
              <widget class="QLabel" name="label_16" >
               <property name="text" >
                <string>Table of Contents &amp;recursion</string>
@@ -637,6 +631,13 @@ p, li { white-space: pre-wrap; }
               </property>
              </widget>
             </item>
+            <item row="0" column="0" >
+             <widget class="QCheckBox" name="opt_use_auto_toc" >
+              <property name="text" >
+               <string>&amp;Force use of auto-generated Table of Contents</string>
+              </property>
+             </widget>
+            </item>
            </layout>
           </widget>
          </item>