cotnent indexing - bugs.

[wolnelektury.git] / apps / catalogue / models.py

diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py
index a5b0ebf..fbae111 100644 (file)
--- a/apps/catalogue/models.py
+++ b/apps/catalogue/models.py
@@ -2,7 +2,7 @@
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #

-from datetime import datetime
+from collections import namedtuple

 
 from django.db import models
 from django.db.models import permalink, Q
 
 from django.db import models
 from django.db.models import permalink, Q


@@ -23,13 +23,17 @@ from django.conf import settings
 from newtagging.models import TagBase, tags_updated
 from newtagging import managers
 from catalogue.fields import JSONField, OverwritingFileField
 from newtagging.models import TagBase, tags_updated
 from newtagging import managers
 from catalogue.fields import JSONField, OverwritingFileField

-from catalogue.utils import create_zip
+from catalogue.utils import create_zip, split_tags
+from catalogue.tasks import touch_tag, index_book

 from shutil import copy
 from glob import glob
 import re
 from os import path
 
 
 from shutil import copy
 from glob import glob
 import re
 from os import path
 
 

+import search
+
+# Those are hard-coded here so that makemessages sees them.

 TAG_CATEGORIES = (
     ('author', _('author')),
     ('epoch', _('epoch')),
 TAG_CATEGORIES = (
     ('author', _('author')),
     ('epoch', _('epoch')),


@@ -40,13 +44,6 @@ TAG_CATEGORIES = (
     ('book', _('book')),
 )
 
     ('book', _('book')),
 )
 

-MEDIA_FORMATS = (
-    ('odt', _('ODT file')),
-    ('mp3', _('MP3 file')),
-    ('ogg', _('OGG file')),
-    ('daisy', _('DAISY file')), 
-)
-

 # not quite, but Django wants you to set a timeout
 CACHE_FOREVER = 2419200  # 28 days
 
 # not quite, but Django wants you to set a timeout
 CACHE_FOREVER = 2419200  # 28 days
 


@@ -67,7 +64,6 @@ class Tag(TagBase):
     category = models.CharField(_('category'), max_length=50, blank=False, null=False,
         db_index=True, choices=TAG_CATEGORIES)
     description = models.TextField(_('description'), blank=True)
     category = models.CharField(_('category'), max_length=50, blank=False, null=False,
         db_index=True, choices=TAG_CATEGORIES)
     description = models.TextField(_('description'), blank=True)

-    main_page = models.BooleanField(_('main page'), default=False, db_index=True, help_text=_('Show tag on main page'))

 
     user = models.ForeignKey(User, blank=True, null=True)
     book_count = models.IntegerField(_('book count'), blank=True, null=True)
 
     user = models.ForeignKey(User, blank=True, null=True)
     book_count = models.IntegerField(_('book count'), blank=True, null=True)


@@ -112,25 +108,22 @@ class Tag(TagBase):
     has_description.boolean = True
 
     def get_count(self):
     has_description.boolean = True
 
     def get_count(self):

-        """ returns global book count for book tags, fragment count for themes """
-
-        if self.book_count is None:
-            if self.category == 'book':
-                # never used
-                objects = Book.objects.none()
-            elif self.category == 'theme':
-                objects = Fragment.tagged.with_all((self,))
-            else:
-                objects = Book.tagged.with_all((self,)).order_by()
-                if self.category != 'set':
-                    # eliminate descendants
-                    l_tags = Tag.objects.filter(slug__in=[book.book_tag_slug() for book in objects])
-                    descendants_keys = [book.pk for book in Book.tagged.with_any(l_tags)]
-                    if descendants_keys:
-                        objects = objects.exclude(pk__in=descendants_keys)
-            self.book_count = objects.count()
-            self.save()
-        return self.book_count
+        """Returns global book count for book tags, fragment count for themes."""
+
+        if self.category == 'book':
+            # never used
+            objects = Book.objects.none()
+        elif self.category == 'theme':
+            objects = Fragment.tagged.with_all((self,))
+        else:
+            objects = Book.tagged.with_all((self,)).order_by()
+            if self.category != 'set':
+                # eliminate descendants
+                l_tags = Tag.objects.filter(slug__in=[book.book_tag_slug() for book in objects])
+                descendants_keys = [book.pk for book in Book.tagged.with_any(l_tags)]
+                if descendants_keys:
+                    objects = objects.exclude(pk__in=descendants_keys)
+        return objects.count()

 
     @staticmethod
     def get_tag_list(tags):
 
     @staticmethod
     def get_tag_list(tags):


@@ -174,16 +167,43 @@ class Tag(TagBase):
     def url_chunk(self):
         return '/'.join((Tag.categories_dict[self.category], self.slug))
 
     def url_chunk(self):
         return '/'.join((Tag.categories_dict[self.category], self.slug))
 

+    @staticmethod
+    def tags_from_info(info):
+        from slughifi import slughifi
+        from sortify import sortify
+        meta_tags = []
+        categories = (('kinds', 'kind'), ('genres', 'genre'), ('authors', 'author'), ('epochs', 'epoch'))
+        for field_name, category in categories:
+            try:
+                tag_names = getattr(info, field_name)
+            except:
+                try:
+                    tag_names = [getattr(info, category)]
+                except:
+                    # For instance, Pictures do not have 'genre' field.
+                    continue
+            for tag_name in tag_names:
+                tag_sort_key = tag_name
+                if category == 'author':
+                    tag_sort_key = tag_name.last_name
+                    tag_name = tag_name.readable()
+                tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
+                if created:
+                    tag.name = tag_name
+                    tag.sort_key = sortify(tag_sort_key.lower())
+                    tag.save()
+                meta_tags.append(tag)
+        return meta_tags
+
+

 
 def get_dynamic_path(media, filename, ext=None, maxlen=100):
     from slughifi import slughifi
 
     # how to put related book's slug here?
     if not ext:
 
 def get_dynamic_path(media, filename, ext=None, maxlen=100):
     from slughifi import slughifi
 
     # how to put related book's slug here?
     if not ext:

-        if media.type == 'daisy':
-            ext = 'daisy.zip'
-        else:
-            ext = media.type
+        # BookMedia case
+        ext = media.formats[media.type].ext

     if media is None or not media.name:
         name = slughifi(filename.split(".")[0])
     else:
     if media is None or not media.name:
         name = slughifi(filename.split(".")[0])
     else:


@@ -203,7 +223,7 @@ def get_customized_pdf_path(book, customizations):
     customizations.sort()
     h = hash(tuple(customizations))
 
     customizations.sort()
     h = hash(tuple(customizations))
 

-    pdf_name = '%s-custom-%s' % (book.fileid(), h)
+    pdf_name = '%s-custom-%s' % (book.slug, h)

     pdf_file = get_dynamic_path(None, pdf_name, ext='pdf')
 
     return pdf_file
     pdf_file = get_dynamic_path(None, pdf_name, ext='pdf')
 
     return pdf_file


@@ -213,14 +233,23 @@ def get_existing_customized_pdf(book):
     """
     Returns a list of paths to generated customized pdf of a book
     """
     """
     Returns a list of paths to generated customized pdf of a book
     """

-    pdf_glob = '%s-custom-' % (book.fileid(),)
+    pdf_glob = '%s-custom-' % (book.slug,)

     pdf_glob = get_dynamic_path(None, pdf_glob, ext='pdf')
     pdf_glob = re.sub(r"[.]([a-z0-9]+)$", "*.\\1", pdf_glob)
     return glob(path.join(settings.MEDIA_ROOT, pdf_glob))
 
 
 class BookMedia(models.Model):
     pdf_glob = get_dynamic_path(None, pdf_glob, ext='pdf')
     pdf_glob = re.sub(r"[.]([a-z0-9]+)$", "*.\\1", pdf_glob)
     return glob(path.join(settings.MEDIA_ROOT, pdf_glob))
 
 
 class BookMedia(models.Model):

-    type        = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
+    FileFormat = namedtuple("FileFormat", "name ext")
+    formats = SortedDict([
+        ('mp3', FileFormat(name='MP3', ext='mp3')),
+        ('ogg', FileFormat(name='Ogg Vorbis', ext='ogg')),
+        ('daisy', FileFormat(name='DAISY', ext='daisy.zip')),
+    ])
+    format_choices = [(k, _('%s file') % t.name)
+            for k, t in formats.items()]
+
+    type        = models.CharField(_('type'), choices=format_choices, max_length="100")

     name        = models.CharField(_('name'), max_length="100")
     file        = OverwritingFileField(_('file'), upload_to=book_upload_path())
     uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)
     name        = models.CharField(_('name'), max_length="100")
     file        = OverwritingFileField(_('file'), upload_to=book_upload_path())
     uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)


@@ -243,7 +272,7 @@ class BookMedia(models.Model):
         try:
             old = BookMedia.objects.get(pk=self.pk)
         except BookMedia.DoesNotExist, e:
         try:
             old = BookMedia.objects.get(pk=self.pk)
         except BookMedia.DoesNotExist, e:

-            pass
+            old = None

         else:
             # if name changed, change the file name, too
             if slughifi(self.name) != slughifi(old.name):
         else:
             # if name changed, change the file name, too
             if slughifi(self.name) != slughifi(old.name):


@@ -252,7 +281,9 @@ class BookMedia(models.Model):
         super(BookMedia, self).save(*args, **kwargs)
 
         # remove the zip package for book with modified media
         super(BookMedia, self).save(*args, **kwargs)
 
         # remove the zip package for book with modified media

-        remove_zip(self.book.fileid())
+        if old:
+            remove_zip("%s_%s" % (old.book.slug, old.type))
+        remove_zip("%s_%s" % (self.book.slug, self.type))

 
         extra_info = self.get_extra_info_value()
         extra_info.update(self.read_meta())
 
         extra_info = self.get_extra_info_value()
         extra_info.update(self.read_meta())


@@ -321,7 +352,9 @@ class BookMedia(models.Model):
 class Book(models.Model):
     title         = models.CharField(_('title'), max_length=120)
     sort_key = models.CharField(_('sort key'), max_length=120, db_index=True, editable=False)
 class Book(models.Model):
     title         = models.CharField(_('title'), max_length=120)
     sort_key = models.CharField(_('sort key'), max_length=120, db_index=True, editable=False)

-    slug          = models.SlugField(_('slug'), max_length=120, db_index=True)
+    slug = models.SlugField(_('slug'), max_length=120, db_index=True,
+            unique=True)
+    common_slug = models.SlugField(_('slug'), max_length=120, db_index=True)

     language = models.CharField(_('language code'), max_length=3, db_index=True,
                     default=settings.CATALOGUE_DEFAULT_LANGUAGE)
     description   = models.TextField(_('description'), blank=True)
     language = models.CharField(_('language code'), max_length=3, db_index=True,
                     default=settings.CATALOGUE_DEFAULT_LANGUAGE)
     description   = models.TextField(_('description'), blank=True)


@@ -333,8 +366,11 @@ class Book(models.Model):
     wiki_link     = models.CharField(blank=True, max_length=240)
     # files generated during publication
 
     wiki_link     = models.CharField(blank=True, max_length=240)
     # files generated during publication
 

-    file_types = ['epub', 'html', 'mobi', 'pdf', 'txt', 'xml']
-    
+    cover = models.FileField(_('cover'), upload_to=book_upload_path('png'),
+                null=True, blank=True)
+    ebook_formats = ['pdf', 'epub', 'mobi', 'txt']
+    formats = ebook_formats + ['html', 'xml']
+

     parent        = models.ForeignKey('self', blank=True, null=True, related_name='children')
     objects  = models.Manager()
     tagged   = managers.ModelTaggedItemManager(Tag)
     parent        = models.ForeignKey('self', blank=True, null=True, related_name='children')
     objects  = models.Manager()
     tagged   = managers.ModelTaggedItemManager(Tag)


@@ -343,14 +379,10 @@ class Book(models.Model):
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
 
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
 

-    URLID_RE = r'[a-z0-9-]+(?:/[a-z]{3})?'
-    FILEID_RE = r'[a-z0-9-]+(?:_[a-z]{3})?'
-

     class AlreadyExists(Exception):
         pass
 
     class Meta:
     class AlreadyExists(Exception):
         pass
 
     class Meta:

-        unique_together = [['slug', 'language']]

         ordering = ('sort_key',)
         verbose_name = _('book')
         verbose_name_plural = _('books')
         ordering = ('sort_key',)
         verbose_name = _('book')
         verbose_name_plural = _('books')


@@ -358,42 +390,6 @@ class Book(models.Model):
     def __unicode__(self):
         return self.title
 
     def __unicode__(self):
         return self.title
 

-    def urlid(self, sep='/'):
-        stem = self.slug
-        if self.language != settings.CATALOGUE_DEFAULT_LANGUAGE:
-            stem += sep + self.language
-        return stem
-
-    def fileid(self):
-        return self.urlid('_')
-
-    @staticmethod
-    def split_urlid(urlid, sep='/', default_lang=settings.CATALOGUE_DEFAULT_LANGUAGE):
-        """Splits a URL book id into slug and language code.
-        
-        Returns a dictionary usable i.e. for object lookup, or None.
-
-        >>> Book.split_urlid("a-slug/pol", default_lang="eng")
-        {'slug': 'a-slug', 'language': 'pol'}
-        >>> Book.split_urlid("a-slug", default_lang="eng")
-        {'slug': 'a-slug', 'language': 'eng'}
-        >>> Book.split_urlid("a-slug_pol", "_", default_lang="eng")
-        {'slug': 'a-slug', 'language': 'pol'}
-        >>> Book.split_urlid("a-slug/eng", default_lang="eng")
-
-        """
-        parts = urlid.rsplit(sep, 1)
-        if len(parts) == 2:
-            if parts[1] == default_lang:
-                return None
-            return {'slug': parts[0], 'language': parts[1]}
-        else:
-            return {'slug': urlid, 'language': default_lang}
-
-    @classmethod
-    def split_fileid(cls, fileid):
-        return cls.split_urlid(fileid, '_')
-

     def save(self, force_insert=False, force_update=False, reset_short_html=True, **kwargs):
         from sortify import sortify
 
     def save(self, force_insert=False, force_update=False, reset_short_html=True, **kwargs):
         from sortify import sortify
 


@@ -408,18 +404,14 @@ class Book(models.Model):
 
     @permalink
     def get_absolute_url(self):
 
     @permalink
     def get_absolute_url(self):

-        return ('catalogue.views.book_detail', [self.urlid()])
+        return ('catalogue.views.book_detail', [self.slug])

 
     @property
     def name(self):
         return self.title
 
     def book_tag_slug(self):
 
     @property
     def name(self):
         return self.title
 
     def book_tag_slug(self):

-        stem = 'l-' + self.slug
-        if self.language != settings.CATALOGUE_DEFAULT_LANGUAGE:
-            return stem[:116] + ' ' + self.language
-        else:
-            return stem[:120]
+        return ('l-' + self.slug)[:120]

 
     def book_tag(self):
         slug = self.book_tag_slug()
 
     def book_tag(self):
         slug = self.book_tag_slug()


@@ -431,14 +423,14 @@ class Book(models.Model):
         return book_tag
 
     def has_media(self, type):
         return book_tag
 
     def has_media(self, type):

-        if type in Book.file_types:
+        if type in Book.formats:

             return bool(getattr(self, "%s_file" % type))
         else:
             return self.media.filter(type=type).exists()
 
     def get_media(self, type):
         if self.has_media(type):
             return bool(getattr(self, "%s_file" % type))
         else:
             return self.media.filter(type=type).exists()
 
     def get_media(self, type):
         if self.has_media(type):

-            if type in Book.file_types:
+            if type in Book.formats:

                 return getattr(self, "%s_file" % type)
             else:                                             
                 return self.media.filter(type=type)
                 return getattr(self, "%s_file" % type)
             else:                                             
                 return self.media.filter(type=type)


@@ -461,6 +453,7 @@ class Book(models.Model):
         cache_key = "Book.short_html/%d/%s"
         for lang, langname in settings.LANGUAGES:
             cache.delete(cache_key % (self.id, lang))
         cache_key = "Book.short_html/%d/%s"
         for lang, langname in settings.LANGUAGES:
             cache.delete(cache_key % (self.id, lang))

+        cache.delete("Book.mini_box/%d" % (self.id, ))

         # Fragment.short_html relies on book's tags, so reset it here too
         for fragm in self.fragments.all():
             fragm.reset_short_html()
         # Fragment.short_html relies on book's tags, so reset it here too
         for fragm in self.fragments.all():
             fragm.reset_short_html()


@@ -475,26 +468,15 @@ class Book(models.Model):
         if short_html is not None:
             return mark_safe(short_html)
         else:
         if short_html is not None:
             return mark_safe(short_html)
         else:

-            tags = self.tags.filter(~Q(category__in=('set', 'theme', 'book')))
-            tags = [mark_safe(u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name)) for tag in tags]
+            tags = self.tags.filter(category__in=('author', 'kind', 'genre', 'epoch'))
+            tags = split_tags(tags)

 
 

-            formats = []
+            formats = {}

             # files generated during publication
             # files generated during publication

-            if self.has_media("html"):
-                formats.append(u'<a href="%s">%s</a>' % (reverse('book_text', args=[self.fileid()]), _('Read online')))
-            if self.has_media("pdf"):
-                formats.append(u'<a href="%s">PDF</a>' % self.get_media('pdf').url)
-            if self.has_media("mobi"):
-                formats.append(u'<a href="%s">MOBI</a>' % self.get_media('mobi').url)
-            if self.root_ancestor.has_media("epub"):
-                formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.get_media('epub').url)
-            if self.has_media("txt"):
-                formats.append(u'<a href="%s">TXT</a>' % self.get_media('txt').url)
-            # other files
-            for m in self.media.order_by('type'):
-                formats.append(u'<a href="%s">%s</a>' % (m.file.url, m.type.upper()))
-
-            formats = [mark_safe(format) for format in formats]
+            for ebook_format in self.ebook_formats:
+                if self.has_media(ebook_format):
+                    formats[ebook_format] = self.get_media(ebook_format)
+

 
             short_html = unicode(render_to_string('catalogue/book_short.html',
                 {'book': self, 'tags': tags, 'formats': formats}))
 
             short_html = unicode(render_to_string('catalogue/book_short.html',
                 {'book': self, 'tags': tags, 'formats': formats}))


@@ -505,7 +487,7 @@ class Book(models.Model):
 
     def mini_box(self):
         if self.id:
 
     def mini_box(self):
         if self.id:

-            cache_key = "Book.mini_boxs/%d" % (self.id, )
+            cache_key = "Book.mini_box/%d" % (self.id, )

             short_html = cache.get(cache_key)
         else:
             short_html = None
             short_html = cache.get(cache_key)
         else:
             short_html = None


@@ -520,29 +502,12 @@ class Book(models.Model):
                 cache.set(cache_key, short_html, CACHE_FOREVER)
         return mark_safe(short_html)
 
                 cache.set(cache_key, short_html, CACHE_FOREVER)
         return mark_safe(short_html)
 

-    @property
-    def root_ancestor(self):
-        """ returns the oldest ancestor """
-
-        if not hasattr(self, '_root_ancestor'):
-            book = self
-            while book.parent:
-                book = book.parent
-            self._root_ancestor = book
-        return self._root_ancestor
-
-

     def has_description(self):
         return len(self.description) > 0
     has_description.short_description = _('description')
     has_description.boolean = True
 
     # ugly ugly ugly
     def has_description(self):
         return len(self.description) > 0
     has_description.short_description = _('description')
     has_description.boolean = True
 
     # ugly ugly ugly

-    def has_odt_file(self):
-        return bool(self.has_media("odt"))
-    has_odt_file.short_description = 'ODT'
-    has_odt_file.boolean = True
-

     def has_mp3_file(self):
         return bool(self.has_media("mp3"))
     has_mp3_file.short_description = 'MP3'
     def has_mp3_file(self):
         return bool(self.has_media("mp3"))
     has_mp3_file.short_description = 'MP3'


@@ -559,13 +524,27 @@ class Book(models.Model):
     has_daisy_file.boolean = True
 
     def wldocument(self, parse_dublincore=True):
     has_daisy_file.boolean = True
 
     def wldocument(self, parse_dublincore=True):

-        from catalogue.utils import ORMDocProvider
+        from catalogue.import_utils import ORMDocProvider

         from librarian.parser import WLDocument
 
         return WLDocument.from_file(self.xml_file.path,
                 provider=ORMDocProvider(self),
                 parse_dublincore=parse_dublincore)
 
         from librarian.parser import WLDocument
 
         return WLDocument.from_file(self.xml_file.path,
                 provider=ORMDocProvider(self),
                 parse_dublincore=parse_dublincore)
 

+    def build_cover(self, book_info=None):
+        """(Re)builds the cover image."""
+        from StringIO import StringIO
+        from django.core.files.base import ContentFile
+        from librarian.cover import WLCover
+
+        if book_info is None:
+            book_info = self.wldocument().book_info
+
+        cover = WLCover(book_info).image()
+        imgstr = StringIO()
+        cover.save(imgstr, 'png')
+        self.cover.save(None, ContentFile(imgstr.getvalue()))
+

     def build_pdf(self, customizations=None, file_name=None):
         """ (Re)builds the pdf file.
         customizations - customizations which are passed to LaTeX class file.
     def build_pdf(self, customizations=None, file_name=None):
         """ (Re)builds the pdf file.
         customizations - customizations which are passed to LaTeX class file.


@@ -581,7 +560,7 @@ class Book(models.Model):
             # we'd like to be sure not to overwrite changes happening while
             # (timely) pdf generation is taking place (async celery scenario)
             current_self = Book.objects.get(id=self.id)
             # we'd like to be sure not to overwrite changes happening while
             # (timely) pdf generation is taking place (async celery scenario)
             current_self = Book.objects.get(id=self.id)

-            current_self.pdf_file.save('%s.pdf' % self.fileid(),
+            current_self.pdf_file.save('%s.pdf' % self.slug,

                     File(open(pdf.get_filename())))
             self.pdf_file = current_self.pdf_file
 
                     File(open(pdf.get_filename())))
             self.pdf_file = current_self.pdf_file
 


@@ -604,39 +583,20 @@ class Book(models.Model):
 
         mobi = self.wldocument().as_mobi()
 
 
         mobi = self.wldocument().as_mobi()
 

-        self.mobi_file.save('%s.mobi' % self.fileid(), File(open(mobi.get_filename())))
+        self.mobi_file.save('%s.mobi' % self.slug, File(open(mobi.get_filename())))

 
         # remove zip with all mobi files
         remove_zip(settings.ALL_MOBI_ZIP)
 
 
         # remove zip with all mobi files
         remove_zip(settings.ALL_MOBI_ZIP)
 

-    def build_epub(self, remove_descendants=True):
-        """ (Re)builds the epub file.
-            If book has a parent, does nothing.
-            Unless remove_descendants is False, descendants' epubs are removed.
-        """
+    def build_epub(self):
+        """(Re)builds the epub file."""

         from django.core.files import File
         from catalogue.utils import remove_zip
 
         from django.core.files import File
         from catalogue.utils import remove_zip
 

-        if self.parent:
-            # don't need an epub
-            return
-

         epub = self.wldocument().as_epub()
 
         epub = self.wldocument().as_epub()
 

-        try:
-            epub.transform(ORMDocProvider(self), self.fileid(), output_file=epub_file)
-            self.epub_file.save('%s.epub' % self.fileid(), File(open(epub.get_filename())))
-        except NoDublinCore:
-            pass
-
-        book_descendants = list(self.children.all())
-        while len(book_descendants) > 0:
-            child_book = book_descendants.pop(0)
-            if remove_descendants and child_book.has_epub_file():
-                child_book.epub_file.delete()
-            # save anyway, to refresh short_html
-            child_book.save()
-            book_descendants += list(child_book.children.all())
+        self.epub_file.save('%s.epub' % self.slug,
+                File(open(epub.get_filename())))

 
         # remove zip package with all epub files
         remove_zip(settings.ALL_EPUB_ZIP)
 
         # remove zip package with all epub files
         remove_zip(settings.ALL_EPUB_ZIP)


@@ -645,7 +605,7 @@ class Book(models.Model):
         from django.core.files.base import ContentFile
 
         text = self.wldocument().as_text()
         from django.core.files.base import ContentFile
 
         text = self.wldocument().as_text()

-        self.txt_file.save('%s.txt' % self.fileid(), ContentFile(text.get_string()))
+        self.txt_file.save('%s.txt' % self.slug, ContentFile(text.get_string()))

 
 
     def build_html(self):
 
 
     def build_html(self):


@@ -660,7 +620,7 @@ class Book(models.Model):
 
         html_output = self.wldocument(parse_dublincore=False).as_html()
         if html_output:
 
         html_output = self.wldocument(parse_dublincore=False).as_html()
         if html_output:

-            self.html_file.save('%s.html' % self.fileid(),
+            self.html_file.save('%s.html' % self.slug,

                     ContentFile(html_output.get_string()))
 
             # get ancestor l-tags for adding to new fragments
                     ContentFile(html_output.get_string()))
 
             # get ancestor l-tags for adding to new fragments


@@ -711,7 +671,7 @@ class Book(models.Model):
         def pretty_file_name(book):
             return "%s/%s.%s" % (
                 b.get_extra_info_value()['author'],
         def pretty_file_name(book):
             return "%s/%s.%s" % (
                 b.get_extra_info_value()['author'],

-                b.fileid(),
+                b.slug,

                 format_)
 
         field_name = "%s_file" % format_
                 format_)
 
         field_name = "%s_file" % format_


@@ -722,12 +682,25 @@ class Book(models.Model):
                     getattr(settings, "ALL_%s_ZIP" % format_.upper()))
         return result.wait()
 
                     getattr(settings, "ALL_%s_ZIP" % format_.upper()))
         return result.wait()
 

-    def zip_audiobooks(self):
-        bm = BookMedia.objects.filter(book=self, type='mp3')
+    def zip_audiobooks(self, format_):
+        bm = BookMedia.objects.filter(book=self, type=format_)

         paths = map(lambda bm: (None, bm.file.path), bm)
         paths = map(lambda bm: (None, bm.file.path), bm)

-        result = create_zip.delay(paths, self.fileid())
+        result = create_zip.delay(paths, "%s_%s" % (self.slug, format_))

         return result.wait()
 
         return result.wait()
 

+    def search_index(self, book_info=None, reuse_index=False):
+        if reuse_index:
+            idx = search.ReusableIndex()
+        else:
+            idx = search.Index()
+            
+        idx.open()
+        try:
+            idx.index_book(self, book_info)
+            idx.index_tags()
+        finally:
+            idx.close()
+

     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
         from django.core.files import File
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
         from django.core.files import File


@@ -746,9 +719,9 @@ class Book(models.Model):
 
     @classmethod
     def from_text_and_meta(cls, raw_file, book_info, overwrite=False,
 
     @classmethod
     def from_text_and_meta(cls, raw_file, book_info, overwrite=False,

-            build_epub=True, build_txt=True, build_pdf=True, build_mobi=True):
+            build_epub=True, build_txt=True, build_pdf=True, build_mobi=True,
+            search_index=True):

         import re
         import re

-        from slughifi import slughifi

         from sortify import sortify
 
         # check for parts before we do anything
         from sortify import sortify
 
         # check for parts before we do anything


@@ -756,51 +729,37 @@ class Book(models.Model):
         if hasattr(book_info, 'parts'):
             for part_url in book_info.parts:
                 try:
         if hasattr(book_info, 'parts'):
             for part_url in book_info.parts:
                 try:

-                    children.append(Book.objects.get(
-                        slug=part_url.slug, language=part_url.language))
+                    children.append(Book.objects.get(slug=part_url.slug))

                 except Book.DoesNotExist, e:
                 except Book.DoesNotExist, e:

-                    raise Book.DoesNotExist(_('Book "%s/%s" does not exist.') %
-                            (part_url.slug, part_url.language))
+                    raise Book.DoesNotExist(_('Book "%s" does not exist.') %
+                            part_url.slug)

 
 
         # Read book metadata
         book_slug = book_info.url.slug
 
 
         # Read book metadata
         book_slug = book_info.url.slug

-        language = book_info.language
-        if re.search(r'[^a-zA-Z0-9-]', book_slug):
+        if re.search(r'[^a-z0-9-]', book_slug):

             raise ValueError('Invalid characters in slug')
             raise ValueError('Invalid characters in slug')

-        book, created = Book.objects.get_or_create(slug=book_slug, language=language)
+        book, created = Book.objects.get_or_create(slug=book_slug)

 
         if created:
             book_shelves = []
         else:
             if not overwrite:
 
         if created:
             book_shelves = []
         else:
             if not overwrite:

-                raise Book.AlreadyExists(_('Book %s/%s already exists') % (
-                        book_slug, language))
+                raise Book.AlreadyExists(_('Book %s already exists') % (
+                        book_slug))

             # Save shelves for this book
             book_shelves = list(book.tags.filter(category='set'))
 
             # Save shelves for this book
             book_shelves = list(book.tags.filter(category='set'))
 

+        book.language = book_info.language

         book.title = book_info.title
         book.title = book_info.title

+        if book_info.variant_of:
+            book.common_slug = book_info.variant_of.slug
+        else:
+            book.common_slug = book.slug

         book.set_extra_info_value(book_info.to_dict())
         book.save()
 
         book.set_extra_info_value(book_info.to_dict())
         book.save()
 

-        meta_tags = []
-        categories = (('kinds', 'kind'), ('genres', 'genre'), ('authors', 'author'), ('epochs', 'epoch'))
-        for field_name, category in categories:
-            try:
-                tag_names = getattr(book_info, field_name)
-            except:
-                tag_names = [getattr(book_info, category)]
-            for tag_name in tag_names:
-                tag_sort_key = tag_name
-                if category == 'author':
-                    tag_sort_key = tag_name.last_name
-                    tag_name = ' '.join(tag_name.first_names) + ' ' + tag_name.last_name
-                tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
-                if created:
-                    tag.name = tag_name
-                    tag.sort_key = sortify(tag_sort_key.lower())
-                    tag.save()
-                meta_tags.append(tag)
+        meta_tags = Tag.tags_from_info(book_info)

 
         book.tags = set(meta_tags + book_shelves)
 
 
         book.tags = set(meta_tags + book_shelves)
 


@@ -821,26 +780,35 @@ class Book(models.Model):
             if not settings.NO_BUILD_TXT and build_txt:
                 book.build_txt()
 
             if not settings.NO_BUILD_TXT and build_txt:
                 book.build_txt()
 

+        book.build_cover(book_info)
+

         if not settings.NO_BUILD_EPUB and build_epub:
         if not settings.NO_BUILD_EPUB and build_epub:

-            book.root_ancestor.build_epub()
+            book.build_epub()

 
         if not settings.NO_BUILD_PDF and build_pdf:
 
         if not settings.NO_BUILD_PDF and build_pdf:

-            book.root_ancestor.build_pdf()
+            book.build_pdf()

 
         if not settings.NO_BUILD_MOBI and build_mobi:
             book.build_mobi()
 
 
         if not settings.NO_BUILD_MOBI and build_mobi:
             book.build_mobi()
 

+        if not settings.NO_SEARCH_INDEX and search_index:
+            index_book.delay(book.id, book_info)
+

         book_descendants = list(book.children.all())
         book_descendants = list(book.children.all())

+        descendants_tags = set()

         # add l-tag to descendants and their fragments
         # add l-tag to descendants and their fragments

-        # delete unnecessary EPUB files

         while len(book_descendants) > 0:
             child_book = book_descendants.pop(0)
         while len(book_descendants) > 0:
             child_book = book_descendants.pop(0)

+            descendants_tags.update(child_book.tags)

             child_book.tags = list(child_book.tags) + [book_tag]
             child_book.save()
             for fragment in child_book.fragments.all():
                 fragment.tags = set(list(fragment.tags) + [book_tag])
             book_descendants += list(child_book.children.all())
 
             child_book.tags = list(child_book.tags) + [book_tag]
             child_book.save()
             for fragment in child_book.fragments.all():
                 fragment.tags = set(list(fragment.tags) + [book_tag])
             book_descendants += list(child_book.children.all())
 

+        for tag in descendants_tags:
+            touch_tag.delay(tag)
+

         book.save()
 
         # refresh cache
         book.save()
 
         # refresh cache


@@ -951,7 +919,7 @@ class Book(models.Model):
 
         books_by_parent = {}
         books = cls.objects.all().order_by('parent_number', 'sort_key').only(
 
         books_by_parent = {}
         books = cls.objects.all().order_by('parent_number', 'sort_key').only(

-                'title', 'parent', 'slug', 'language')
+                'title', 'parent', 'slug')

         if filter:
             books = books.filter(filter).distinct()
             book_ids = set((book.pk for book in books))
         if filter:
             books = books.filter(filter).distinct()
             book_ids = set((book.pk for book in books))


@@ -1002,7 +970,7 @@ def _has_factory(ftype):
 
     
 # add the file fields
 
     
 # add the file fields

-for t in Book.file_types:
+for t in Book.formats:

     field_name = "%s_file" % t
     models.FileField(_("%s file" % t.upper()),
             upload_to=book_upload_path(t),
     field_name = "%s_file" % t
     models.FileField(_("%s file" % t.upper()),
             upload_to=book_upload_path(t),


@@ -1027,7 +995,7 @@ class Fragment(models.Model):
         verbose_name_plural = _('fragments')
 
     def get_absolute_url(self):
         verbose_name_plural = _('fragments')
 
     def get_absolute_url(self):

-        return '%s#m%s' % (self.book.get_html_url(), self.anchor)
+        return '%s#m%s' % (reverse('book_text', args=[self.book.slug]), self.anchor)

 
     def reset_short_html(self):
         if self.id is None:
 
     def reset_short_html(self):
         if self.id is None:


@@ -1064,7 +1032,8 @@ class Fragment(models.Model):
 def _tags_updated_handler(sender, affected_tags, **kwargs):
     # reset tag global counter
     # we want Tag.changed_at updated for API to know the tag was touched
 def _tags_updated_handler(sender, affected_tags, **kwargs):
     # reset tag global counter
     # we want Tag.changed_at updated for API to know the tag was touched

-    Tag.objects.filter(pk__in=[tag.pk for tag in affected_tags]).update(book_count=None, changed_at=datetime.now())
+    for tag in affected_tags:
+        touch_tag.delay(tag)

 
     # if book tags changed, reset book tag counter
     if isinstance(sender, Book) and \
 
     # if book tags changed, reset book tag counter
     if isinstance(sender, Book) and \