Search - some tests, change in SearchResults cleaning.
[wolnelektury.git] / apps / catalogue / models.py
index 359a89e..fbae111 100644 (file)
@@ -23,14 +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 catalogue.utils import create_zip
-from catalogue.tasks import touch_tag
+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
 
 
+import search
+
+# Those are hard-coded here so that makemessages sees them.
 TAG_CATEGORIES = (
     ('author', _('author')),
     ('epoch', _('epoch')),
@@ -61,7 +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)
-    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)
@@ -175,12 +177,16 @@ class Tag(TagBase):
             try:
                 tag_names = getattr(info, field_name)
             except:
-                tag_names = [getattr(info, category)]
+                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 = ' '.join(tag_name.first_names) + ' ' + 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
@@ -217,7 +223,7 @@ def get_customized_pdf_path(book, 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
@@ -227,7 +233,7 @@ def get_existing_customized_pdf(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))
@@ -266,7 +272,7 @@ class BookMedia(models.Model):
         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):
@@ -275,7 +281,9 @@ class BookMedia(models.Model):
         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())
@@ -344,7 +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)
-    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)
@@ -356,6 +366,8 @@ class Book(models.Model):
     wiki_link     = models.CharField(blank=True, max_length=240)
     # files generated during publication
 
+    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']
 
@@ -367,14 +379,10 @@ class Book(models.Model):
     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:
-        unique_together = [['slug', 'language']]
         ordering = ('sort_key',)
         verbose_name = _('book')
         verbose_name_plural = _('books')
@@ -382,42 +390,6 @@ class Book(models.Model):
     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
 
@@ -432,18 +404,14 @@ class Book(models.Model):
 
     @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):
-        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()
@@ -485,6 +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.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()
@@ -499,24 +468,15 @@ class Book(models.Model):
         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
-            if self.has_media("html"):
-                formats.append(u'<a href="%s">%s</a>' % (reverse('book_text', args=[self.fileid()]), _('Read online')))
             for ebook_format in self.ebook_formats:
                 if self.has_media(ebook_format):
-                    formats.append(u'<a href="%s">%s</a>' % (
-                        self.get_media(ebook_format).url,
-                        ebook_format.upper()
-                    ))
-            # 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[ebook_format] = self.get_media(ebook_format)
 
-            formats = [mark_safe(format) for format in formats]
 
             short_html = unicode(render_to_string('catalogue/book_short.html',
                 {'book': self, 'tags': tags, 'formats': formats}))
@@ -527,7 +487,7 @@ class Book(models.Model):
 
     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
@@ -564,13 +524,27 @@ class Book(models.Model):
     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)
 
+    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.
@@ -586,7 +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)
-            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
 
@@ -609,7 +583,7 @@ class Book(models.Model):
 
         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)
@@ -621,7 +595,7 @@ class Book(models.Model):
 
         epub = self.wldocument().as_epub()
 
-        self.epub_file.save('%s.epub' % self.fileid(),
+        self.epub_file.save('%s.epub' % self.slug,
                 File(open(epub.get_filename())))
 
         # remove zip package with all epub files
@@ -631,7 +605,7 @@ class Book(models.Model):
         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):
@@ -646,7 +620,7 @@ class Book(models.Model):
 
         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
@@ -697,7 +671,7 @@ class Book(models.Model):
         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_
@@ -708,12 +682,25 @@ class Book(models.Model):
                     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)
-        result = create_zip.delay(paths, self.fileid())
+        result = create_zip.delay(paths, "%s_%s" % (self.slug, format_))
         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
@@ -732,7 +719,8 @@ class Book(models.Model):
 
     @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
         from sortify import sortify
 
@@ -741,30 +729,33 @@ class Book(models.Model):
         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:
-                    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
-        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')
-        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:
-                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'))
 
+        book.language = book_info.language
         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()
 
@@ -789,6 +780,8 @@ class Book(models.Model):
             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:
             book.build_epub()
 
@@ -798,6 +791,9 @@ class Book(models.Model):
         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())
         descendants_tags = set()
         # add l-tag to descendants and their fragments
@@ -923,7 +919,7 @@ class Book(models.Model):
 
         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))
@@ -999,7 +995,7 @@ class Fragment(models.Model):
         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: