fnp / wolnelektury.git / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
fix
[wolnelektury.git] / src / catalogue / models / book.py
diff --git a/src/catalogue/models/book.py b/src/catalogue/models/book.py
index 1524c34..35d0e8c 100644 (file)
--- a/src/catalogue/models/book.py
+++ b/src/catalogue/models/book.py
@@ -1,134 +1,292 @@
-# -*- coding: utf-8 -*-
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from collections import OrderedDict
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from collections import OrderedDict
+import json
+from datetime import date, timedelta
 from random import randint
 from random import randint
+import os.path
 import re
 import re
+from urllib.request import urlretrieve
+from django.apps import apps
 from django.conf import settings
 from django.db import connection, models, transaction
 from django.conf import settings
 from django.db import connection, models, transaction
-from django.db.models import permalink
 import django.dispatch
 from django.contrib.contenttypes.fields import GenericRelation
 import django.dispatch
 from django.contrib.contenttypes.fields import GenericRelation
-from django.core.urlresolvers import reverse
-from django.utils.translation import ugettext_lazy as _
-import jsonfield
+from django.template.loader import render_to_string
+from django.urls import reverse
+from django.utils.translation import gettext_lazy as _, get_language
 from fnpdjango.storage import BofhFileSystemStorage
 from fnpdjango.storage import BofhFileSystemStorage
-from ssify import flush_ssi_includes
+from lxml import html
+from librarian.cover import WLCover
+from librarian.html import transform_abstrakt
+from librarian.builders import builders
 from newtagging import managers
 from catalogue import constants
 from newtagging import managers
 from catalogue import constants
-from catalogue.fields import EbookField
+from catalogue import fields
 from catalogue.models import Tag, Fragment, BookMedia
 from catalogue.models import Tag, Fragment, BookMedia
-from catalogue.utils import create_zip
+from catalogue.utils import create_zip, gallery_url, gallery_path, split_tags, get_random_hash
+from catalogue.models.tag import prefetched_relations
 from catalogue import app_settings
 from catalogue import app_settings
-from catalogue import tasks
+from wolnelektury.utils import makedirs, cached_render, clear_cached_renders
 
 bofh_storage = BofhFileSystemStorage()
 
 
 
 bofh_storage = BofhFileSystemStorage()
 
 
-def _cover_upload_to(i, n):
-    return 'book/cover/%s.jpg' % i.slug
-
-def _cover_thumb_upload_to(i, n):
-    return 'book/cover_thumb/%s.jpg' % i.slug
-
-def _ebook_upload_to(upload_path):
-    def _upload_to(i, n):
-        return upload_path % i.slug
-    return _upload_to
-
-
 class Book(models.Model):
     """Represents a book imported from WL-XML."""
 class Book(models.Model):
     """Represents a book imported from WL-XML."""
-    title         = models.CharField(_('title'), max_length=32767)
-    sort_key = models.CharField(_('sort key'), max_length=120, db_index=True, editable=False)
-    sort_key_author = models.CharField(_('sort key by author'), max_length=120, db_index=True, editable=False, default=u'')
-    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=app_settings.DEFAULT_LANGUAGE)
-    description   = models.TextField(_('description'), blank=True)
-    created_at    = models.DateTimeField(_('creation date'), auto_now_add=True, db_index=True)
-    changed_at    = models.DateTimeField(_('creation date'), auto_now=True, db_index=True)
-    parent_number = models.IntegerField(_('parent number'), default=0)
-    extra_info    = jsonfield.JSONField(_('extra information'), default={})
-    gazeta_link   = models.CharField(blank=True, max_length=240)
-    wiki_link     = models.CharField(blank=True, max_length=240)
+    title = models.CharField('tytuł', max_length=32767)
+    sort_key = models.CharField('klucz sortowania', max_length=120, db_index=True, editable=False)
+    sort_key_author = models.CharField(
+        'klucz sortowania wg autora', max_length=120, db_index=True, editable=False, default='')
+    slug = models.SlugField('slug', max_length=120, db_index=True, unique=True)
+    common_slug = models.SlugField('wspólny slug', max_length=120, db_index=True)
+    language = models.CharField('kod języka', max_length=3, db_index=True, default=app_settings.DEFAULT_LANGUAGE)
+    description = models.TextField('opis', blank=True)
+    abstract = models.TextField('abstrakt', blank=True)
+    toc = models.TextField('spis treści', blank=True)
+    created_at = models.DateTimeField('data utworzenia', auto_now_add=True, db_index=True)
+    changed_at = models.DateTimeField('data motyfikacji', auto_now=True, db_index=True)
+    parent_number = models.IntegerField('numer w ramach rodzica', default=0)
+    extra_info = models.TextField('dodatkowe informacje', default='{}')
+    gazeta_link = models.CharField(blank=True, max_length=240)
+    wiki_link = models.CharField(blank=True, max_length=240)
+    print_on_demand = models.BooleanField('druk na żądanie', default=False)
+    recommended = models.BooleanField('polecane', default=False)
+    audio_length = models.CharField('długość audio', blank=True, max_length=8)
+    preview = models.BooleanField('prapremiera', default=False)
+    preview_until = models.DateField('prapremiera do', blank=True, null=True)
+    preview_key = models.CharField(max_length=32, blank=True, null=True)
+    findable = models.BooleanField('wyszukiwalna', default=True, db_index=True)
 
     # files generated during publication
 
     # files generated during publication
-    cover = EbookField('cover', _('cover'),
-            null=True, blank=True,
-            upload_to=_cover_upload_to,
-            storage=bofh_storage, max_length=255)
+    xml_file = fields.XmlField(storage=bofh_storage, with_etag=False)
+    html_file = fields.HtmlField(storage=bofh_storage)
+    fb2_file = fields.Fb2Field(storage=bofh_storage)
+    txt_file = fields.TxtField(storage=bofh_storage)
+    epub_file = fields.EpubField(storage=bofh_storage)
+    mobi_file = fields.MobiField(storage=bofh_storage)
+    pdf_file = fields.PdfField(storage=bofh_storage)
+
+    cover = fields.CoverField('okładka', storage=bofh_storage)
     # Cleaner version of cover for thumbs
     # Cleaner version of cover for thumbs
-    cover_thumb = EbookField('cover_thumb', _('cover thumbnail'),
-            null=True, blank=True,
-            upload_to=_cover_thumb_upload_to,
-            max_length=255)
+    cover_clean = fields.CoverCleanField('czysta okładka')
+    cover_thumb = fields.CoverThumbField('miniatura okładki')
+    cover_api_thumb = fields.CoverApiThumbField(
+        'mniaturka okładki dla aplikacji')
+    simple_cover = fields.SimpleCoverField('okładka dla aplikacji')
+    cover_ebookpoint = fields.CoverEbookpointField(
+        'okładka dla Ebookpoint')
+
     ebook_formats = constants.EBOOK_FORMATS
     formats = ebook_formats + ['html', 'xml']
 
     ebook_formats = constants.EBOOK_FORMATS
     formats = ebook_formats + ['html', 'xml']
 
-    parent = models.ForeignKey('self', blank=True, null=True,
-        related_name='children')
-    ancestor = models.ManyToManyField('self', blank=True,
-        editable=False, related_name='descendant', symmetrical=False)
+    parent = models.ForeignKey('self', models.CASCADE, blank=True, null=True, related_name='children')
+    ancestor = models.ManyToManyField('self', blank=True, editable=False, related_name='descendant', symmetrical=False)
+
+    cached_author = models.CharField(blank=True, max_length=240, db_index=True)
+    has_audience = models.BooleanField(default=False)
 
 
-    objects  = models.Manager()
-    tagged   = managers.ModelTaggedItemManager(Tag)
-    tags     = managers.TagDescriptor(Tag)
+    objects = models.Manager()
+    tagged = managers.ModelTaggedItemManager(Tag)
+    tags = managers.TagDescriptor(Tag)
     tag_relations = GenericRelation(Tag.intermediary_table_model)
 
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
 
     tag_relations = GenericRelation(Tag.intermediary_table_model)
 
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
 
-    short_html_url_name = 'catalogue_book_short'
+    SORT_KEY_SEP = '$'
+
+    is_book = True
 
     class AlreadyExists(Exception):
         pass
 
     class Meta:
 
     class AlreadyExists(Exception):
         pass
 
     class Meta:
-        ordering = ('sort_key',)
-        verbose_name = _('book')
-        verbose_name_plural = _('books')
+        ordering = ('sort_key_author', 'sort_key')
+        verbose_name = 'książka'
+        verbose_name_plural = 'książki'
         app_label = 'catalogue'
 
         app_label = 'catalogue'
 
-    def __unicode__(self):
+    def __str__(self):
         return self.title
 
         return self.title
 
+    def get_extra_info_json(self):
+        return json.loads(self.extra_info or '{}')
+
     def get_initial(self):
         try:
             return re.search(r'\w', self.title, re.U).group(0)
         except AttributeError:
     def get_initial(self):
         try:
             return re.search(r'\w', self.title, re.U).group(0)
         except AttributeError:
+            return ''
+
+    def authors(self):
+        return self.tags.filter(category='author')
+
+    def epochs(self):
+        return self.tags.filter(category='epoch')
+
+    def genres(self):
+        return self.tags.filter(category='genre')
+
+    def kinds(self):
+        return self.tags.filter(category='kind')
+
+    def tag_unicode(self, category):
+        relations = prefetched_relations(self, category)
+        if relations:
+            return ', '.join(rel.tag.name for rel in relations)
+        else:
+            return ', '.join(self.tags.filter(category=category).values_list('name', flat=True))
+
+    def tags_by_category(self):
+        return split_tags(self.tags.exclude(category__in=('set', 'theme')))
+
+    def author_unicode(self):
+        return self.cached_author
+
+    def kind_unicode(self):
+        return self.tag_unicode('kind')
+
+    def epoch_unicode(self):
+        return self.tag_unicode('epoch')
+
+    def genre_unicode(self):
+        return self.tag_unicode('genre')
+
+    def translators(self):
+        translators = self.get_extra_info_json().get('translators') or []
+        return [
+            '\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators
+        ]
+
+    def translator(self):
+        translators = self.get_extra_info_json().get('translators')
+        if not translators:
             return None
             return None
+        if len(translators) > 3:
+            translators = translators[:2]
+            others = ' i inni'
+        else:
+            others = ''
+        return ', '.join('\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators) + others
+
+    def cover_source(self):
+        return self.get_extra_info_json().get('cover_source', self.parent.cover_source() if self.parent else '')
 
 
-    def author_str(self):
-        return ", ".join(str(t) for t in self.tags.filter(category='author'))
+    @property
+    def isbn_pdf(self):
+        return self.get_extra_info_json().get('isbn_pdf')
+
+    @property
+    def isbn_epub(self):
+        return self.get_extra_info_json().get('isbn_epub')
+
+    @property
+    def isbn_mobi(self):
+        return self.get_extra_info_json().get('isbn_mobi')
+
+    def is_accessible_to(self, user):
+        if not self.preview:
+            return True
+        if not user.is_authenticated:
+            return False
+        Membership = apps.get_model('club', 'Membership')
+        if Membership.is_active_for(user):
+            return True
+        Funding = apps.get_model('funding', 'Funding')
+        if Funding.objects.filter(user=user, offer__book=self):
+            return True
+        return False
 
     def save(self, force_insert=False, force_update=False, **kwargs):
         from sortify import sortify
 
         self.sort_key = sortify(self.title)[:120]
 
     def save(self, force_insert=False, force_update=False, **kwargs):
         from sortify import sortify
 
         self.sort_key = sortify(self.title)[:120]
-        self.title = unicode(self.title) # ???
+        self.title = str(self.title)  # ???
 
         try:
 
         try:
-            author = self.tags.filter(category='author')[0].sort_key
-        except IndexError:
-            author = u''
+            author = self.authors().first().sort_key
+        except AttributeError:
+            author = ''
         self.sort_key_author = author
 
         self.sort_key_author = author
 
+        self.cached_author = self.tag_unicode('author')
+        self.has_audience = 'audience' in self.get_extra_info_json()
+
+        if self.preview and not self.preview_key:
+            self.preview_key = get_random_hash(self.slug)[:32]
+
         ret = super(Book, self).save(force_insert, force_update, **kwargs)
 
         return ret
 
         ret = super(Book, self).save(force_insert, force_update, **kwargs)
 
         return ret
 
-    @permalink
     def get_absolute_url(self):
     def get_absolute_url(self):
-        return ('catalogue.views.book_detail', [self.slug])
+        return reverse('book_detail', args=[self.slug])
+
+    def gallery_path(self):
+        return gallery_path(self.slug)
+
+    def gallery_url(self):
+        return gallery_url(self.slug)
+
+    def get_first_text(self):
+        if self.html_file:
+            return self
+        child = self.children.all().order_by('parent_number').first()
+        if child is not None:
+            return child.get_first_text()
+
+    def get_last_text(self):
+        if self.html_file:
+            return self
+        child = self.children.all().order_by('parent_number').last()
+        if child is not None:
+            return child.get_last_text()
+
+    def get_prev_text(self):
+        if not self.parent:
+            return None
+        sibling = self.parent.children.filter(parent_number__lt=self.parent_number).order_by('-parent_number').first()
+        if sibling is not None:
+            return sibling.get_last_text()
 
 
-    @staticmethod
-    @permalink
-    def create_url(slug):
-        return ('catalogue.views.book_detail', [slug])
+        if self.parent.html_file:
+            return self.parent
+
+        return self.parent.get_prev_text()
+
+    def get_next_text(self, inside=True):
+        if inside:
+            child = self.children.order_by('parent_number').first()
+            if child is not None:
+                return child.get_first_text()
+
+        if not self.parent:
+            return None
+        sibling = self.parent.children.filter(parent_number__gt=self.parent_number).order_by('parent_number').first()
+        if sibling is not None:
+            return sibling.get_first_text()
+        return self.parent.get_next_text(inside=False)
+
+    def get_child_audiobook(self):
+        BookMedia = apps.get_model('catalogue', 'BookMedia')
+        if not BookMedia.objects.filter(book__ancestor=self).exists():
+            return None
+        for child in self.children.order_by('parent_number').all():
+            if child.has_mp3_file():
+                return child
+            child_sub = child.get_child_audiobook()
+            if child_sub is not None:
+                return child_sub
+
+    def get_siblings(self):
+        if not self.parent:
+            return []
+        return self.parent.children.all().order_by('parent_number')
+
+    def get_children(self):
+        return self.children.all().order_by('parent_number')
 
     @property
     def name(self):
 
     @property
     def name(self):
@@ -140,12 +298,48 @@ class Book(models.Model):
     def language_name(self):
         return dict(settings.LANGUAGES).get(self.language_code(), "")
 
     def language_name(self):
         return dict(settings.LANGUAGES).get(self.language_code(), "")
 
+    def is_foreign(self):
+        return self.language_code() != settings.LANGUAGE_CODE
+
+    def set_audio_length(self):
+        length = self.get_audio_length()
+        if length > 0:
+            self.audio_length = self.format_audio_length(length)
+            self.save()
+
+    @staticmethod
+    def format_audio_length(seconds):
+        """
+        >>> Book.format_audio_length(1)
+        '0:01'
+        >>> Book.format_audio_length(3661)
+        '1:01:01'
+        """
+        if seconds < 60*60:
+            minutes = seconds // 60
+            seconds = seconds % 60
+            return '%d:%02d' % (minutes, seconds)
+        else:
+            hours = seconds // 3600
+            minutes = seconds % 3600 // 60
+            seconds = seconds % 60
+            return '%d:%02d:%02d' % (hours, minutes, seconds)
+
+    def get_audio_length(self):
+        total = 0
+        for media in self.get_mp3() or ():
+            total += app_settings.GET_MP3_LENGTH(media.file.path)
+        return int(total)
+
     def has_media(self, type_):
         if type_ in Book.formats:
             return bool(getattr(self, "%s_file" % type_))
         else:
             return self.media.filter(type=type_).exists()
 
     def has_media(self, type_):
         if type_ in Book.formats:
             return bool(getattr(self, "%s_file" % type_))
         else:
             return self.media.filter(type=type_).exists()
 
+    def has_audio(self):
+        return self.has_media('mp3')
+
     def get_media(self, type_):
         if self.has_media(type_):
             if type_ in Book.formats:
     def get_media(self, type_):
         if self.has_media(type_):
             if type_ in Book.formats:
@@ -157,34 +351,129 @@ class Book(models.Model):
 
     def get_mp3(self):
         return self.get_media("mp3")
 
     def get_mp3(self):
         return self.get_media("mp3")
+
     def get_odt(self):
         return self.get_media("odt")
     def get_odt(self):
         return self.get_media("odt")
+
     def get_ogg(self):
         return self.get_media("ogg")
     def get_ogg(self):
         return self.get_media("ogg")
+
     def get_daisy(self):
         return self.get_media("daisy")
 
     def get_daisy(self):
         return self.get_media("daisy")
 
+    def get_audio_epub(self):
+        return self.get_media("audio.epub")
+
+    def media_url(self, format_):
+        media = self.get_media(format_)
+        if media:
+            if self.preview:
+                return reverse('embargo_link', kwargs={'key': self.preview_key, 'slug': self.slug, 'format_': format_})
+            else:
+                return media.url
+        else:
+            return None
+
+    def html_url(self):
+        return self.media_url('html')
+
+    def pdf_url(self):
+        return self.media_url('pdf')
+
+    def epub_url(self):
+        return self.media_url('epub')
+
+    def mobi_url(self):
+        return self.media_url('mobi')
+
+    def txt_url(self):
+        return self.media_url('txt')
+
+    def fb2_url(self):
+        return self.media_url('fb2')
+
+    def xml_url(self):
+        return self.media_url('xml')
+
     def has_description(self):
         return len(self.description) > 0
     def has_description(self):
         return len(self.description) > 0
-    has_description.short_description = _('description')
+    has_description.short_description = 'opis'
     has_description.boolean = True
 
     has_description.boolean = True
 
-    # ugly ugly ugly
     def has_mp3_file(self):
     def has_mp3_file(self):
-        return bool(self.has_media("mp3"))
+        return self.has_media("mp3")
     has_mp3_file.short_description = 'MP3'
     has_mp3_file.boolean = True
 
     def has_ogg_file(self):
     has_mp3_file.short_description = 'MP3'
     has_mp3_file.boolean = True
 
     def has_ogg_file(self):
-        return bool(self.has_media("ogg"))
+        return self.has_media("ogg")
     has_ogg_file.short_description = 'OGG'
     has_ogg_file.boolean = True
 
     def has_daisy_file(self):
     has_ogg_file.short_description = 'OGG'
     has_ogg_file.boolean = True
 
     def has_daisy_file(self):
-        return bool(self.has_media("daisy"))
+        return self.has_media("daisy")
     has_daisy_file.short_description = 'DAISY'
     has_daisy_file.boolean = True
 
     has_daisy_file.short_description = 'DAISY'
     has_daisy_file.boolean = True
 
+    def has_sync_file(self):
+        return self.has_media("sync")
+
+    def get_sync(self):
+        with self.get_media('sync').first().file.open('r') as f:
+            sync = f.read().split('\n')
+        offset = float(sync[0])
+        items = []
+        for line in sync[1:]:
+            if not line:
+                continue
+            start, end, elid = line.split()
+            items.append([elid, float(start) + offset])
+        return json.dumps(items)
+    
+    def has_audio_epub_file(self):
+        return self.has_media("audio.epub")
+
+    @property
+    def media_daisy(self):
+        return self.get_media('daisy')
+
+    @property
+    def media_audio_epub(self):
+        return self.get_media('audio.epub')
+
+    def get_audiobooks(self):
+        ogg_files = {}
+        for m in self.media.filter(type='ogg').order_by().iterator():
+            ogg_files[m.name] = m
+
+        audiobooks = []
+        projects = set()
+        total_duration = 0
+        for mp3 in self.media.filter(type='mp3').iterator():
+            # ogg files are always from the same project
+            meta = mp3.get_extra_info_json()
+            project = meta.get('project')
+            if not project:
+                # temporary fallback
+                project = 'CzytamySłuchając'
+
+            projects.add((project, meta.get('funded_by', '')))
+            total_duration += mp3.duration or 0
+
+            media = {'mp3': mp3}
+
+            ogg = ogg_files.get(mp3.name)
+            if ogg:
+                media['ogg'] = ogg
+            audiobooks.append(media)
+
+        projects = sorted(projects)
+        total_duration = '%d:%02d' % (
+            total_duration // 60,
+            total_duration % 60
+        )
+        return audiobooks, projects, total_duration
+
     def wldocument(self, parse_dublincore=True, inherit=True):
         from catalogue.import_utils import ORMDocProvider
         from librarian.parser import WLDocument
     def wldocument(self, parse_dublincore=True, inherit=True):
         from catalogue.import_utils import ORMDocProvider
         from librarian.parser import WLDocument
@@ -194,44 +483,98 @@ class Book(models.Model):
         else:
             meta_fallbacks = None
 
         else:
             meta_fallbacks = None
 
-        return WLDocument.from_file(self.xml_file.path,
-                provider=ORMDocProvider(self),
-                parse_dublincore=parse_dublincore,
-                meta_fallbacks=meta_fallbacks)
+        return WLDocument.from_file(
+            self.xml_file.path,
+            provider=ORMDocProvider(self),
+            parse_dublincore=parse_dublincore,
+            meta_fallbacks=meta_fallbacks)
+
+    def wldocument2(self):
+        from catalogue.import_utils import ORMDocProvider
+        from librarian.document import WLDocument
+        doc = WLDocument(
+            self.xml_file.path,
+            provider=ORMDocProvider(self)
+        )
+        doc.meta.update(self.cover_info())
+        return doc
+
 
     @staticmethod
     def zip_format(format_):
         def pretty_file_name(book):
             return "%s/%s.%s" % (
 
     @staticmethod
     def zip_format(format_):
         def pretty_file_name(book):
             return "%s/%s.%s" % (
-                book.extra_info['author'],
+                book.get_extra_info_json()['author'],
                 book.slug,
                 format_)
 
         field_name = "%s_file" % format_
                 book.slug,
                 format_)
 
         field_name = "%s_file" % format_
-        books = Book.objects.filter(parent=None).exclude(**{field_name: ""})
-        paths = [(pretty_file_name(b), getattr(b, field_name).path)
-                    for b in books.iterator()]
-        return create_zip(paths, app_settings.FORMAT_ZIPS[format_])
+        field = getattr(Book, field_name)
+        books = Book.objects.filter(parent=None).exclude(**{field_name: ""}).exclude(preview=True).exclude(findable=False)
+        paths = [(pretty_file_name(b), getattr(b, field_name).path) for b in books.iterator()]
+        return create_zip(paths, field.ZIP)
 
     def zip_audiobooks(self, format_):
         bm = BookMedia.objects.filter(book=self, type=format_)
 
     def zip_audiobooks(self, format_):
         bm = BookMedia.objects.filter(book=self, type=format_)
-        paths = map(lambda bm: (None, bm.file.path), bm)
-        return create_zip(paths, "%s_%s" % (self.slug, format_))
-
-    def search_index(self, book_info=None, index=None, index_tags=True, commit=True):
-        if index is None:
-            from search.index import Index
-            index = Index()
-        try:
-            index.index_book(self, book_info)
-            if index_tags:
-                index.index_tags()
-            if commit:
-                index.index.commit()
-        except Exception, e:
-            index.index.rollback()
-            raise e
-
+        paths = map(lambda bm: (bm.get_nice_filename(), bm.file.path), bm)
+        licenses = set()
+        for m in bm:
+            license = constants.LICENSES.get(
+                m.get_extra_info_json().get('license'), {}
+            ).get('locative')
+            if license:
+                licenses.add(license)
+        readme = render_to_string('catalogue/audiobook_zip_readme.txt', {
+            'licenses': licenses,
+            'meta': self.wldocument2().meta,
+        })
+        return create_zip(paths, "%s_%s" % (self.slug, format_), {'informacje.txt': readme})
+
+    def search_index(self, index=None):
+        if not self.findable:
+            return
+        from search.index import Index
+        Index.index_book(self)
+
+    # will make problems in conjunction with paid previews
+    def download_pictures(self, remote_gallery_url):
+        # This is only needed for legacy relative image paths.
+        gallery_path = self.gallery_path()
+        # delete previous files, so we don't include old files in ebooks
+        if os.path.isdir(gallery_path):
+            for filename in os.listdir(gallery_path):
+                file_path = os.path.join(gallery_path, filename)
+                os.unlink(file_path)
+        ilustr_elements = list(self.wldocument().edoc.findall('//ilustr'))
+        if ilustr_elements:
+            makedirs(gallery_path)
+            for ilustr in ilustr_elements:
+                ilustr_src = ilustr.get('src')
+                if '/' in ilustr_src:
+                    continue
+                ilustr_path = os.path.join(gallery_path, ilustr_src)
+                urlretrieve('%s/%s' % (remote_gallery_url, ilustr_src), ilustr_path)
+
+    def load_abstract(self):
+        abstract = self.wldocument(parse_dublincore=False).edoc.getroot().find('.//abstrakt')
+        if abstract is not None:
+            self.abstract = transform_abstrakt(abstract)
+        else:
+            self.abstract = ''
+
+    def load_toc(self):
+        self.toc = ''
+        if self.html_file:
+            parser = html.HTMLParser(encoding='utf-8')
+            tree = html.parse(self.html_file.path, parser=parser)
+            toc = tree.find('//div[@id="toc"]/ol')
+            if toc is None or not len(toc):
+                return
+            html_link = reverse('book_text', args=[self.slug])
+            for a in toc.findall('.//a'):
+                a.attrib['href'] = html_link + a.attrib['href']
+            self.toc = html.tostring(toc, encoding='unicode')
+            # div#toc
 
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
 
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
@@ -250,9 +593,10 @@ class Book(models.Model):
             xml_file.close()
 
     @classmethod
             xml_file.close()
 
     @classmethod
-    def from_text_and_meta(cls, raw_file, book_info, overwrite=False,
-            dont_build=None, search_index=True,
-            search_index_tags=True):
+    def from_text_and_meta(cls, raw_file, book_info, overwrite=False, dont_build=None, search_index=True,
+                           remote_gallery_url=None, days=0, findable=True):
+        from catalogue import tasks
+
         if dont_build is None:
             dont_build = set()
         dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
         if dont_build is None:
             dont_build = set()
         dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
@@ -264,8 +608,7 @@ class Book(models.Model):
                 try:
                     children.append(Book.objects.get(slug=part_url.slug))
                 except Book.DoesNotExist:
                 try:
                     children.append(Book.objects.get(slug=part_url.slug))
                 except Book.DoesNotExist:
-                    raise Book.DoesNotExist(_('Book "%s" does not exist.') %
-                            part_url.slug)
+                    raise Book.DoesNotExist('Książka "%s" nie istnieje.' % part_url.slug)
 
         # Read book metadata
         book_slug = book_info.url.slug
 
         # Read book metadata
         book_slug = book_info.url.slug
@@ -276,29 +619,42 @@ class Book(models.Model):
         if created:
             book_shelves = []
             old_cover = None
         if created:
             book_shelves = []
             old_cover = None
+            book.preview = bool(days)
+            if book.preview:
+                book.preview_until = date.today() + timedelta(days)
         else:
             if not overwrite:
         else:
             if not overwrite:
-                raise Book.AlreadyExists(_('Book %s already exists') % (
-                        book_slug))
+                raise Book.AlreadyExists('Książka %s już istnieje' % book_slug)
             # Save shelves for this book
             book_shelves = list(book.tags.filter(category='set'))
             old_cover = book.cover_info()
 
         # Save XML file
         book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
             # Save shelves for this book
             book_shelves = list(book.tags.filter(category='set'))
             old_cover = book.cover_info()
 
         # Save XML file
         book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
+        if book.preview:
+            book.xml_file.set_readable(False)
 
 
+        book.findable = findable
         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.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.extra_info = book_info.to_dict()
+        book.extra_info = json.dumps(book_info.to_dict())
+        book.load_abstract()
+        book.load_toc()
         book.save()
 
         meta_tags = Tag.tags_from_info(book_info)
 
         book.save()
 
         meta_tags = Tag.tags_from_info(book_info)
 
+        for tag in meta_tags:
+            if not tag.for_books:
+                tag.for_books = True
+                tag.save()
+
         book.tags = set(meta_tags + book_shelves)
         book.tags = set(meta_tags + book_shelves)
+        book.save()  # update sort_key_author
 
         cover_changed = old_cover != book.cover_info()
         obsolete_children = set(b for b in book.children.all()
 
         cover_changed = old_cover != book.cover_info()
         obsolete_children = set(b for b in book.children.all()
@@ -322,12 +678,19 @@ class Book(models.Model):
         cls.repopulate_ancestors()
         tasks.update_counters.delay()
 
         cls.repopulate_ancestors()
         tasks.update_counters.delay()
 
+        if remote_gallery_url:
+            book.download_pictures(remote_gallery_url)
+
         # No saves beyond this point.
 
         # Build cover.
         if 'cover' not in dont_build:
             book.cover.build_delay()
         # No saves beyond this point.
 
         # Build cover.
         if 'cover' not in dont_build:
             book.cover.build_delay()
+            book.cover_clean.build_delay()
             book.cover_thumb.build_delay()
             book.cover_thumb.build_delay()
+            book.cover_api_thumb.build_delay()
+            book.simple_cover.build_delay()
+            book.cover_ebookpoint.build_delay()
 
         # Build HTML and ebooks.
         book.html_file.build_delay()
 
         # Build HTML and ebooks.
         book.html_file.build_delay()
@@ -339,65 +702,104 @@ class Book(models.Model):
             if format_ not in dont_build:
                 getattr(book, '%s_file' % format_).build_delay()
 
             if format_ not in dont_build:
                 getattr(book, '%s_file' % format_).build_delay()
 
-        if not settings.NO_SEARCH_INDEX and search_index:
-            tasks.index_book.delay(book.id, book_info=book_info, index_tags=search_index_tags)
+        if not settings.NO_SEARCH_INDEX and search_index and findable:
+            tasks.index_book.delay(book.id)
 
         for child in notify_cover_changed:
             child.parent_cover_changed()
 
 
         for child in notify_cover_changed:
             child.parent_cover_changed()
 
+        book.update_popularity()
+        tasks.update_references.delay(book.id)
+
         cls.published.send(sender=cls, instance=book)
         return book
 
         cls.published.send(sender=cls, instance=book)
         return book
 
+    def update_references(self):
+        Entity = apps.get_model('references', 'Entity')
+        doc = self.wldocument2()
+        doc._compat_assign_section_ids()
+        doc._compat_assign_ordered_ids()
+        refs = {}
+        for ref_elem in doc.references():
+            uri = ref_elem.attrib.get('href', '')
+            if not uri:
+                continue
+            if uri in refs:
+                ref = refs[uri]
+            else:
+                entity, entity_created = Entity.objects.get_or_create(uri=uri)
+                if entity_created:
+                    entity.populate()
+                    entity.save()
+                ref, ref_created = entity.reference_set.get_or_create(book=self)
+                refs[uri] = ref
+                if not ref_created:
+                    ref.occurence_set.all().delete()
+            sec = ref_elem.get_link()
+            m = re.match(r'sec(\d+)', sec)
+            assert m is not None
+            sec = int(m.group(1))
+            snippet = ref_elem.get_snippet()
+            b = builders['html-snippet']()
+            for s in snippet:
+                s.html_build(b)
+            html = b.output().get_bytes().decode('utf-8')
+
+            ref.occurence_set.create(
+                section=sec,
+                html=html
+            )
+        self.reference_set.exclude(entity__uri__in=refs).delete()
+
+    @property
+    def references(self):
+        return self.reference_set.all().select_related('entity')
+
     @classmethod
     @classmethod
+    @transaction.atomic
     def repopulate_ancestors(cls):
         """Fixes the ancestry cache."""
         # TODO: table names
     def repopulate_ancestors(cls):
         """Fixes the ancestry cache."""
         # TODO: table names
-        with transaction.atomic():
-            cursor = connection.cursor()
-            if connection.vendor == 'postgres':
-                cursor.execute("TRUNCATE catalogue_book_ancestor")
-                cursor.execute("""
-                    WITH RECURSIVE ancestry AS (
-                        SELECT book.id, book.parent_id
-                        FROM catalogue_book AS book
-                        WHERE book.parent_id IS NOT NULL
-                        UNION
-                        SELECT ancestor.id, book.parent_id
-                        FROM ancestry AS ancestor, catalogue_book AS book
-                        WHERE ancestor.parent_id = book.id
-                            AND book.parent_id IS NOT NULL
-                        )
-                    INSERT INTO catalogue_book_ancestor
-                        (from_book_id, to_book_id)
-                        SELECT id, parent_id
-                        FROM ancestry
-                        ORDER BY id;
-                    """)
-            else:
-                cursor.execute("DELETE FROM catalogue_book_ancestor")
-                for b in cls.objects.exclude(parent=None):
-                    parent = b.parent
-                    while parent is not None:
-                        b.ancestor.add(parent)
-                        parent = parent.parent
-
-    def flush_includes(self, languages=True):
-        if not languages:
-            return
-        if languages is True:
-            languages = [lc for (lc, _ln) in settings.LANGUAGES]
-        flush_ssi_includes([
-            template % (self.pk, lang)
-            for template in [
-                '/katalog/b/%d/mini.%s.html',
-                '/katalog/b/%d/mini_nolink.%s.html',
-                '/katalog/b/%d/short.%s.html',
-                '/katalog/b/%d/wide.%s.html',
-                '/api/include/book/%d.%s.json',
-                '/api/include/book/%d.%s.xml',
-                ]
-            for lang in languages
-            ])
+        cursor = connection.cursor()
+        if connection.vendor == 'postgres':
+            cursor.execute("TRUNCATE catalogue_book_ancestor")
+            cursor.execute("""
+                WITH RECURSIVE ancestry AS (
+                    SELECT book.id, book.parent_id
+                    FROM catalogue_book AS book
+                    WHERE book.parent_id IS NOT NULL
+                    UNION
+                    SELECT ancestor.id, book.parent_id
+                    FROM ancestry AS ancestor, catalogue_book AS book
+                    WHERE ancestor.parent_id = book.id
+                        AND book.parent_id IS NOT NULL
+                    )
+                INSERT INTO catalogue_book_ancestor
+                    (from_book_id, to_book_id)
+                    SELECT id, parent_id
+                    FROM ancestry
+                    ORDER BY id;
+                """)
+        else:
+            cursor.execute("DELETE FROM catalogue_book_ancestor")
+            for b in cls.objects.exclude(parent=None):
+                parent = b.parent
+                while parent is not None:
+                    b.ancestor.add(parent)
+                    parent = parent.parent
+
+    @property
+    def ancestors(self):
+        if self.parent:
+            for anc in self.parent.ancestors:
+                yield anc
+            yield self.parent
+        else:
+            return []
+
+    def clear_cache(self):
+        clear_cached_renders(self.mini_box)
+        clear_cached_renders(self.mini_box_nolink)
 
     def cover_info(self, inherit=True):
         """Returns a dictionary to serve as fallback for BookInfo.
 
     def cover_info(self, inherit=True):
         """Returns a dictionary to serve as fallback for BookInfo.
@@ -407,7 +809,7 @@ class Book(models.Model):
         need = False
         info = {}
         for field in ('cover_url', 'cover_by', 'cover_source'):
         need = False
         info = {}
         for field in ('cover_url', 'cover_by', 'cover_source'):
-            val = self.extra_info.get(field)
+            val = self.get_extra_info_json().get(field)
             if val:
                 info[field] = val
             else:
             if val:
                 info[field] = val
             else:
@@ -421,14 +823,18 @@ class Book(models.Model):
     def related_themes(self):
         return Tag.objects.usage_for_queryset(
             Fragment.objects.filter(models.Q(book=self) | models.Q(book__ancestor=self)),
     def related_themes(self):
         return Tag.objects.usage_for_queryset(
             Fragment.objects.filter(models.Q(book=self) | models.Q(book__ancestor=self)),
-            counts=True).filter(category='theme')
+            counts=True).filter(category='theme').order_by('-count')
 
     def parent_cover_changed(self):
         """Called when parent book's cover image is changed."""
         if not self.cover_info(inherit=False):
             if 'cover' not in app_settings.DONT_BUILD:
                 self.cover.build_delay()
 
     def parent_cover_changed(self):
         """Called when parent book's cover image is changed."""
         if not self.cover_info(inherit=False):
             if 'cover' not in app_settings.DONT_BUILD:
                 self.cover.build_delay()
+                self.cover_clean.build_delay()
                 self.cover_thumb.build_delay()
                 self.cover_thumb.build_delay()
+                self.cover_api_thumb.build_delay()
+                self.simple_cover.build_delay()
+                self.cover_ebookpoint.build_delay()
             for format_ in constants.EBOOK_FORMATS_WITH_COVERS:
                 if format_ not in app_settings.DONT_BUILD:
                     getattr(self, '%s_file' % format_).build_delay()
             for format_ in constants.EBOOK_FORMATS_WITH_COVERS:
                 if format_ not in app_settings.DONT_BUILD:
                     getattr(self, '%s_file' % format_).build_delay()
@@ -437,7 +843,7 @@ class Book(models.Model):
 
     def other_versions(self):
         """Find other versions (i.e. in other languages) of the book."""
 
     def other_versions(self):
         """Find other versions (i.e. in other languages) of the book."""
-        return type(self).objects.filter(common_slug=self.common_slug).exclude(pk=self.pk)
+        return type(self).objects.filter(common_slug=self.common_slug, findable=True).exclude(pk=self.pk)
 
     def parents(self):
         books = []
 
     def parents(self):
         books = []
@@ -448,8 +854,7 @@ class Book(models.Model):
         return books
 
     def pretty_title(self, html_links=False):
         return books
 
     def pretty_title(self, html_links=False):
-        names = [(tag.name, tag.get_absolute_url())
-            for tag in self.tags.filter(category='author')]
+        names = [(tag.name, tag.get_absolute_url()) for tag in self.authors().only('name', 'category', 'slug')]
         books = self.parents() + [self]
         names.extend([(b.title, b.get_absolute_url()) for b in books])
 
         books = self.parents() + [self]
         names.extend([(b.title, b.get_absolute_url()) for b in books])
 
@@ -459,6 +864,13 @@ class Book(models.Model):
             names = [tag[0] for tag in names]
         return ', '.join(names)
 
             names = [tag[0] for tag in names]
         return ', '.join(names)
 
+    def publisher(self):
+        publisher = self.get_extra_info_json()['publisher']
+        if isinstance(publisher, str):
+            return publisher
+        elif isinstance(publisher, list):
+            return ', '.join(publisher)
+
     @classmethod
     def tagged_top_level(cls, tags):
         """ Returns top-level books tagged with `tags`.
     @classmethod
     def tagged_top_level(cls, tags):
         """ Returns top-level books tagged with `tags`.
@@ -468,10 +880,10 @@ class Book(models.Model):
 
         """
         objects = cls.tagged.with_all(tags)
 
         """
         objects = cls.tagged.with_all(tags)
-        return objects.exclude(ancestor__in=objects)
+        return objects.filter(findable=True).exclude(ancestor__in=objects)
 
     @classmethod
 
     @classmethod
-    def book_list(cls, filter=None):
+    def book_list(cls, book_filter=None):
         """Generates a hierarchical listing of all books.
 
         Books are optionally filtered with a test function.
         """Generates a hierarchical listing of all books.
 
         Books are optionally filtered with a test function.
@@ -479,10 +891,9 @@ class Book(models.Model):
         """
 
         books_by_parent = {}
         """
 
         books_by_parent = {}
-        books = cls.objects.all().order_by('parent_number', 'sort_key').only(
-                'title', 'parent', 'slug')
-        if filter:
-            books = books.filter(filter).distinct()
+        books = cls.objects.filter(findable=True).order_by('parent_number', 'sort_key').only('title', 'parent', 'slug', 'extra_info')
+        if book_filter:
+            books = books.filter(book_filter).distinct()
 
             book_ids = set(b['pk'] for b in books.values("pk").iterator())
             for book in books.iterator():
 
             book_ids = set(b['pk'] for b in books.values("pk").iterator())
             for book in books.iterator():
@@ -500,7 +911,7 @@ class Book(models.Model):
             books_by_author[tag] = []
 
         for book in books_by_parent.get(None, ()):
             books_by_author[tag] = []
 
         for book in books_by_parent.get(None, ()):
-            authors = list(book.tags.filter(category='author'))
+            authors = list(book.authors().only('pk'))
             if authors:
                 for author in authors:
                     books_by_author[author].append(book)
             if authors:
                 for author in authors:
                     books_by_author[author].append(book)
@@ -510,53 +921,106 @@ class Book(models.Model):
         return books_by_author, orphans, books_by_parent
 
     _audiences_pl = {
         return books_by_author, orphans, books_by_parent
 
     _audiences_pl = {
-        "SP": (1, u"szkoła podstawowa"),
-        "SP1": (1, u"szkoła podstawowa"),
-        "SP2": (1, u"szkoła podstawowa"),
-        "P": (1, u"szkoła podstawowa"),
-        "G": (2, u"gimnazjum"),
-        "L": (3, u"liceum"),
-        "LP": (3, u"liceum"),
+        "SP": (1, "szkoła podstawowa"),
+        "SP1": (1, "szkoła podstawowa"),
+        "SP2": (1, "szkoła podstawowa"),
+        "SP3": (1, "szkoła podstawowa"),
+        "P": (1, "szkoła podstawowa"),
+        "G": (2, "gimnazjum"),
+        "L": (3, "liceum"),
+        "LP": (3, "liceum"),
     }
     }
+
     def audiences_pl(self):
     def audiences_pl(self):
-        audiences = self.extra_info.get('audiences', [])
+        audiences = self.get_extra_info_json().get('audiences', [])
         audiences = sorted(set([self._audiences_pl.get(a, (99, a)) for a in audiences]))
         return [a[1] for a in audiences]
 
     def stage_note(self):
         audiences = sorted(set([self._audiences_pl.get(a, (99, a)) for a in audiences]))
         return [a[1] for a in audiences]
 
     def stage_note(self):
-        stage = self.extra_info.get('stage')
+        stage = self.get_extra_info_json().get('stage')
         if stage and stage < '0.4':
         if stage and stage < '0.4':
-            return (_('This work needs modernisation'),
+            return (_('Ten utwór wymaga uwspółcześnienia'),
                     reverse('infopage', args=['wymagajace-uwspolczesnienia']))
         else:
             return None, None
 
                     reverse('infopage', args=['wymagajace-uwspolczesnienia']))
         else:
             return None, None
 
-    def choose_fragment(self):
+    def choose_fragments(self, number):
         fragments = self.fragments.order_by()
         fragments_count = fragments.count()
         if not fragments_count and self.children.exists():
             fragments = Fragment.objects.filter(book__ancestor=self).order_by()
             fragments_count = fragments.count()
         if fragments_count:
         fragments = self.fragments.order_by()
         fragments_count = fragments.count()
         if not fragments_count and self.children.exists():
             fragments = Fragment.objects.filter(book__ancestor=self).order_by()
             fragments_count = fragments.count()
         if fragments_count:
-            return fragments[randint(0, fragments_count - 1)]
+            if fragments_count > number:
+                offset = randint(0, fragments_count - number)
+            else:
+                offset = 0
+            return fragments[offset : offset + number]
         elif self.parent:
         elif self.parent:
-            return self.parent.choose_fragment()
+            return self.parent.choose_fragments(number)
+        else:
+            return []
+
+    def choose_fragment(self):
+        fragments = self.choose_fragments(1)
+        if fragments:
+            return fragments[0]
         else:
             return None
 
         else:
             return None
 
+    def fragment_data(self):
+        fragment = self.choose_fragment()
+        if fragment:
+            return {
+                'title': fragment.book.pretty_title(),
+                'html': re.sub('</?blockquote[^>]*>', '', fragment.get_short_text()),
+            }
+        else:
+            return None
 
 
-# add the file fields
-for format_ in Book.formats:
-    field_name = "%s_file" % format_
-    # This weird globals() assignment makes Django migrations comfortable.
-    _upload_to = _ebook_upload_to('book/%s/%%s.%s' % (format_, format_))
-    _upload_to.__name__ = '_%s_upload_to' % format_
-    globals()[_upload_to.__name__] = _upload_to
-
-    EbookField(format_, _("%s file" % format_.upper()),
-        upload_to=_upload_to,
-        storage=bofh_storage,
-        max_length=255,
-        blank=True,
-        default=''
-    ).contribute_to_class(Book, field_name)
+    def update_popularity(self):
+        count = self.tags.filter(category='set').values('user').order_by('user').distinct().count()
+        try:
+            pop = self.popularity
+            pop.count = count
+            pop.save()
+        except BookPopularity.DoesNotExist:
+            BookPopularity.objects.create(book=self, count=count)
+
+    def ridero_link(self):
+        return 'https://ridero.eu/%s/books/wl_%s/' % (get_language(), self.slug.replace('-', '_'))
+
+    def like(self, user):
+        from social.utils import likes, get_set, set_sets
+        if not likes(user, self):
+            tag = get_set(user, '')
+            set_sets(user, self, [tag])
+
+    def unlike(self, user):
+        from social.utils import likes, set_sets
+        if likes(user, self):
+            set_sets(user, self, [])
+
+    def full_sort_key(self):
+        return self.SORT_KEY_SEP.join((self.sort_key_author, self.sort_key, str(self.id)))
+
+    def cover_color(self):
+        return WLCover.epoch_colors.get(self.get_extra_info_json().get('epoch'), '#000000')
+
+    @cached_render('catalogue/book_mini_box.html')
+    def mini_box(self):
+        return {
+            'book': self
+        }
+
+    @cached_render('catalogue/book_mini_box.html')
+    def mini_box_nolink(self):
+        return {
+            'book': self,
+            'no_link': True,
+        }
+
+
+class BookPopularity(models.Model):
+    book = models.OneToOneField(Book, models.CASCADE, related_name='popularity')
+    count = models.IntegerField(default=0, db_index=True)