More tolerance for bad url metadata.
[wolnelektury.git] / apps / catalogue / models / book.py
index 6a31f4b..27240fb 100644 (file)
@@ -3,22 +3,25 @@
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 import re
-from django.conf import settings as settings
+from django.conf import settings
 from django.core.cache import get_cache
 from django.db import models
 from django.db.models import permalink
 import django.dispatch
+from django.core.urlresolvers import reverse
 from django.utils.datastructures import SortedDict
 from django.utils.translation import ugettext_lazy as _
 import jsonfield
+from fnpdjango.storage import BofhFileSystemStorage
 from catalogue import constants
 from catalogue.fields import EbookField
 from catalogue.models import Tag, Fragment, BookMedia
-from catalogue.utils import create_zip, split_tags, book_upload_path
+from catalogue.utils import create_zip, split_tags, related_tag_name
 from catalogue import app_settings
 from catalogue import tasks
 from newtagging import managers
 
+bofh_storage = BofhFileSystemStorage()
 
 permanent_cache = get_cache('permanent')
 
@@ -27,6 +30,7 @@ class Book(models.Model):
     """Represents a book imported from WL-XML."""
     title         = models.CharField(_('title'), max_length=120)
     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)
@@ -36,13 +40,20 @@ class Book(models.Model):
     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='{}')
+    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)
     # files generated during publication
 
     cover = EbookField('cover', _('cover'),
-                upload_to=book_upload_path('jpg'), null=True, blank=True)
+            null=True, blank=True,
+            upload_to=lambda i, n: 'book/cover/%s.jpg' % i.slug,
+            storage=bofh_storage, max_length=255)
+    # Cleaner version of cover for thumbs
+    cover_thumb = EbookField('cover_thumb', _('cover thumbnail'), 
+            null=True, blank=True,
+            upload_to=lambda i, n: 'book/cover_thumb/%s.jpg' % i.slug,
+            max_length=255)
     ebook_formats = constants.EBOOK_FORMATS
     formats = ebook_formats + ['html', 'xml']
 
@@ -74,8 +85,9 @@ class Book(models.Model):
         from sortify import sortify
 
         self.sort_key = sortify(self.title)
+        self.title = unicode(self.title) # ???
 
-        ret = super(Book, self).save(force_insert, force_update)
+        ret = super(Book, self).save(force_insert, force_update, **kwargs)
 
         if reset_short_html:
             self.reset_short_html()
@@ -86,10 +98,21 @@ class Book(models.Model):
     def get_absolute_url(self):
         return ('catalogue.views.book_detail', [self.slug])
 
+    @staticmethod
+    @permalink
+    def create_url(slug):
+        return ('catalogue.views.book_detail', [slug])
+
     @property
     def name(self):
         return self.title
 
+    def language_code(self):
+        return constants.LANGUAGES_3TO2.get(self.language, self.language)
+
+    def language_name(self):
+        return dict(settings.LANGUAGES).get(self.language_code(), "")
+
     def book_tag_slug(self):
         return ('l-' + self.slug)[:120]
 
@@ -112,7 +135,7 @@ class Book(models.Model):
         if self.has_media(type_):
             if type_ in Book.formats:
                 return getattr(self, "%s_file" % type_)
-            else:                                             
+            else:
                 return self.media.filter(type=type_)
         else:
             return None
@@ -124,7 +147,7 @@ class Book(models.Model):
     def get_ogg(self):
         return self.get_media("ogg")
     def get_daisy(self):
-        return self.get_media("daisy")                       
+        return self.get_media("daisy")
 
     def reset_short_html(self):
         if self.id is None:
@@ -135,6 +158,14 @@ class Book(models.Model):
         for fragm in self.fragments.all().iterator():
             fragm.reset_short_html()
 
+        try:
+            author = self.tags.filter(category='author')[0].sort_key
+        except IndexError:
+            author = u''
+        type(self).objects.filter(pk=self.pk).update(sort_key_author=author)
+
+
+
     def has_description(self):
         return len(self.description) > 0
     has_description.short_description = _('description')
@@ -189,20 +220,20 @@ class Book(models.Model):
         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, reuse_index=False, index_tags=True):
+    def search_index(self, book_info=None, index=None, index_tags=True, commit=True):
         import search
-        if reuse_index:
-            idx = search.ReusableIndex()
-        else:
-            idx = search.Index()
-            
-        idx.open()
+        if index is None:
+            index = search.Index()
         try:
-            idx.index_book(self, book_info)
+            index.index_book(self, book_info)
             if index_tags:
-                idx.index_tags()
-        finally:
-            idx.close()
+                index.index_tags()
+            if commit:
+                index.index.commit()
+        except Exception, e:
+            index.index.rollback()
+            raise e
+
 
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
@@ -223,7 +254,7 @@ class Book(models.Model):
     @classmethod
     def from_text_and_meta(cls, raw_file, book_info, overwrite=False,
             dont_build=None, search_index=True,
-            search_index_tags=True, search_index_reuse=False):
+            search_index_tags=True):
         if dont_build is None:
             dont_build = set()
         dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
@@ -298,7 +329,8 @@ class Book(models.Model):
         tasks.fix_tree_tags.delay(book)
         if 'cover' not in dont_build:
             book.cover.build_delay()
-        
+            book.cover_thumb.build_delay()
+
         # No saves behind this point.
 
         if has_own_text:
@@ -310,8 +342,7 @@ class Book(models.Model):
                 getattr(book, '%s_file' % format_).build_delay()
 
         if not settings.NO_SEARCH_INDEX and search_index:
-            book.search_index(index_tags=search_index_tags, reuse_index=search_index_reuse)
-            #index_book.delay(book.id, book_info)
+            tasks.index_book.delay(book.id, book_info=book_info, index_tags=search_index_tags)
 
         for child in notify_cover_changed:
             child.parent_cover_changed()
@@ -381,12 +412,17 @@ class Book(models.Model):
         if not self.cover_info(inherit=False):
             if 'cover' not in app_settings.DONT_BUILD:
                 self.cover.build_delay()
+                self.cover_thumb.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 child in self.children.all():
                 child.parent_cover_changed()
 
+    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)
+
     def related_info(self):
         """Keeps info about related objects (tags, media) in cache field."""
         if self._related_info is not None:
@@ -398,8 +434,15 @@ class Book(models.Model):
                     'author', 'kind', 'genre', 'epoch'))
             tags = split_tags(tags)
             for category in tags:
-                rel['tags'][category] = [
-                        (t.name, t.slug) for t in tags[category]]
+                cat = []
+                for tag in tags[category]:
+                    tag_info = {'slug': tag.slug, 'name': tag.name}
+                    for lc, ln in settings.LANGUAGES:
+                        tag_name = getattr(tag, "name_%s" % lc)
+                        if tag_name:
+                            tag_info["name_%s" % lc] = tag_name
+                    cat.append(tag_info)
+                rel['tags'][category] = cat
 
             for media_format in BookMedia.formats:
                 rel['media'][media_format] = self.has_media(media_format)
@@ -482,19 +525,19 @@ class Book(models.Model):
 
     def pretty_title(self, html_links=False):
         book = self
-        names = list(book.tags.filter(category='author'))
-
-        books = []
-        while book:
-            books.append(book)
-            book = book.parent
-        names.extend(reversed(books))
+        rel_info = book.related_info()
+        names = [(related_tag_name(tag), Tag.create_url('author', tag['slug']))
+                    for tag in rel_info['tags'].get('author', ())]
+        if 'parents' in rel_info:
+            books = [(name, Book.create_url(slug))
+                        for name, slug in rel_info['parents']]
+            names.extend(reversed(books))
+        names.append((self.title, self.get_absolute_url()))
 
         if html_links:
-            names = ['<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) for tag in names]
+            names = ['<a href="%s">%s</a>' % (tag[1], tag[0]) for tag in names]
         else:
-            names = [tag.name for tag in names]
-
+            names = [tag[0] for tag in names]
         return ', '.join(names)
 
     @classmethod
@@ -507,10 +550,11 @@ class Book(models.Model):
         """
         # get relevant books and their tags
         objects = cls.tagged.with_all(tags)
+        parents = objects.exclude(children=None).only('slug')
         # eliminate descendants
         l_tags = Tag.objects.filter(category='book',
-            slug__in=[book.book_tag_slug() for book in objects.iterator()])
-        descendants_keys = [book.pk for book in cls.tagged.with_any(l_tags).iterator()]
+            slug__in=[book.book_tag_slug() for book in parents.iterator()])
+        descendants_keys = [book.pk for book in cls.tagged.with_any(l_tags).only('pk').iterator()]
         if descendants_keys:
             objects = objects.exclude(pk__in=descendants_keys)
 
@@ -529,7 +573,7 @@ class Book(models.Model):
                 'title', 'parent', 'slug')
         if filter:
             books = books.filter(filter).distinct()
-            
+
             book_ids = set(b['pk'] for b in books.values("pk").iterator())
             for book in books.iterator():
                 parent = book.parent_id
@@ -545,7 +589,7 @@ class Book(models.Model):
         for tag in Tag.objects.filter(category='author').iterator():
             books_by_author[tag] = []
 
-        for book in books_by_parent.get(None,()):
+        for book in books_by_parent.get(None, ()):
             authors = list(book.tags.filter(category='author'))
             if authors:
                 for author in authors:
@@ -556,6 +600,7 @@ class Book(models.Model):
         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"),
@@ -565,9 +610,17 @@ class Book(models.Model):
     }
     def audiences_pl(self):
         audiences = self.extra_info.get('audiences', [])
-        audiences = sorted(set([self._audiences_pl[a] for a in 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):
+        stage = self.extra_info.get('stage')
+        if stage and stage < '0.4':
+            return (_('This work needs modernisation'),
+                    reverse('infopage', args=['wymagajace-uwspolczesnienia']))
+        else:
+            return None, None
+
     def choose_fragment(self):
         tag = self.book_tag()
         fragments = Fragment.tagged.with_any([tag])
@@ -582,6 +635,13 @@ class Book(models.Model):
 # add the file fields
 for format_ in Book.formats:
     field_name = "%s_file" % format_
+    upload_to = (lambda upload_path:
+            lambda i, n: upload_path % i.slug
+        )('book/%s/%%s.%s' % (format_, format_))
     EbookField(format_, _("%s file" % format_.upper()),
-            upload_to=book_upload_path(format_),
-            blank=True, default='').contribute_to_class(Book, field_name)
+        upload_to=upload_to,
+        storage=bofh_storage,
+        max_length=255,
+        blank=True,
+        default=''
+    ).contribute_to_class(Book, field_name)