+
+ @classmethod
+ def from_xml_file(cls, xml_file, **kwargs):
+ # use librarian to parse meta-data
+ book_info = dcparser.parse(xml_file)
+
+ if not isinstance(xml_file, File):
+ xml_file = File(open(xml_file))
+
+ try:
+ return cls.from_text_and_meta(xml_file, book_info, **kwargs)
+ finally:
+ xml_file.close()
+
+ @classmethod
+ def from_text_and_meta(cls, raw_file, book_info, overwrite=False, build_epub=True, build_txt=True):
+ import re
+
+ # check for parts before we do anything
+ children = []
+ if hasattr(book_info, 'parts'):
+ for part_url in book_info.parts:
+ base, slug = part_url.rsplit('/', 1)
+ try:
+ children.append(Book.objects.get(slug=slug))
+ except Book.DoesNotExist, e:
+ raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug)
+
+
+ # Read book metadata
+ book_base, book_slug = book_info.url.rsplit('/', 1)
+ if re.search(r'[^a-zA-Z0-9-]', book_slug):
+ raise ValueError('Invalid characters in slug')
+ book, created = Book.objects.get_or_create(slug=book_slug)
+
+ if created:
+ book_shelves = []
+ else:
+ if not overwrite:
+ raise Book.AlreadyExists(_('Book %s already exists') % book_slug)
+ # Save shelves for this book
+ book_shelves = list(book.tags.filter(category='set'))
+
+ book.title = book_info.title
+ book.set_extra_info_value(book_info.to_dict())
+ book._short_html = ''
+ 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)
+
+ book.tags = set(meta_tags + book_shelves)
+
+ book_tag = book.book_tag()
+
+ for n, child_book in enumerate(children):
+ child_book.parent = book
+ child_book.parent_number = n
+ child_book.save()
+
+ # Save XML and HTML files
+ book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
+
+ # delete old fragments when overwriting
+ book.fragments.all().delete()
+
+ if book.build_html():
+ if not settings.NO_BUILD_TXT and build_txt:
+ book.build_txt()
+
+ if not settings.NO_BUILD_EPUB and build_epub:
+ book.root_ancestor.build_epub()
+
+ book_descendants = list(book.children.all())
+ # add l-tag to descendants and their fragments
+ # delete unnecessary EPUB files
+ while len(book_descendants) > 0:
+ child_book = book_descendants.pop(0)
+ 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())
+
+ # refresh cache
+ book.reset_tag_counter()
+ book.reset_theme_counter()
+
+ book.save()
+ return book
+
+
+ def refresh_tag_counter(self):
+ tags = {}
+ for child in self.children.all().order_by():
+ for tag_pk, value in child.tag_counter.iteritems():
+ tags[tag_pk] = tags.get(tag_pk, 0) + value
+ for tag in self.tags.exclude(category__in=('book', 'theme', 'set')).order_by():
+ tags[tag.pk] = 1
+ self.set__tag_counter_value(tags)
+ self.save(reset_short_html=False)
+ return tags
+
+ def reset_tag_counter(self):
+ self._tag_counter = None
+ self.save(reset_short_html=False)
+ if self.parent:
+ self.parent.reset_tag_counter()
+
+ @property
+ def tag_counter(self):
+ if self._tag_counter is None:
+ return self.refresh_tag_counter()
+ return dict((int(k), v) for k, v in self.get__tag_counter_value().iteritems())
+
+ def refresh_theme_counter(self):
+ tags = {}
+ for fragment in Fragment.tagged.with_any([self.book_tag()]).order_by():
+ for tag in fragment.tags.filter(category='theme').order_by():
+ tags[tag.pk] = tags.get(tag.pk, 0) + 1
+ self.set__theme_counter_value(tags)
+ self.save(reset_short_html=False)
+ return tags
+
+ def reset_theme_counter(self):
+ self._theme_counter = None
+ self.save(reset_short_html=False)
+ if self.parent:
+ self.parent.reset_theme_counter()
+
+ @property
+ def theme_counter(self):
+ if self._theme_counter is None:
+ return self.refresh_theme_counter()
+ return dict((int(k), v) for k, v in self.get__theme_counter_value().iteritems())
+
+ 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))
+
+ if html_links:
+ names = ['<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) for tag in names]
+ else:
+ names = [tag.name for tag in names]
+
+ return ', '.join(names)
+
+ @classmethod
+ def tagged_top_level(cls, tags):
+ """ Returns top-level books tagged with `tags'.
+
+ It only returns those books which don't have ancestors which are
+ also tagged with those tags.
+
+ """
+ # get relevant books and their tags
+ objects = cls.tagged.with_all(tags)
+ # eliminate descendants
+ l_tags = Tag.objects.filter(category='book', slug__in=[book.book_tag_slug() for book in objects])
+ descendants_keys = [book.pk for book in cls.tagged.with_any(l_tags)]
+ if descendants_keys:
+ objects = objects.exclude(pk__in=descendants_keys)
+
+ return objects
fnp / wolnelektury.git / blobdiff