Update to new librarian api for html, txt.

[wolnelektury.git] / src / catalogue / models / book.py

diff --git a/src/catalogue/models/book.py b/src/catalogue/models/book.py
index 10e9d22..7d36662 100644 (file)
--- a/src/catalogue/models/book.py
+++ b/src/catalogue/models/book.py
@@ -19,7 +19,7 @@ from django.utils.translation import gettext_lazy as _, get_language
 from fnpdjango.storage import BofhFileSystemStorage
 from lxml import html
 from librarian.cover import WLCover
-from librarian.html import transform_abstrakt
+from librarian.builders.html import AbstraktHtmlBuilder
 from librarian.builders import builders
 from newtagging import managers
 from catalogue import constants
@@ -91,6 +91,7 @@ class Book(models.Model):
     tagged = managers.ModelTaggedItemManager(Tag)
     tags = managers.TagDescriptor(Tag)
     tag_relations = GenericRelation(Tag.intermediary_table_model)
+    translators = models.ManyToManyField(Tag, blank=True)
 
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
@@ -154,12 +155,6 @@ class Book(models.Model):
     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:
@@ -332,7 +327,10 @@ class Book(models.Model):
         return int(total)
 
     def get_time(self):
-        return round(self.xml_file.size / 1000 * 40)
+        try:
+            return round(self.xml_file.size / 1000 * 40)
+        except ValueError:
+            return 0
     
     def has_media(self, type_):
         if type_ in Book.formats:
@@ -419,7 +417,7 @@ class Book(models.Model):
     has_daisy_file.boolean = True
 
     def has_sync_file(self):
-        return self.has_media("sync")
+        return settings.FEATURE_SYNCHRO and self.has_media("sync")
 
     def get_sync(self):
         with self.get_media('sync').first().file.open('r') as f:
@@ -559,11 +557,8 @@ class Book(models.Model):
                 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 = ''
+        self.abstract = AbstraktHtmlBuilder().build(
+            self.wldocument2()).get_bytes().decode('utf-8')
 
     def load_toc(self):
         self.toc = ''
@@ -658,14 +653,12 @@ class Book(models.Model):
 
         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)
+        just_tags = [t for (t, rel) in meta_tags if not rel]
+        book.tags = set(just_tags + book_shelves)
         book.save()  # update sort_key_author
 
+        book.translators.set([t for (t, rel) in meta_tags if rel == 'translator'])
+
         cover_changed = old_cover != book.cover_info()
         obsolete_children = set(b for b in book.children.all()
                                 if b not in children)
@@ -724,13 +717,14 @@ class Book(models.Model):
         cls.published.send(sender=cls, instance=book)
         return book
 
+    # TODO TEST
     def update_references(self):
         Entity = apps.get_model('references', 'Entity')
         doc = self.wldocument2()
-        doc._compat_assign_section_ids()
-        doc._compat_assign_ordered_ids()
+        doc.assign_ids()
+
         refs = {}
-        for ref_elem in doc.references():
+        for i, ref_elem in enumerate(doc.references()):
             uri = ref_elem.attrib.get('href', '')
             if not uri:
                 continue
@@ -749,10 +743,8 @@ class Book(models.Model):
                 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))
+            anchor = ref_elem.get_link()
+
             snippet = ref_elem.get_snippet()
             b = builders['html-snippet']()
             for s in snippet:
@@ -760,7 +752,8 @@ class Book(models.Model):
             html = b.output().get_bytes().decode('utf-8')
 
             ref.occurence_set.create(
-                section=sec,
+                section=i,
+                anchor=anchor,
                 html=html
             )
         self.reference_set.exclude(entity__uri__in=refs).delete()