audio fix

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

diff --git a/src/catalogue/models/book.py b/src/catalogue/models/book.py
index ac8c02b..ff09853 100644 (file)
--- a/src/catalogue/models/book.py
+++ b/src/catalogue/models/book.py
@@ -43,6 +43,7 @@ class Book(models.Model):
     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)
+    license = models.CharField('licencja', max_length=255, blank=True, db_index=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)
@@ -62,6 +63,7 @@ class Book(models.Model):
     # files generated during publication
     xml_file = fields.XmlField(storage=bofh_storage, with_etag=False)
     html_file = fields.HtmlField(storage=bofh_storage)
+    html_nonotes_file = fields.HtmlNonotesField(storage=bofh_storage)
     fb2_file = fields.Fb2Field(storage=bofh_storage)
     txt_file = fields.TxtField(storage=bofh_storage)
     epub_file = fields.EpubField(storage=bofh_storage)
@@ -79,7 +81,7 @@ class Book(models.Model):
         'okładka dla Ebookpoint')
 
     ebook_formats = constants.EBOOK_FORMATS
-    formats = ebook_formats + ['html', 'xml']
+    formats = ebook_formats + ['html', 'xml', 'html_nonotes']
 
     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)
@@ -375,6 +377,9 @@ class Book(models.Model):
     def html_url(self):
         return self.media_url('html')
 
+    def html_nonotes_url(self):
+        return self.media_url('html_nonotes')
+
     def pdf_url(self):
         return self.media_url('pdf')
 
@@ -416,6 +421,67 @@ class Book(models.Model):
     def has_sync_file(self):
         return settings.FEATURE_SYNCHRO and self.has_media("sync")
 
+    def build_sync_file(self):
+        from lxml import html
+        from django.core.files.base import ContentFile
+        with self.html_file.open('rb') as f:
+            h = html.fragment_fromstring(f.read().decode('utf-8'))
+
+        durations = [
+            m['mp3'].duration
+            for m in self.get_audiobooks()[0]
+        ]
+        if settings.MOCK_DURATIONS:
+            durations = settings.MOCK_DURATIONS
+
+        sync = []
+        ts = None
+        sid = 1
+        dirty = False
+        for elem in h.iter():
+            if elem.get('data-audio-ts'):
+                part, ts = int(elem.get('data-audio-part')), float(elem.get('data-audio-ts'))
+                ts = str(round(sum(durations[:part - 1]) + ts, 3))
+                # check if inside verse
+                p = elem.getparent()
+                while p is not None:
+                    # Workaround for missing ids.
+                    if 'verse' in p.get('class', ''):
+                        if not p.get('id'):
+                            p.set('id', f'syn{sid}')
+                            dirty = True
+                            sid += 1
+                        sync.append((ts, p.get('id')))
+                        ts = None
+                        break
+                    p = p.getparent()
+            elif ts:
+                cls = elem.get('class', '')
+                # Workaround for missing ids.
+                if 'paragraph' in cls or 'verse' in cls or elem.tag in ('h1', 'h2', 'h3', 'h4'):
+                    if not elem.get('id'):
+                        elem.set('id', f'syn{sid}')
+                        dirty = True
+                        sid += 1
+                    sync.append((ts, elem.get('id')))
+                    ts = None
+        if dirty:
+            htext = html.tostring(h, encoding='utf-8')
+            with open(self.html_file.path, 'wb') as f:
+                f.write(htext)
+        try:
+            bm = self.media.get(type='sync')
+        except:
+            bm = BookMedia(book=self, type='sync')
+        sync = (
+            '27\n' + '\n'.join(
+                f'{s[0]}\t{sync[i+1][0]}\t{s[1]}' for i, s in enumerate(sync[:-1])
+            )).encode('latin1')
+        bm.file.save(
+            None, ContentFile(sync)
+            )
+
+    
     def get_sync(self):
         with self.get_media('sync').first().file.open('r') as f:
             sync = f.read().split('\n')
@@ -635,6 +701,7 @@ class Book(models.Model):
         book.findable = findable
         book.language = book_info.language
         book.title = book_info.title
+        book.license = book_info.license or ''
         if book_info.variant_of:
             book.common_slug = book_info.variant_of.slug
         else:
@@ -653,11 +720,6 @@ class Book(models.Model):
 
         meta_tags = Tag.tags_from_info(book_info)
 
-        for tag, relationship in meta_tags:
-            if not tag.for_books:
-                tag.for_books = True
-                tag.save()
-
         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
@@ -709,6 +771,7 @@ class Book(models.Model):
         for format_ in constants.EBOOK_FORMATS_WITH_CHILDREN:
             if format_ not in dont_build:
                 getattr(book, '%s_file' % format_).build_delay()
+        book.html_nonotes_file.build_delay()
 
         if not settings.NO_SEARCH_INDEX and search_index and findable:
             tasks.index_book.delay(book.id)