from random import randint
import os.path
import re
+from slugify import slugify
+from sortify import sortify
from urllib.request import urlretrieve
from django.apps import apps
from django.conf import settings
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)
# 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)
'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)
tagged = managers.ModelTaggedItemManager(Tag)
tags = managers.TagDescriptor(Tag)
tag_relations = GenericRelation(Tag.intermediary_table_model)
+ translators = models.ManyToManyField(Tag, blank=True)
+ narrators = models.ManyToManyField(Tag, blank=True, related_name='narrated')
+ has_audio = models.BooleanField(default=False)
html_built = django.dispatch.Signal()
published = django.dispatch.Signal()
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 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 []
total += app_settings.GET_MP3_LENGTH(media.file.path)
return int(total)
+ def get_time(self):
+ return round(self.xml_file.size / 1000 * 40)
+
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 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')
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 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')
def media_audio_epub(self):
return self.get_media('audio.epub')
- def get_audiobooks(self):
+ def get_audiobooks(self, with_children=False, processing=False):
ogg_files = {}
for m in self.media.filter(type='ogg').order_by().iterator():
ogg_files[m.name] = m
media['ogg'] = ogg
audiobooks.append(media)
- projects = sorted(projects)
- total_duration = '%d:%02d' % (
- total_duration // 60,
- total_duration % 60
- )
+ if with_children:
+ for child in self.get_children():
+ ch_audiobooks, ch_projects, ch_duration = child.get_audiobooks(
+ with_children=True, processing=True)
+ audiobooks.append({'part': child})
+ audiobooks += ch_audiobooks
+ projects.update(ch_projects)
+ total_duration += ch_duration
+
+ if not processing:
+ projects = sorted(projects)
+ total_duration = '%d:%02d' % (
+ total_duration // 60,
+ total_duration % 60
+ )
+
return audiobooks, projects, total_duration
+ def get_audiobooks_with_children(self):
+ return self.get_audiobooks(with_children=True)
+
def wldocument(self, parse_dublincore=True, inherit=True):
from catalogue.import_utils import ORMDocProvider
from librarian.parser import WLDocument
@classmethod
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):
+ remote_gallery_url=None, days=0, findable=True, logo=None, logo_mono=None, logo_alt=None):
from catalogue import tasks
if dont_build is None:
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:
book.common_slug = book.slug
- book.extra_info = json.dumps(book_info.to_dict())
+ extra = book_info.to_dict()
+ if logo:
+ extra['logo'] = logo
+ if logo_mono:
+ extra['logo_mono'] = logo_mono
+ if logo_alt:
+ extra['logo_alt'] = logo_alt
+ book.extra_info = json.dumps(extra)
book.load_abstract()
book.load_toc()
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)
+ 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)
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)
else:
entity, entity_created = Entity.objects.get_or_create(uri=uri)
if entity_created:
- entity.populate()
- entity.save()
+ try:
+ entity.populate()
+ except:
+ pass
+ else:
+ entity.save()
ref, ref_created = entity.reference_set.get_or_create(book=self)
refs[uri] = ref
if not ref_created:
def references(self):
return self.reference_set.all().select_related('entity')
+ def update_has_audio(self):
+ self.has_audio = False
+ if self.media.filter(type='mp3').exists():
+ self.has_audio = True
+ if self.descendant.filter(has_audio=True).exists():
+ self.has_audio = True
+ self.save(update_fields=['has_audio'])
+ if self.parent is not None:
+ self.parent.update_has_audio()
+
+ def update_narrators(self):
+ narrator_names = set()
+ for bm in self.media.filter(type='mp3'):
+ narrator_names.update(set(
+ a.strip() for a in re.split(r',|\si\s', bm.artist)
+ ))
+ narrators = []
+
+ for name in narrator_names:
+ if not name: continue
+ slug = slugify(name)
+ try:
+ t = Tag.objects.get(category='author', slug=slug)
+ except Tag.DoesNotExist:
+ sort_key = sortify(
+ ' '.join(name.rsplit(' ', 1)[::-1]).lower()
+ )
+ t = Tag.objects.create(
+ category='author',
+ name_pl=name,
+ slug=slug,
+ sort_key=sort_key,
+ )
+ narrators.append(t)
+ self.narrators.set(narrators)
+
@classmethod
@transaction.atomic
def repopulate_ancestors(cls):
fnp / wolnelektury.git / blobdiff