# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
from collections import OrderedDict
+import json
from datetime import date, timedelta
from random import randint
import os.path
import re
from urllib.request import urlretrieve
+from django.apps import apps
from django.conf import settings
from django.db import connection, models, transaction
import django.dispatch
from django.contrib.contenttypes.fields import GenericRelation
+from django.template.loader import render_to_string
from django.urls import reverse
-from django.utils.translation import ugettext_lazy as _, get_language
-from django.utils.deconstruct import deconstructible
-import jsonfield
+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 newtagging import managers
from catalogue import constants
-from catalogue.fields import EbookField
+from catalogue import fields
from catalogue.models import Tag, Fragment, BookMedia
from catalogue.utils import create_zip, gallery_url, gallery_path, split_tags, get_random_hash
from catalogue.models.tag import prefetched_relations
from catalogue import app_settings
-from catalogue import tasks
from wolnelektury.utils import makedirs, cached_render, clear_cached_renders
bofh_storage = BofhFileSystemStorage()
-@deconstructible
-class UploadToPath(object):
- def __init__(self, path):
- self.path = path
-
- def __call__(self, instance, filename):
- return self.path % instance.slug
-
-
-_cover_upload_to = UploadToPath('book/cover/%s.jpg')
-_cover_thumb_upload_to = UploadToPath('book/cover_thumb/%s.jpg')
-_cover_api_thumb_upload_to = UploadToPath('book/cover_api_thumb/%s.jpg')
-_simple_cover_upload_to = UploadToPath('book/cover_simple/%s.jpg')
-
-
-def _ebook_upload_to(upload_path):
- return UploadToPath(upload_path)
-
-
class Book(models.Model):
"""Represents a book imported from WL-XML."""
title = models.CharField(_('title'), max_length=32767)
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'')
+ _('sort key by author'), max_length=120, db_index=True, editable=False, default='')
slug = models.SlugField(_('slug'), max_length=120, db_index=True, unique=True)
common_slug = models.SlugField(_('slug'), max_length=120, db_index=True)
language = models.CharField(_('language code'), max_length=3, db_index=True, default=app_settings.DEFAULT_LANGUAGE)
description = models.TextField(_('description'), blank=True)
abstract = models.TextField(_('abstract'), blank=True)
+ toc = models.TextField(_('toc'), blank=True)
created_at = models.DateTimeField(_('creation date'), auto_now_add=True, db_index=True)
changed_at = models.DateTimeField(_('change date'), auto_now=True, db_index=True)
parent_number = models.IntegerField(_('parent number'), default=0)
- extra_info = jsonfield.JSONField(_('extra information'), default={})
+ extra_info = models.TextField(_('extra information'), default='{}')
gazeta_link = models.CharField(blank=True, max_length=240)
wiki_link = models.CharField(blank=True, max_length=240)
print_on_demand = models.BooleanField(_('print on demand'), default=False)
preview = models.BooleanField(_('preview'), default=False)
preview_until = models.DateField(_('preview until'), blank=True, null=True)
preview_key = models.CharField(max_length=32, blank=True, null=True)
+ findable = models.BooleanField(_('findable'), default=True, db_index=True)
# files generated during publication
- cover = EbookField(
- 'cover', _('cover'),
- null=True, blank=True,
- upload_to=_cover_upload_to,
- storage=bofh_storage, max_length=255)
+ xml_file = fields.XmlField(storage=bofh_storage, with_etag=False)
+ html_file = fields.HtmlField(storage=bofh_storage)
+ fb2_file = fields.Fb2Field(storage=bofh_storage)
+ txt_file = fields.TxtField(storage=bofh_storage)
+ epub_file = fields.EpubField(storage=bofh_storage)
+ mobi_file = fields.MobiField(storage=bofh_storage)
+ pdf_file = fields.PdfField(storage=bofh_storage)
+
+ cover = fields.CoverField(_('cover'), storage=bofh_storage)
# Cleaner version of cover for thumbs
- cover_thumb = EbookField(
- 'cover_thumb', _('cover thumbnail'),
- null=True, blank=True,
- upload_to=_cover_thumb_upload_to,
- max_length=255)
- cover_api_thumb = EbookField(
- 'cover_api_thumb', _('cover thumbnail for mobile app'),
- null=True, blank=True,
- upload_to=_cover_api_thumb_upload_to,
- max_length=255)
- simple_cover = EbookField(
- 'simple_cover', _('cover for mobile app'),
- null=True, blank=True,
- upload_to=_simple_cover_upload_to,
- max_length=255)
+ cover_clean = fields.CoverCleanField(_('clean cover'))
+ cover_thumb = fields.CoverThumbField(_('cover thumbnail'))
+ cover_api_thumb = fields.CoverApiThumbField(
+ _('cover thumbnail for mobile app'))
+ simple_cover = fields.SimpleCoverField(_('cover for mobile app'))
+ cover_ebookpoint = fields.CoverEbookpointField(
+ _('cover for Ebookpoint'))
+
ebook_formats = constants.EBOOK_FORMATS
formats = ebook_formats + ['html', 'xml']
SORT_KEY_SEP = '$'
+ is_book = True
+
class AlreadyExists(Exception):
pass
def __str__(self):
return self.title
+ def get_extra_info_json(self):
+ return json.loads(self.extra_info or '{}')
+
def get_initial(self):
try:
return re.search(r'\w', self.title, re.U).group(0)
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.extra_info.get('translators')
+ translators = self.get_extra_info_json().get('translators')
if not translators:
return None
if len(translators) > 3:
others = ' i inni'
else:
others = ''
- return ', '.join(u'\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators) + others
+ return ', '.join('\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators) + others
def cover_source(self):
- return self.extra_info.get('cover_source', self.parent.cover_source() if self.parent else '')
+ return self.get_extra_info_json().get('cover_source', self.parent.cover_source() if self.parent else '')
+
+ @property
+ def isbn_pdf(self):
+ return self.get_extra_info_json().get('isbn_pdf')
+
+ @property
+ def isbn_epub(self):
+ return self.get_extra_info_json().get('isbn_epub')
+
+ @property
+ def isbn_mobi(self):
+ return self.get_extra_info_json().get('isbn_mobi')
+
+ def is_accessible_to(self, user):
+ if not self.preview:
+ return True
+ if not user.is_authenticated:
+ return False
+ Membership = apps.get_model('club', 'Membership')
+ if Membership.is_active_for(user):
+ return True
+ Funding = apps.get_model('funding', 'Funding')
+ if Funding.objects.filter(user=user, offer__book=self):
+ return True
+ return False
def save(self, force_insert=False, force_update=False, **kwargs):
from sortify import sortify
try:
author = self.authors().first().sort_key
except AttributeError:
- author = u''
+ author = ''
self.sort_key_author = author
self.cached_author = self.tag_unicode('author')
- self.has_audience = 'audience' in self.extra_info
+ self.has_audience = 'audience' in self.get_extra_info_json()
if self.preview and not self.preview_key:
self.preview_key = get_random_hash(self.slug)[:32]
def gallery_url(self):
return gallery_url(self.slug)
+ def get_first_text(self):
+ if self.html_file:
+ return self
+ child = self.children.all().order_by('parent_number').first()
+ if child is not None:
+ return child.get_first_text()
+
+ def get_last_text(self):
+ if self.html_file:
+ return self
+ child = self.children.all().order_by('parent_number').last()
+ if child is not None:
+ return child.get_last_text()
+
+ def get_prev_text(self):
+ if not self.parent:
+ return None
+ sibling = self.parent.children.filter(parent_number__lt=self.parent_number).order_by('-parent_number').first()
+ if sibling is not None:
+ return sibling.get_last_text()
+
+ if self.parent.html_file:
+ return self.parent
+
+ return self.parent.get_prev_text()
+
+ def get_next_text(self, inside=True):
+ if inside:
+ child = self.children.order_by('parent_number').first()
+ if child is not None:
+ return child.get_first_text()
+
+ if not self.parent:
+ return None
+ sibling = self.parent.children.filter(parent_number__gt=self.parent_number).order_by('parent_number').first()
+ if sibling is not None:
+ 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 []
+ return self.parent.children.all().order_by('parent_number')
+
+ def get_children(self):
+ return self.children.all().order_by('parent_number')
+
@property
def name(self):
return self.title
@staticmethod
def format_audio_length(seconds):
+ """
+ >>> Book.format_audio_length(1)
+ '0:01'
+ >>> Book.format_audio_length(3661)
+ '1:01:01'
+ """
if seconds < 60*60:
minutes = seconds // 60
seconds = seconds % 60
def get_daisy(self):
return self.get_media("daisy")
+ def get_audio_epub(self):
+ return self.get_media("audio.epub")
+
def media_url(self, format_):
media = self.get_media(format_)
if media:
has_daisy_file.short_description = 'DAISY'
has_daisy_file.boolean = True
+ def has_sync_file(self):
+ return self.has_media("sync")
+
+ def get_sync(self):
+ with self.get_media('sync').first().file.open('r') as f:
+ sync = f.read().split('\n')
+ offset = float(sync[0])
+ items = []
+ for line in sync[1:]:
+ if not line:
+ continue
+ start, end, elid = line.split()
+ items.append([elid, float(start) + offset])
+ return json.dumps(items)
+
+ def has_audio_epub_file(self):
+ return self.has_media("audio.epub")
+
+ @property
+ def media_daisy(self):
+ return self.get_media('daisy')
+
+ @property
+ def media_audio_epub(self):
+ return self.get_media('audio.epub')
+
def get_audiobooks(self):
ogg_files = {}
for m in self.media.filter(type='ogg').order_by().iterator():
audiobooks = []
projects = set()
+ total_duration = 0
for mp3 in self.media.filter(type='mp3').iterator():
# ogg files are always from the same project
- meta = mp3.extra_info
+ meta = mp3.get_extra_info_json()
project = meta.get('project')
if not project:
# temporary fallback
- project = u'CzytamySłuchając'
+ project = 'CzytamySłuchając'
projects.add((project, meta.get('funded_by', '')))
+ total_duration += mp3.duration or 0
media = {'mp3': mp3}
audiobooks.append(media)
projects = sorted(projects)
- return audiobooks, projects
+ total_duration = '%d:%02d' % (
+ total_duration // 60,
+ total_duration % 60
+ )
+ return audiobooks, projects, total_duration
def wldocument(self, parse_dublincore=True, inherit=True):
from catalogue.import_utils import ORMDocProvider
parse_dublincore=parse_dublincore,
meta_fallbacks=meta_fallbacks)
+ def wldocument2(self):
+ from catalogue.import_utils import ORMDocProvider
+ from librarian.document import WLDocument
+ doc = WLDocument(
+ self.xml_file.path,
+ provider=ORMDocProvider(self)
+ )
+ doc.meta.update(self.cover_info())
+ return doc
+
+
@staticmethod
def zip_format(format_):
def pretty_file_name(book):
return "%s/%s.%s" % (
- book.extra_info['author'],
+ book.get_extra_info_json()['author'],
book.slug,
format_)
field_name = "%s_file" % format_
- books = Book.objects.filter(parent=None).exclude(**{field_name: ""}).exclude(preview=True)
+ field = getattr(Book, field_name)
+ books = Book.objects.filter(parent=None).exclude(**{field_name: ""}).exclude(preview=True).exclude(findable=False)
paths = [(pretty_file_name(b), getattr(b, field_name).path) for b in books.iterator()]
- return create_zip(paths, app_settings.FORMAT_ZIPS[format_])
+ return create_zip(paths, field.ZIP)
def zip_audiobooks(self, format_):
bm = BookMedia.objects.filter(book=self, type=format_)
- paths = map(lambda bm: (None, bm.file.path), bm)
- return create_zip(paths, "%s_%s" % (self.slug, format_))
+ paths = map(lambda bm: (bm.get_nice_filename(), bm.file.path), bm)
+ licenses = set()
+ for m in bm:
+ license = constants.LICENSES.get(
+ m.get_extra_info_json().get('license'), {}
+ ).get('locative')
+ if license:
+ licenses.add(license)
+ readme = render_to_string('catalogue/audiobook_zip_readme.txt', {
+ 'licenses': licenses,
+ 'meta': self.wldocument2().meta,
+ })
+ return create_zip(paths, "%s_%s" % (self.slug, format_), {'informacje.txt': readme})
def search_index(self, book_info=None, index=None, index_tags=True, commit=True):
+ if not self.findable:
+ return
if index is None:
from search.index import Index
index = Index()
# will make problems in conjunction with paid previews
def download_pictures(self, remote_gallery_url):
+ # This is only needed for legacy relative image paths.
gallery_path = self.gallery_path()
# delete previous files, so we don't include old files in ebooks
if os.path.isdir(gallery_path):
makedirs(gallery_path)
for ilustr in ilustr_elements:
ilustr_src = ilustr.get('src')
+ if '/' in ilustr_src:
+ continue
ilustr_path = os.path.join(gallery_path, ilustr_src)
urlretrieve('%s/%s' % (remote_gallery_url, ilustr_src), ilustr_path)
else:
self.abstract = ''
+ def load_toc(self):
+ self.toc = ''
+ if self.html_file:
+ parser = html.HTMLParser(encoding='utf-8')
+ tree = html.parse(self.html_file.path, parser=parser)
+ toc = tree.find('//div[@id="toc"]/ol')
+ if toc is None or not len(toc):
+ return
+ html_link = reverse('book_text', args=[self.slug])
+ for a in toc.findall('.//a'):
+ a.attrib['href'] = html_link + a.attrib['href']
+ self.toc = html.tostring(toc, encoding='unicode')
+ # div#toc
+
@classmethod
def from_xml_file(cls, xml_file, **kwargs):
from django.core.files import File
@classmethod
def from_text_and_meta(cls, raw_file, book_info, overwrite=False, dont_build=None, search_index=True,
- search_index_tags=True, remote_gallery_url=None, days=0):
+ search_index_tags=True, remote_gallery_url=None, days=0, findable=True):
+ from catalogue import tasks
+
if dont_build is None:
dont_build = set()
dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
if book.preview:
book.xml_file.set_readable(False)
+ book.findable = findable
book.language = book_info.language
book.title = book_info.title
if book_info.variant_of:
book.common_slug = book_info.variant_of.slug
else:
book.common_slug = book.slug
- book.extra_info = book_info.to_dict()
+ book.extra_info = json.dumps(book_info.to_dict())
book.load_abstract()
+ book.load_toc()
book.save()
meta_tags = Tag.tags_from_info(book_info)
# Build cover.
if 'cover' not in dont_build:
book.cover.build_delay()
+ book.cover_clean.build_delay()
book.cover_thumb.build_delay()
book.cover_api_thumb.build_delay()
book.simple_cover.build_delay()
+ book.cover_ebookpoint.build_delay()
# Build HTML and ebooks.
book.html_file.build_delay()
if format_ not in dont_build:
getattr(book, '%s_file' % format_).build_delay()
- if not settings.NO_SEARCH_INDEX and search_index:
+ if not settings.NO_SEARCH_INDEX and search_index and findable:
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()
book.update_popularity()
+ tasks.update_references.delay(book.id)
+
cls.published.send(sender=cls, instance=book)
return book
+ def get_master(self):
+ master_tags = [
+ 'opowiadanie',
+ 'powiesc',
+ 'dramat_wierszowany_l',
+ 'dramat_wierszowany_lp',
+ 'dramat_wspolczesny', 'liryka_l', 'liryka_lp',
+ 'wywiad',
+ ]
+ from librarian.parser import WLDocument
+ wld = WLDocument.from_file(self.xml_file.path, parse_dublincore=False)
+ root = wld.edoc.getroot()
+ for master in root.iter():
+ if master.tag in master_tags:
+ return master
+
+ def update_references(self):
+ from references.models import Entity, Reference
+ master = self.get_master()
+ if master is None:
+ master = []
+ found = set()
+ for i, sec in enumerate(master):
+ for ref in sec.findall('.//ref'):
+ href = ref.attrib.get('href', '')
+ if not href or href in found:
+ continue
+ found.add(href)
+ entity, created = Entity.objects.get_or_create(
+ uri=href
+ )
+ ref, created = Reference.objects.get_or_create(
+ book=self,
+ entity=entity
+ )
+ ref.first_section = 'sec%d' % (i + 1)
+ entity.populate()
+ entity.save()
+ Reference.objects.filter(book=self).exclude(entity__uri__in=found).delete()
+
+ @property
+ def references(self):
+ return self.reference_set.all().select_related('entity')
+
@classmethod
@transaction.atomic
def repopulate_ancestors(cls):
b.ancestor.add(parent)
parent = parent.parent
+ @property
+ def ancestors(self):
+ if self.parent:
+ for anc in self.parent.ancestors:
+ yield anc
+ yield self.parent
+ else:
+ return []
+
def clear_cache(self):
clear_cached_renders(self.mini_box)
clear_cached_renders(self.mini_box_nolink)
need = False
info = {}
for field in ('cover_url', 'cover_by', 'cover_source'):
- val = self.extra_info.get(field)
+ val = self.get_extra_info_json().get(field)
if val:
info[field] = val
else:
def related_themes(self):
return Tag.objects.usage_for_queryset(
Fragment.objects.filter(models.Q(book=self) | models.Q(book__ancestor=self)),
- counts=True).filter(category='theme')
+ counts=True).filter(category='theme').order_by('-count')
def parent_cover_changed(self):
"""Called when parent book's cover image is changed."""
if not self.cover_info(inherit=False):
if 'cover' not in app_settings.DONT_BUILD:
self.cover.build_delay()
+ self.cover_clean.build_delay()
self.cover_thumb.build_delay()
self.cover_api_thumb.build_delay()
self.simple_cover.build_delay()
+ self.cover_ebookpoint.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()
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)
+ return type(self).objects.filter(common_slug=self.common_slug, findable=True).exclude(pk=self.pk)
def parents(self):
books = []
return ', '.join(names)
def publisher(self):
- publisher = self.extra_info['publisher']
+ publisher = self.get_extra_info_json()['publisher']
if isinstance(publisher, str):
return publisher
elif isinstance(publisher, list):
"""
objects = cls.tagged.with_all(tags)
- return objects.exclude(ancestor__in=objects)
+ return objects.filter(findable=True).exclude(ancestor__in=objects)
@classmethod
def book_list(cls, book_filter=None):
"""
books_by_parent = {}
- books = cls.objects.order_by('parent_number', 'sort_key').only('title', 'parent', 'slug', 'extra_info')
+ books = cls.objects.filter(findable=True).order_by('parent_number', 'sort_key').only('title', 'parent', 'slug', 'extra_info')
if book_filter:
books = books.filter(book_filter).distinct()
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"),
- "SP3": (1, u"szkoła podstawowa"),
- "P": (1, u"szkoła podstawowa"),
- "G": (2, u"gimnazjum"),
- "L": (3, u"liceum"),
- "LP": (3, u"liceum"),
+ "SP": (1, "szkoła podstawowa"),
+ "SP1": (1, "szkoła podstawowa"),
+ "SP2": (1, "szkoła podstawowa"),
+ "SP3": (1, "szkoła podstawowa"),
+ "P": (1, "szkoła podstawowa"),
+ "G": (2, "gimnazjum"),
+ "L": (3, "liceum"),
+ "LP": (3, "liceum"),
}
def audiences_pl(self):
- audiences = self.extra_info.get('audiences', [])
+ audiences = self.get_extra_info_json().get('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')
+ stage = self.get_extra_info_json().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):
+ def choose_fragments(self, number):
fragments = self.fragments.order_by()
fragments_count = fragments.count()
if not fragments_count and self.children.exists():
fragments = Fragment.objects.filter(book__ancestor=self).order_by()
fragments_count = fragments.count()
if fragments_count:
- return fragments[randint(0, fragments_count - 1)]
+ if fragments_count > number:
+ offset = randint(0, fragments_count - number)
+ else:
+ offset = 0
+ return fragments[offset : offset + number]
elif self.parent:
- return self.parent.choose_fragment()
+ return self.parent.choose_fragments(number)
+ else:
+ return []
+
+ def choose_fragment(self):
+ fragments = self.choose_fragments(1)
+ if fragments:
+ return fragments[0]
else:
return None
return self.SORT_KEY_SEP.join((self.sort_key_author, self.sort_key, str(self.id)))
def cover_color(self):
- return WLCover.epoch_colors.get(self.extra_info.get('epoch'), '#000000')
+ return WLCover.epoch_colors.get(self.get_extra_info_json().get('epoch'), '#000000')
@cached_render('catalogue/book_mini_box.html')
def mini_box(self):
'no_link': True,
}
-def add_file_fields():
- for format_ in Book.formats:
- field_name = "%s_file" % format_
- # This weird globals() assignment makes Django migrations comfortable.
- _upload_to = _ebook_upload_to('book/%s/%%s.%s' % (format_, format_))
- _upload_to.__name__ = '_%s_upload_to' % format_
- globals()[_upload_to.__name__] = _upload_to
-
- EbookField(
- format_, _("%s file" % format_.upper()),
- upload_to=_upload_to,
- storage=bofh_storage,
- max_length=255,
- blank=True,
- default=''
- ).contribute_to_class(Book, field_name)
-
-
-add_file_fields()
-
class BookPopularity(models.Model):
book = models.OneToOneField(Book, models.CASCADE, related_name='popularity')
fnp / wolnelektury.git / blobdiff