# This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
+from datetime import datetime
+
from django.db import models
from django.db.models import permalink, Q
from django.utils.translation import ugettext_lazy as _
from django.contrib.auth.models import User
from django.core.files import File
from django.template.loader import render_to_string
-from django.template.defaultfilters import slugify
from django.utils.safestring import mark_safe
from django.utils.translation import get_language
from django.core.urlresolvers import reverse
-from datetime import datetime
+from django.db.models.signals import post_save, m2m_changed, pre_delete
from django.conf import settings
from librarian import dcparser, html, epub, NoDublinCore
from mutagen import id3
+from slughifi import slughifi
+from sortify import sortify
TAG_CATEGORIES = (
main_page = models.BooleanField(_('main page'), default=False, db_index=True, help_text=_('Show tag on main page'))
user = models.ForeignKey(User, blank=True, null=True)
- book_count = models.IntegerField(_('book count'), blank=False, null=True)
- death = models.IntegerField(_(u'year of death'), blank=True, null=True)
+ book_count = models.IntegerField(_('book count'), blank=True, null=True)
gazeta_link = models.CharField(blank=True, max_length=240)
wiki_link = models.CharField(blank=True, max_length=240)
+ created_at = models.DateTimeField(_('creation date'), auto_now_add=True, db_index=True)
+ changed_at = models.DateTimeField(_('creation date'), auto_now=True, db_index=True)
+
categories_rev = {
'autor': 'author',
'epoka': 'epoch',
has_description.short_description = _('description')
has_description.boolean = True
- def alive(self):
- return self.death is None
-
- def in_pd(self):
- """ tests whether an author is in public domain """
- return self.death is not None and self.goes_to_pd() <= datetime.now().year
-
- def goes_to_pd(self):
- """ calculates the year of public domain entry for an author """
- return self.death + 71 if self.death is not None else None
-
def get_count(self):
""" returns global book count for book tags, fragment count for themes """
# TODO: why is this hard-coded ?
-def book_upload_path(ext=None):
+def book_upload_path(ext=None, maxlen=100):
def get_dynamic_path(media, filename, ext=ext):
# how to put related book's slug here?
if not ext:
- ext = media.type
+ if media.type == 'daisy':
+ ext = 'daisy.zip'
+ else:
+ ext = media.type
if not media.name:
- name = slugify(filename.split(".")[0])
+ name = slughifi(filename.split(".")[0])
else:
- name = slugify(media.name)
- return 'lektura/%s.%s' % (name, ext)
+ name = slughifi(media.name)
+ return 'lektura/%s.%s' % (name[:maxlen-len('lektura/.%s' % ext)-4], ext)
return get_dynamic_path
class BookMedia(models.Model):
type = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
- name = models.CharField(_('name'), max_length="100", blank=True)
- file = models.FileField(_('file'), upload_to=book_upload_path(), blank=True)
+ name = models.CharField(_('name'), max_length="100")
+ file = models.FileField(_('file'), upload_to=book_upload_path())
uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)
+ extra_info = JSONField(_('extra information'), default='{}')
+
+ def book_count(self):
+ return self.book_set.count()
+ book_count.short_description = _('book count')
+
+ def books(self):
+ return mark_safe('<br/>'.join("<a href='%s'>%s</a>" % (reverse('admin:catalogue_book_change', args=[b.id]), b.title) for b in self.book_set.all()))
+ books.short_description = _('books')
def __unicode__(self):
return "%s (%s)" % (self.name, self.file.name.split("/")[-1])
verbose_name = _('book media')
verbose_name_plural = _('book media')
+ def save(self, force_insert=False, force_update=False, **kwargs):
+ media = super(BookMedia, self).save(force_insert, force_update, **kwargs)
+ if self.type == 'mp3':
+ file = self.file
+ extra_info = self.get_extra_info_value()
+ extra_info.update(self.get_mp3_info())
+ self.set_extra_info_value(extra_info)
+ media = super(BookMedia, self).save(force_insert, force_update, **kwargs)
+ return media
+
+ def get_mp3_info(self):
+ """Retrieves artist and director names from audio ID3 tags."""
+ try:
+ audio = id3.ID3(self.file.path)
+ artist_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE1'))
+ director_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE3'))
+ except:
+ artist_name = director_name = ''
+ return {'artist_name': artist_name, 'director_name': director_name}
+
+
class Book(models.Model):
title = models.CharField(_('title'), max_length=120)
+ sort_key = models.CharField(_('sort_key'), max_length=120, db_index=True, editable=False)
slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
description = models.TextField(_('description'), blank=True)
- created_at = models.DateTimeField(_('creation date'), auto_now_add=True)
+ created_at = models.DateTimeField(_('creation date'), auto_now_add=True, db_index=True)
+ changed_at = models.DateTimeField(_('creation date'), auto_now=True, db_index=True)
_short_html = models.TextField(_('short HTML'), editable=False)
parent_number = models.IntegerField(_('parent number'), default=0)
- extra_info = JSONField(_('extra information'))
+ extra_info = JSONField(_('extra information'), default='{}')
gazeta_link = models.CharField(blank=True, max_length=240)
wiki_link = models.CharField(blank=True, max_length=240)
# files generated during publication
pass
class Meta:
- ordering = ('title',)
+ ordering = ('sort_key',)
verbose_name = _('book')
verbose_name_plural = _('books')
def __unicode__(self):
return self.title
- def save(self, force_insert=False, force_update=False, reset_short_html=True, refresh_mp3=True, **kwargs):
+ def save(self, force_insert=False, force_update=False, reset_short_html=True, **kwargs):
+ self.sort_key = sortify(self.title)
+
if reset_short_html:
# Reset _short_html during save
update = {}
# Fragment.short_html relies on book's tags, so reset it here too
self.fragments.all().update(**update)
- book = super(Book, self).save(force_insert, force_update)
-
- if refresh_mp3 and self.has_media('mp3'):
- file = self.get_media('mp3')[0]
- #print file, file.path
- extra_info = self.get_extra_info_value()
- extra_info.update(self.get_mp3_info())
- self.set_extra_info_value(extra_info)
- book = super(Book, self).save(force_insert, force_update)
- return book
+ return super(Book, self).save(force_insert, force_update)
@permalink
def get_absolute_url(self):
elif type == "html":
return self.html_file
elif type == "epub":
- return self.html_file
+ return self.epub_file
elif type == "txt":
return self.txt_file
elif type == "pdf":
formats.append(u'<a href="%s">PDF</a>' % self.get_media('pdf').url)
if self.root_ancestor.has_media("epub"):
formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.get_media('epub').url)
- if self.has_media("odt"):
- formats.append(u'<a href="%s">ODT</a>' % self.get_media('odt').url)
if self.has_media("txt"):
formats.append(u'<a href="%s">TXT</a>' % self.get_media('txt').url)
# other files
- for m in self.media.order_by('type'):
- formats.append(u'<a href="%s">%s</a>' % m.type, m.file.url)
-
+ for m in self.medias.order_by('type'):
+ formats.append(u'<a href="%s">%s</a>' % (m.file.url, m.type.upper()))
+
formats = [mark_safe(format) for format in formats]
setattr(self, key, unicode(render_to_string('catalogue/book_short.html',
return self._root_ancestor
- def get_mp3_info(self):
- """Retrieves artist and director names from audio ID3 tags."""
- audio = id3.ID3(self.get_media('mp3')[0].file.path)
- artist_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE1'))
- director_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE3'))
- return {'artist_name': artist_name, 'director_name': director_name}
-
def has_description(self):
return len(self.description) > 0
has_description.short_description = _('description')
has_epub_file.short_description = 'EPUB'
has_epub_file.boolean = True
+ def has_txt_file(self):
+ return bool(self.txt_file)
+ has_txt_file.short_description = 'HTML'
+ has_txt_file.boolean = True
+
def has_html_file(self):
return bool(self.html_file)
has_html_file.short_description = 'HTML'
epub_file = StringIO()
try:
- epub.transform(BookImportDocProvider(self), self.slug, epub_file)
- self.epub_file.save('%s.epub' % self.slug, ContentFile(epub_file.getvalue()), save=False)
- self.save(refresh_mp3=False)
+ epub.transform(BookImportDocProvider(self), self.slug, output_file=epub_file)
+ self.epub_file.save('%s.epub' % self.slug, ContentFile(epub_file.getvalue()))
FileRecord(slug=self.slug, type='epub', sha1=sha1(epub_file.getvalue()).hexdigest()).save()
except NoDublinCore:
pass
if remove_descendants and child_book.has_epub_file():
child_book.epub_file.delete()
# save anyway, to refresh short_html
- child_book.save(refresh_mp3=False)
+ child_book.save()
book_descendants += list(child_book.children.all())
+ def build_txt(self):
+ from StringIO import StringIO
+ from django.core.files.base import ContentFile
+ from librarian import text
+
+ out = StringIO()
+ text.transform(open(self.xml_file.path), out)
+ self.txt_file.save('%s.txt' % self.slug, ContentFile(out.getvalue()))
+ self.save()
+
@classmethod
- def from_xml_file(cls, xml_file, overwrite=False):
+ def from_xml_file(cls, xml_file, **kwargs):
# use librarian to parse meta-data
book_info = dcparser.parse(xml_file)
xml_file = File(open(xml_file))
try:
- return cls.from_text_and_meta(xml_file, book_info, overwrite)
+ 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):
+ def from_text_and_meta(cls, raw_file, book_info, overwrite=False, build_epub=True, build_txt=True):
+ import re
from tempfile import NamedTemporaryFile
- from slughifi import slughifi
from markupstring import MarkupString
from django.core.files.storage import default_storage
# 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:
tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
if created:
tag.name = tag_name
- tag.sort_key = tag_sort_key.lower()
+ tag.sort_key = sortify(tag_sort_key.lower())
tag.save()
book_tags.append(tag)
- book.tags = book_tags + book_shelves
+ book.tags = set(book_tags + book_shelves)
book_tag = book.book_tag()
new_fragment.save()
new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags)
- if not settings.NO_BUILD_EPUB:
- book.root_ancestor().build_epub()
+ 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
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, refresh_mp3=False)
+ self.save(reset_short_html=False)
return tags
def reset_tag_counter(self):
self._tag_counter = None
- self.save(reset_short_html=False, refresh_mp3=False)
+ self.save(reset_short_html=False)
if self.parent:
self.parent.reset_tag_counter()
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, refresh_mp3=False)
+ self.save(reset_short_html=False)
return tags
def reset_theme_counter(self):
self._theme_counter = None
- self.save(reset_short_html=False, refresh_mp3=False)
+ self.save(reset_short_html=False)
if self.parent:
self.parent.reset_theme_counter()
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
+
+
class Fragment(models.Model):
text = models.TextField()
short_text = models.TextField(editable=False)
return mark_safe(getattr(self, key))
-class BookStub(models.Model):
- title = models.CharField(_('title'), max_length=120)
- author = models.CharField(_('author'), max_length=120)
- pd = models.IntegerField(_('goes to public domain'), null=True, blank=True)
- slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
- translator = models.TextField(_('translator'), blank=True)
- translator_death = models.TextField(_('year of translator\'s death'), blank=True)
-
- class Meta:
- ordering = ('title',)
- verbose_name = _('book stub')
- verbose_name_plural = _('book stubs')
-
- def __unicode__(self):
- return self.title
-
- @permalink
- def get_absolute_url(self):
- return ('catalogue.views.book_detail', [self.slug])
-
- def in_pd(self):
- return self.pd is not None and self.pd <= datetime.now().year
-
- @property
- def name(self):
- return self.title
-
-
class FileRecord(models.Model):
slug = models.SlugField(_('slug'), max_length=120, db_index=True)
type = models.CharField(_('type'), max_length=20, db_index=True)
def __unicode__(self):
return "%s %s.%s" % (self.sha1, self.slug, self.type)
+###########
+#
+# SIGNALS
+#
+###########
+
def _tags_updated_handler(sender, affected_tags, **kwargs):
# reset tag global counter
- Tag.objects.filter(pk__in=[tag.pk for tag in affected_tags]).update(book_count=None)
+ # we want Tag.changed_at updated for API to know the tag was touched
+ Tag.objects.filter(pk__in=[tag.pk for tag in affected_tags]).update(book_count=None, changed_at=datetime.now())
# if book tags changed, reset book tag counter
if isinstance(sender, Book) and \
sender.book.reset_theme_counter()
tags_updated.connect(_tags_updated_handler)
+
+def _m2m_changed_handler(sender, instance, action, reverse, pk_set, **kwargs):
+ """ refresh all the short_html stuff on BookMedia delete """
+ if sender == Book.medias.through and reverse and action == 'pre_clear':
+ for book in instance.book_set.all():
+ book.save()
+m2m_changed.connect(_m2m_changed_handler)
+
+def _pre_delete_handler(sender, instance, **kwargs):
+ """ explicitly clear m2m, so that Books can be refreshed """
+ if sender == BookMedia:
+ instance.book_set.clear()
+pre_delete.connect(_pre_delete_handler)
+
+def _post_save_handler(sender, instance, **kwargs):
+ """ refresh all the short_html stuff on BookMedia update """
+ if sender == BookMedia:
+ for book in instance.book_set.all():
+ book.save()
+post_save.connect(_post_save_handler)
+
fnp / wolnelektury.git / blobdiff