X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/9e41863c3cff14cbbc65b783ee57120fb59fa22c..afdd9bc42c8e08e0e2fbade6b9c934a3d65f4885:/apps/catalogue/models.py
diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py
index a5b0ebf0c..a180c650c 100644
--- a/apps/catalogue/models.py
+++ b/apps/catalogue/models.py
@@ -2,7 +2,7 @@
# 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 collections import namedtuple
from django.db import models
from django.db.models import permalink, Q
@@ -23,13 +23,16 @@ from django.conf import settings
from newtagging.models import TagBase, tags_updated
from newtagging import managers
from catalogue.fields import JSONField, OverwritingFileField
-from catalogue.utils import create_zip
+from catalogue.utils import create_zip, split_tags
+from catalogue.tasks import touch_tag, index_book
from shutil import copy
from glob import glob
import re
from os import path
+import search
+
TAG_CATEGORIES = (
('author', _('author')),
('epoch', _('epoch')),
@@ -40,13 +43,6 @@ TAG_CATEGORIES = (
('book', _('book')),
)
-MEDIA_FORMATS = (
- ('odt', _('ODT file')),
- ('mp3', _('MP3 file')),
- ('ogg', _('OGG file')),
- ('daisy', _('DAISY file')),
-)
-
# not quite, but Django wants you to set a timeout
CACHE_FOREVER = 2419200 # 28 days
@@ -67,7 +63,6 @@ class Tag(TagBase):
category = models.CharField(_('category'), max_length=50, blank=False, null=False,
db_index=True, choices=TAG_CATEGORIES)
description = models.TextField(_('description'), blank=True)
- 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=True, null=True)
@@ -112,25 +107,22 @@ class Tag(TagBase):
has_description.boolean = True
def get_count(self):
- """ returns global book count for book tags, fragment count for themes """
-
- if self.book_count is None:
- if self.category == 'book':
- # never used
- objects = Book.objects.none()
- elif self.category == 'theme':
- objects = Fragment.tagged.with_all((self,))
- else:
- objects = Book.tagged.with_all((self,)).order_by()
- if self.category != 'set':
- # eliminate descendants
- l_tags = Tag.objects.filter(slug__in=[book.book_tag_slug() for book in objects])
- descendants_keys = [book.pk for book in Book.tagged.with_any(l_tags)]
- if descendants_keys:
- objects = objects.exclude(pk__in=descendants_keys)
- self.book_count = objects.count()
- self.save()
- return self.book_count
+ """Returns global book count for book tags, fragment count for themes."""
+
+ if self.category == 'book':
+ # never used
+ objects = Book.objects.none()
+ elif self.category == 'theme':
+ objects = Fragment.tagged.with_all((self,))
+ else:
+ objects = Book.tagged.with_all((self,)).order_by()
+ if self.category != 'set':
+ # eliminate descendants
+ l_tags = Tag.objects.filter(slug__in=[book.book_tag_slug() for book in objects])
+ descendants_keys = [book.pk for book in Book.tagged.with_any(l_tags)]
+ if descendants_keys:
+ objects = objects.exclude(pk__in=descendants_keys)
+ return objects.count()
@staticmethod
def get_tag_list(tags):
@@ -174,16 +166,43 @@ class Tag(TagBase):
def url_chunk(self):
return '/'.join((Tag.categories_dict[self.category], self.slug))
+ @staticmethod
+ def tags_from_info(info):
+ from slughifi import slughifi
+ from sortify import sortify
+ meta_tags = []
+ categories = (('kinds', 'kind'), ('genres', 'genre'), ('authors', 'author'), ('epochs', 'epoch'))
+ for field_name, category in categories:
+ try:
+ tag_names = getattr(info, field_name)
+ except:
+ try:
+ tag_names = [getattr(info, category)]
+ except:
+ # For instance, Pictures do not have 'genre' field.
+ continue
+ for tag_name in tag_names:
+ tag_sort_key = tag_name
+ if category == 'author':
+ tag_sort_key = tag_name.last_name
+ tag_name = tag_name.readable()
+ tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
+ if created:
+ tag.name = tag_name
+ tag.sort_key = sortify(tag_sort_key.lower())
+ tag.save()
+ meta_tags.append(tag)
+ return meta_tags
+
+
def get_dynamic_path(media, filename, ext=None, maxlen=100):
from slughifi import slughifi
# how to put related book's slug here?
if not ext:
- if media.type == 'daisy':
- ext = 'daisy.zip'
- else:
- ext = media.type
+ # BookMedia case
+ ext = media.formats[media.type].ext
if media is None or not media.name:
name = slughifi(filename.split(".")[0])
else:
@@ -220,7 +239,16 @@ def get_existing_customized_pdf(book):
class BookMedia(models.Model):
- type = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
+ FileFormat = namedtuple("FileFormat", "name ext")
+ formats = SortedDict([
+ ('mp3', FileFormat(name='MP3', ext='mp3')),
+ ('ogg', FileFormat(name='Ogg Vorbis', ext='ogg')),
+ ('daisy', FileFormat(name='DAISY', ext='daisy.zip')),
+ ])
+ format_choices = [(k, _('%s file') % t.name)
+ for k, t in formats.items()]
+
+ type = models.CharField(_('type'), choices=format_choices, max_length="100")
name = models.CharField(_('name'), max_length="100")
file = OverwritingFileField(_('file'), upload_to=book_upload_path())
uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)
@@ -333,8 +361,11 @@ class Book(models.Model):
wiki_link = models.CharField(blank=True, max_length=240)
# files generated during publication
- file_types = ['epub', 'html', 'mobi', 'pdf', 'txt', 'xml']
-
+ cover = models.FileField(_('cover'), upload_to=book_upload_path('png'),
+ null=True, blank=True)
+ ebook_formats = ['pdf', 'epub', 'mobi', 'txt']
+ formats = ebook_formats + ['html', 'xml']
+
parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
objects = models.Manager()
tagged = managers.ModelTaggedItemManager(Tag)
@@ -431,14 +462,14 @@ class Book(models.Model):
return book_tag
def has_media(self, type):
- if type in Book.file_types:
+ if type in Book.formats:
return bool(getattr(self, "%s_file" % type))
else:
return self.media.filter(type=type).exists()
def get_media(self, type):
if self.has_media(type):
- if type in Book.file_types:
+ if type in Book.formats:
return getattr(self, "%s_file" % type)
else:
return self.media.filter(type=type)
@@ -461,6 +492,7 @@ class Book(models.Model):
cache_key = "Book.short_html/%d/%s"
for lang, langname in settings.LANGUAGES:
cache.delete(cache_key % (self.id, lang))
+ cache.delete("Book.mini_box/%d" % (self.id, ))
# Fragment.short_html relies on book's tags, so reset it here too
for fragm in self.fragments.all():
fragm.reset_short_html()
@@ -475,24 +507,17 @@ class Book(models.Model):
if short_html is not None:
return mark_safe(short_html)
else:
- tags = self.tags.filter(~Q(category__in=('set', 'theme', 'book')))
- tags = [mark_safe(u'%s' % (tag.get_absolute_url(), tag.name)) for tag in tags]
+ tags = self.tags.filter(category__in=('author', 'kind', 'genre', 'epoch'))
+ tags = split_tags(tags)
formats = []
# files generated during publication
- if self.has_media("html"):
- formats.append(u'%s' % (reverse('book_text', args=[self.fileid()]), _('Read online')))
- if self.has_media("pdf"):
- formats.append(u'PDF' % self.get_media('pdf').url)
- if self.has_media("mobi"):
- formats.append(u'MOBI' % self.get_media('mobi').url)
- if self.root_ancestor.has_media("epub"):
- formats.append(u'EPUB' % self.root_ancestor.get_media('epub').url)
- if self.has_media("txt"):
- formats.append(u'TXT' % self.get_media('txt').url)
- # other files
- for m in self.media.order_by('type'):
- formats.append(u'%s' % (m.file.url, m.type.upper()))
+ for ebook_format in self.ebook_formats:
+ if self.has_media(ebook_format):
+ formats.append(u'%s' % (
+ "", #self.get_media(ebook_format).url,
+ ebook_format.upper()
+ ))
formats = [mark_safe(format) for format in formats]
@@ -505,7 +530,7 @@ class Book(models.Model):
def mini_box(self):
if self.id:
- cache_key = "Book.mini_boxs/%d" % (self.id, )
+ cache_key = "Book.mini_box/%d" % (self.id, )
short_html = cache.get(cache_key)
else:
short_html = None
@@ -520,29 +545,12 @@ class Book(models.Model):
cache.set(cache_key, short_html, CACHE_FOREVER)
return mark_safe(short_html)
- @property
- def root_ancestor(self):
- """ returns the oldest ancestor """
-
- if not hasattr(self, '_root_ancestor'):
- book = self
- while book.parent:
- book = book.parent
- self._root_ancestor = book
- return self._root_ancestor
-
-
def has_description(self):
return len(self.description) > 0
has_description.short_description = _('description')
has_description.boolean = True
# ugly ugly ugly
- def has_odt_file(self):
- return bool(self.has_media("odt"))
- has_odt_file.short_description = 'ODT'
- has_odt_file.boolean = True
-
def has_mp3_file(self):
return bool(self.has_media("mp3"))
has_mp3_file.short_description = 'MP3'
@@ -566,6 +574,20 @@ class Book(models.Model):
provider=ORMDocProvider(self),
parse_dublincore=parse_dublincore)
+ def build_cover(self, book_info=None):
+ """(Re)builds the cover image."""
+ from StringIO import StringIO
+ from django.core.files.base import ContentFile
+ from librarian.cover import WLCover
+
+ if book_info is None:
+ book_info = self.wldocument().book_info
+
+ cover = WLCover(book_info).image()
+ imgstr = StringIO()
+ cover.save(imgstr, 'png')
+ self.cover.save(None, ContentFile(imgstr.getvalue()))
+
def build_pdf(self, customizations=None, file_name=None):
""" (Re)builds the pdf file.
customizations - customizations which are passed to LaTeX class file.
@@ -609,34 +631,15 @@ class Book(models.Model):
# remove zip with all mobi files
remove_zip(settings.ALL_MOBI_ZIP)
- def build_epub(self, remove_descendants=True):
- """ (Re)builds the epub file.
- If book has a parent, does nothing.
- Unless remove_descendants is False, descendants' epubs are removed.
- """
+ def build_epub(self):
+ """(Re)builds the epub file."""
from django.core.files import File
from catalogue.utils import remove_zip
- if self.parent:
- # don't need an epub
- return
-
epub = self.wldocument().as_epub()
- try:
- epub.transform(ORMDocProvider(self), self.fileid(), output_file=epub_file)
- self.epub_file.save('%s.epub' % self.fileid(), File(open(epub.get_filename())))
- except NoDublinCore:
- pass
-
- book_descendants = list(self.children.all())
- while len(book_descendants) > 0:
- child_book = book_descendants.pop(0)
- if remove_descendants and child_book.has_epub_file():
- child_book.epub_file.delete()
- # save anyway, to refresh short_html
- child_book.save()
- book_descendants += list(child_book.children.all())
+ self.epub_file.save('%s.epub' % self.fileid(),
+ File(open(epub.get_filename())))
# remove zip package with all epub files
remove_zip(settings.ALL_EPUB_ZIP)
@@ -728,6 +731,19 @@ class Book(models.Model):
result = create_zip.delay(paths, self.fileid())
return result.wait()
+ def search_index(self, book_info=None):
+ if settings.CELERY_ALWAYS_EAGER:
+ idx = search.ReusableIndex()
+ else:
+ idx = search.Index()
+
+ idx.open()
+ try:
+ idx.index_book(self, book_info)
+ idx.index_tags()
+ finally:
+ idx.close()
+
@classmethod
def from_xml_file(cls, xml_file, **kwargs):
from django.core.files import File
@@ -746,9 +762,9 @@ class Book(models.Model):
@classmethod
def from_text_and_meta(cls, raw_file, book_info, overwrite=False,
- build_epub=True, build_txt=True, build_pdf=True, build_mobi=True):
+ build_epub=True, build_txt=True, build_pdf=True, build_mobi=True,
+ search_index=True):
import re
- from slughifi import slughifi
from sortify import sortify
# check for parts before we do anything
@@ -783,24 +799,7 @@ class Book(models.Model):
book.set_extra_info_value(book_info.to_dict())
book.save()
- meta_tags = []
- categories = (('kinds', 'kind'), ('genres', 'genre'), ('authors', 'author'), ('epochs', 'epoch'))
- for field_name, category in categories:
- try:
- tag_names = getattr(book_info, field_name)
- except:
- tag_names = [getattr(book_info, category)]
- for tag_name in tag_names:
- tag_sort_key = tag_name
- if category == 'author':
- tag_sort_key = tag_name.last_name
- tag_name = ' '.join(tag_name.first_names) + ' ' + tag_name.last_name
- tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
- if created:
- tag.name = tag_name
- tag.sort_key = sortify(tag_sort_key.lower())
- tag.save()
- meta_tags.append(tag)
+ meta_tags = Tag.tags_from_info(book_info)
book.tags = set(meta_tags + book_shelves)
@@ -821,26 +820,35 @@ class Book(models.Model):
if not settings.NO_BUILD_TXT and build_txt:
book.build_txt()
+ book.build_cover(book_info)
+
if not settings.NO_BUILD_EPUB and build_epub:
- book.root_ancestor.build_epub()
+ book.build_epub()
if not settings.NO_BUILD_PDF and build_pdf:
- book.root_ancestor.build_pdf()
+ book.build_pdf()
if not settings.NO_BUILD_MOBI and build_mobi:
book.build_mobi()
+ if not settings.NO_SEARCH_INDEX and search_index:
+ index_book.delay(book.id, book_info)
+
book_descendants = list(book.children.all())
+ descendants_tags = set()
# add l-tag to descendants and their fragments
- # delete unnecessary EPUB files
while len(book_descendants) > 0:
child_book = book_descendants.pop(0)
+ descendants_tags.update(child_book.tags)
child_book.tags = list(child_book.tags) + [book_tag]
child_book.save()
for fragment in child_book.fragments.all():
fragment.tags = set(list(fragment.tags) + [book_tag])
book_descendants += list(child_book.children.all())
+ for tag in descendants_tags:
+ touch_tag.delay(tag)
+
book.save()
# refresh cache
@@ -1002,7 +1010,7 @@ def _has_factory(ftype):
# add the file fields
-for t in Book.file_types:
+for t in Book.formats:
field_name = "%s_file" % t
models.FileField(_("%s file" % t.upper()),
upload_to=book_upload_path(t),
@@ -1027,7 +1035,7 @@ class Fragment(models.Model):
verbose_name_plural = _('fragments')
def get_absolute_url(self):
- return '%s#m%s' % (self.book.get_html_url(), self.anchor)
+ return '%s#m%s' % (reverse('book_text', args=[self.book.slug]), self.anchor)
def reset_short_html(self):
if self.id is None:
@@ -1064,7 +1072,8 @@ class Fragment(models.Model):
def _tags_updated_handler(sender, affected_tags, **kwargs):
# reset tag global counter
# 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())
+ for tag in affected_tags:
+ touch_tag.delay(tag)
# if book tags changed, reset book tag counter
if isinstance(sender, Book) and \