1 # -*- coding: utf-8 -*-
2 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
3 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
5 from django.db import models
6 from django.db.models import permalink, Q
7 from django.utils.translation import ugettext_lazy as _
8 from django.contrib.auth.models import User
9 from django.core.files import File
10 from django.template.loader import render_to_string
11 from django.template.defaultfilters import slugify
12 from django.utils.safestring import mark_safe
13 from django.utils.translation import get_language
14 from django.core.urlresolvers import reverse
16 from django.conf import settings
18 from newtagging.models import TagBase, tags_updated
19 from newtagging import managers
20 from catalogue.fields import JSONField
22 from librarian import dcparser, html, epub, NoDublinCore
23 from mutagen import id3
27 ('author', _('author')),
28 ('epoch', _('epoch')),
30 ('genre', _('genre')),
31 ('theme', _('theme')),
37 ('odt', _('ODT file')),
38 ('mp3', _('MP3 file')),
39 ('ogg', _('OGG file')),
40 ('daisy', _('DAISY file')),
43 class TagSubcategoryManager(models.Manager):
44 def __init__(self, subcategory):
45 super(TagSubcategoryManager, self).__init__()
46 self.subcategory = subcategory
48 def get_query_set(self):
49 return super(TagSubcategoryManager, self).get_query_set().filter(category=self.subcategory)
53 name = models.CharField(_('name'), max_length=50, db_index=True)
54 slug = models.SlugField(_('slug'), max_length=120, db_index=True)
55 sort_key = models.CharField(_('sort key'), max_length=120, db_index=True)
56 category = models.CharField(_('category'), max_length=50, blank=False, null=False,
57 db_index=True, choices=TAG_CATEGORIES)
58 description = models.TextField(_('description'), blank=True)
59 main_page = models.BooleanField(_('main page'), default=False, db_index=True, help_text=_('Show tag on main page'))
61 user = models.ForeignKey(User, blank=True, null=True)
62 book_count = models.IntegerField(_('book count'), blank=False, null=True)
63 gazeta_link = models.CharField(blank=True, max_length=240)
64 wiki_link = models.CharField(blank=True, max_length=240)
74 categories_dict = dict((item[::-1] for item in categories_rev.iteritems()))
77 ordering = ('sort_key',)
78 verbose_name = _('tag')
79 verbose_name_plural = _('tags')
80 unique_together = (("slug", "category"),)
82 def __unicode__(self):
86 return "Tag(slug=%r)" % self.slug
89 def get_absolute_url(self):
90 return ('catalogue.views.tagged_object_list', [self.url_chunk])
92 def has_description(self):
93 return len(self.description) > 0
94 has_description.short_description = _('description')
95 has_description.boolean = True
98 """ returns global book count for book tags, fragment count for themes """
100 if self.book_count is None:
101 if self.category == 'book':
103 objects = Book.objects.none()
104 elif self.category == 'theme':
105 objects = Fragment.tagged.with_all((self,))
107 objects = Book.tagged.with_all((self,)).order_by()
108 if self.category != 'set':
109 # eliminate descendants
110 l_tags = Tag.objects.filter(slug__in=[book.book_tag_slug() for book in objects])
111 descendants_keys = [book.pk for book in Book.tagged.with_any(l_tags)]
113 objects = objects.exclude(pk__in=descendants_keys)
114 self.book_count = objects.count()
116 return self.book_count
119 def get_tag_list(tags):
120 if isinstance(tags, basestring):
124 tags_splitted = tags.split('/')
125 for index, name in enumerate(tags_splitted):
126 if name in Tag.categories_rev:
127 category = Tag.categories_rev[name]
130 real_tags.append(Tag.objects.get(slug=name, category=category))
134 real_tags.append(Tag.objects.exclude(category='book').get(slug=name))
135 except Tag.MultipleObjectsReturned, e:
136 ambiguous_slugs.append(name)
139 # something strange left off
140 raise Tag.DoesNotExist()
142 # some tags should be qualified
143 e = Tag.MultipleObjectsReturned()
145 e.ambiguous_slugs = ambiguous_slugs
150 return TagBase.get_tag_list(tags)
154 return '/'.join((Tag.categories_dict[self.category], self.slug))
157 # TODO: why is this hard-coded ?
158 def book_upload_path(ext=None):
159 def get_dynamic_path(media, filename, ext=ext):
160 # how to put related book's slug here?
164 name = slugify(filename.split(".")[0])
166 name = slugify(media.name)
167 return 'lektura/%s.%s' % (name, ext)
168 return get_dynamic_path
171 class BookMedia(models.Model):
172 type = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
173 name = models.CharField(_('name'), max_length="100", blank=True)
174 file = models.FileField(_('file'), upload_to=book_upload_path(), blank=True)
175 uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)
177 def __unicode__(self):
178 return "%s (%s)" % (self.name, self.file.name.split("/")[-1])
181 ordering = ('type', 'name')
182 verbose_name = _('book media')
183 verbose_name_plural = _('book media')
185 class Book(models.Model):
186 title = models.CharField(_('title'), max_length=120)
187 slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
188 description = models.TextField(_('description'), blank=True)
189 created_at = models.DateTimeField(_('creation date'), auto_now_add=True)
190 _short_html = models.TextField(_('short HTML'), editable=False)
191 parent_number = models.IntegerField(_('parent number'), default=0)
192 extra_info = JSONField(_('extra information'))
193 gazeta_link = models.CharField(blank=True, max_length=240)
194 wiki_link = models.CharField(blank=True, max_length=240)
195 # files generated during publication
196 xml_file = models.FileField(_('XML file'), upload_to=book_upload_path('xml'), blank=True)
197 html_file = models.FileField(_('HTML file'), upload_to=book_upload_path('html'), blank=True)
198 pdf_file = models.FileField(_('PDF file'), upload_to=book_upload_path('pdf'), blank=True)
199 epub_file = models.FileField(_('EPUB file'), upload_to=book_upload_path('epub'), blank=True)
200 txt_file = models.FileField(_('TXT file'), upload_to=book_upload_path('txt'), blank=True)
202 medias = models.ManyToManyField(BookMedia, blank=True)
204 parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
205 objects = models.Manager()
206 tagged = managers.ModelTaggedItemManager(Tag)
207 tags = managers.TagDescriptor(Tag)
209 _tag_counter = JSONField(null=True, editable=False)
210 _theme_counter = JSONField(null=True, editable=False)
212 class AlreadyExists(Exception):
216 ordering = ('title',)
217 verbose_name = _('book')
218 verbose_name_plural = _('books')
220 def __unicode__(self):
223 def save(self, force_insert=False, force_update=False, reset_short_html=True, refresh_mp3=True, **kwargs):
225 # Reset _short_html during save
227 for key in filter(lambda x: x.startswith('_short_html'), self.__dict__):
229 self.__setattr__(key, '')
230 # Fragment.short_html relies on book's tags, so reset it here too
231 self.fragments.all().update(**update)
233 book = super(Book, self).save(force_insert, force_update)
235 if refresh_mp3 and self.has_media('mp3'):
236 file = self.get_media('mp3')[0]
237 #print file, file.path
238 extra_info = self.get_extra_info_value()
239 extra_info.update(self.get_mp3_info())
240 self.set_extra_info_value(extra_info)
241 book = super(Book, self).save(force_insert, force_update)
245 def get_absolute_url(self):
246 return ('catalogue.views.book_detail', [self.slug])
252 def book_tag_slug(self):
253 return ('l-' + self.slug)[:120]
256 slug = self.book_tag_slug()
257 book_tag, created = Tag.objects.get_or_create(slug=slug, category='book')
259 book_tag.name = self.title[:50]
260 book_tag.sort_key = self.title.lower()
264 def has_media(self, type):
291 if self.medias.filter(book=self, type=type).count() > 0:
296 def get_media(self, type):
297 if self.has_media(type):
301 return self.html_file
303 return self.html_file
309 return self.medias.filter(book=self, type=type)
314 return self.get_media("mp3")
316 return self.get_media("odt")
318 return self.get_media("ogg")
320 return self.get_media("daisy")
322 def short_html(self):
323 key = '_short_html_%s' % get_language()
324 short_html = getattr(self, key)
326 if short_html and len(short_html):
327 return mark_safe(short_html)
329 tags = self.tags.filter(~Q(category__in=('set', 'theme', 'book')))
330 tags = [mark_safe(u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name)) for tag in tags]
333 # files generated during publication
334 if self.has_media("html"):
335 formats.append(u'<a href="%s">%s</a>' % (reverse('book_text', kwargs={'slug': self.slug}), _('Read online')))
336 if self.has_media("pdf"):
337 formats.append(u'<a href="%s">PDF</a>' % self.get_media('pdf').url)
338 if self.root_ancestor.has_media("epub"):
339 formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.get_media('epub').url)
340 if self.has_media("odt"):
341 formats.append(u'<a href="%s">ODT</a>' % self.get_media('odt').url)
342 if self.has_media("txt"):
343 formats.append(u'<a href="%s">TXT</a>' % self.get_media('txt').url)
345 for m in self.medias.order_by('type'):
346 formats.append(u'<a href="%s">%s</a>' % m.type, m.file.url)
348 formats = [mark_safe(format) for format in formats]
350 setattr(self, key, unicode(render_to_string('catalogue/book_short.html',
351 {'book': self, 'tags': tags, 'formats': formats})))
352 self.save(reset_short_html=False)
353 return mark_safe(getattr(self, key))
357 def root_ancestor(self):
358 """ returns the oldest ancestor """
360 if not hasattr(self, '_root_ancestor'):
364 self._root_ancestor = book
365 return self._root_ancestor
368 def get_mp3_info(self):
369 """Retrieves artist and director names from audio ID3 tags."""
370 audio = id3.ID3(self.get_media('mp3')[0].file.path)
371 artist_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE1'))
372 director_name = ', '.join(', '.join(tag.text) for tag in audio.getall('TPE3'))
373 return {'artist_name': artist_name, 'director_name': director_name}
375 def has_description(self):
376 return len(self.description) > 0
377 has_description.short_description = _('description')
378 has_description.boolean = True
381 def has_pdf_file(self):
382 return bool(self.pdf_file)
383 has_pdf_file.short_description = 'PDF'
384 has_pdf_file.boolean = True
386 def has_epub_file(self):
387 return bool(self.epub_file)
388 has_epub_file.short_description = 'EPUB'
389 has_epub_file.boolean = True
391 def has_html_file(self):
392 return bool(self.html_file)
393 has_html_file.short_description = 'HTML'
394 has_html_file.boolean = True
396 def has_odt_file(self):
397 return bool(self.has_media("odt"))
398 has_odt_file.short_description = 'ODT'
399 has_odt_file.boolean = True
401 def has_mp3_file(self):
402 return bool(self.has_media("mp3"))
403 has_mp3_file.short_description = 'MP3'
404 has_mp3_file.boolean = True
406 def has_ogg_file(self):
407 return bool(self.has_media("ogg"))
408 has_ogg_file.short_description = 'OGG'
409 has_ogg_file.boolean = True
411 def has_daisy_file(self):
412 return bool(self.has_media("daisy"))
413 has_daisy_file.short_description = 'DAISY'
414 has_daisy_file.boolean = True
416 def build_epub(self, remove_descendants=True):
417 """ (Re)builds the epub file.
418 If book has a parent, does nothing.
419 Unless remove_descendants is False, descendants' epubs are removed.
422 from StringIO import StringIO
423 from hashlib import sha1
424 from django.core.files.base import ContentFile
425 from librarian import DocProvider
427 class BookImportDocProvider(DocProvider):
428 """ used for joined EPUBs """
430 def __init__(self, book):
433 def by_slug(self, slug):
434 if slug == self.book.slug:
435 return self.book.xml_file
437 return Book.objects.get(slug=slug).xml_file
443 epub_file = StringIO()
445 epub.transform(BookImportDocProvider(self), self.slug, output_file=epub_file)
446 self.epub_file.save('%s.epub' % self.slug, ContentFile(epub_file.getvalue()), save=False)
447 self.save(refresh_mp3=False)
448 FileRecord(slug=self.slug, type='epub', sha1=sha1(epub_file.getvalue()).hexdigest()).save()
452 book_descendants = list(self.children.all())
453 while len(book_descendants) > 0:
454 child_book = book_descendants.pop(0)
455 if remove_descendants and child_book.has_epub_file():
456 child_book.epub_file.delete()
457 # save anyway, to refresh short_html
458 child_book.save(refresh_mp3=False)
459 book_descendants += list(child_book.children.all())
463 def from_xml_file(cls, xml_file, overwrite=False, build_epub=True):
464 # use librarian to parse meta-data
465 book_info = dcparser.parse(xml_file)
467 if not isinstance(xml_file, File):
468 xml_file = File(open(xml_file))
471 return cls.from_text_and_meta(xml_file, book_info, overwrite, build_epub=build_epub)
476 def from_text_and_meta(cls, raw_file, book_info, overwrite=False, build_epub=True):
477 from tempfile import NamedTemporaryFile
478 from slughifi import slughifi
479 from markupstring import MarkupString
480 from django.core.files.storage import default_storage
482 # check for parts before we do anything
484 if hasattr(book_info, 'parts'):
485 for part_url in book_info.parts:
486 base, slug = part_url.rsplit('/', 1)
488 children.append(Book.objects.get(slug=slug))
489 except Book.DoesNotExist, e:
490 raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug)
494 book_base, book_slug = book_info.url.rsplit('/', 1)
495 book, created = Book.objects.get_or_create(slug=book_slug)
501 raise Book.AlreadyExists(_('Book %s already exists') % book_slug)
502 # Save shelves for this book
503 book_shelves = list(book.tags.filter(category='set'))
505 book.title = book_info.title
506 book.set_extra_info_value(book_info.to_dict())
507 book._short_html = ''
511 categories = (('kinds', 'kind'), ('genres', 'genre'), ('authors', 'author'), ('epochs', 'epoch'))
512 for field_name, category in categories:
514 tag_names = getattr(book_info, field_name)
516 tag_names = [getattr(book_info, category)]
517 for tag_name in tag_names:
518 tag_sort_key = tag_name
519 if category == 'author':
520 tag_sort_key = tag_name.last_name
521 tag_name = ' '.join(tag_name.first_names) + ' ' + tag_name.last_name
522 tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name), category=category)
525 tag.sort_key = tag_sort_key.lower()
527 book_tags.append(tag)
529 book.tags = book_tags + book_shelves
531 book_tag = book.book_tag()
533 for n, child_book in enumerate(children):
534 child_book.parent = book
535 child_book.parent_number = n
538 # Save XML and HTML files
539 book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
541 # delete old fragments when overwriting
542 book.fragments.all().delete()
544 html_file = NamedTemporaryFile()
545 if html.transform(book.xml_file.path, html_file, parse_dublincore=False):
546 book.html_file.save('%s.html' % book.slug, File(html_file), save=False)
548 # get ancestor l-tags for adding to new fragments
552 ancestor_tags.append(p.book_tag())
556 closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
557 for fragment in closed_fragments.values():
559 theme_names = [s.strip() for s in fragment.themes.split(',')]
560 except AttributeError:
563 for theme_name in theme_names:
566 tag, created = Tag.objects.get_or_create(slug=slughifi(theme_name), category='theme')
568 tag.name = theme_name
569 tag.sort_key = theme_name.lower()
575 text = fragment.to_string()
577 if (len(MarkupString(text)) > 240):
578 short_text = unicode(MarkupString(text)[:160])
579 new_fragment, created = Fragment.objects.get_or_create(anchor=fragment.id, book=book,
580 defaults={'text': text, 'short_text': short_text})
583 new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags)
585 if not settings.NO_BUILD_EPUB and build_epub:
586 book.root_ancestor().build_epub()
588 book_descendants = list(book.children.all())
589 # add l-tag to descendants and their fragments
590 # delete unnecessary EPUB files
591 while len(book_descendants) > 0:
592 child_book = book_descendants.pop(0)
593 child_book.tags = list(child_book.tags) + [book_tag]
595 for fragment in child_book.fragments.all():
596 fragment.tags = set(list(fragment.tags) + [book_tag])
597 book_descendants += list(child_book.children.all())
600 book.reset_tag_counter()
601 book.reset_theme_counter()
607 def refresh_tag_counter(self):
609 for child in self.children.all().order_by():
610 for tag_pk, value in child.tag_counter.iteritems():
611 tags[tag_pk] = tags.get(tag_pk, 0) + value
612 for tag in self.tags.exclude(category__in=('book', 'theme', 'set')).order_by():
614 self.set__tag_counter_value(tags)
615 self.save(reset_short_html=False, refresh_mp3=False)
618 def reset_tag_counter(self):
619 self._tag_counter = None
620 self.save(reset_short_html=False, refresh_mp3=False)
622 self.parent.reset_tag_counter()
625 def tag_counter(self):
626 if self._tag_counter is None:
627 return self.refresh_tag_counter()
628 return dict((int(k), v) for k, v in self.get__tag_counter_value().iteritems())
630 def refresh_theme_counter(self):
632 for fragment in Fragment.tagged.with_any([self.book_tag()]).order_by():
633 for tag in fragment.tags.filter(category='theme').order_by():
634 tags[tag.pk] = tags.get(tag.pk, 0) + 1
635 self.set__theme_counter_value(tags)
636 self.save(reset_short_html=False, refresh_mp3=False)
639 def reset_theme_counter(self):
640 self._theme_counter = None
641 self.save(reset_short_html=False, refresh_mp3=False)
643 self.parent.reset_theme_counter()
646 def theme_counter(self):
647 if self._theme_counter is None:
648 return self.refresh_theme_counter()
649 return dict((int(k), v) for k, v in self.get__theme_counter_value().iteritems())
651 def pretty_title(self, html_links=False):
653 names = list(book.tags.filter(category='author'))
659 names.extend(reversed(books))
662 names = ['<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) for tag in names]
664 names = [tag.name for tag in names]
666 return ', '.join(names)
669 class Fragment(models.Model):
670 text = models.TextField()
671 short_text = models.TextField(editable=False)
672 _short_html = models.TextField(editable=False)
673 anchor = models.CharField(max_length=120)
674 book = models.ForeignKey(Book, related_name='fragments')
676 objects = models.Manager()
677 tagged = managers.ModelTaggedItemManager(Tag)
678 tags = managers.TagDescriptor(Tag)
681 ordering = ('book', 'anchor',)
682 verbose_name = _('fragment')
683 verbose_name_plural = _('fragments')
685 def get_absolute_url(self):
686 return '%s#m%s' % (reverse('book_text', kwargs={'slug': self.book.slug}), self.anchor)
688 def short_html(self):
689 key = '_short_html_%s' % get_language()
690 short_html = getattr(self, key)
691 if short_html and len(short_html):
692 return mark_safe(short_html)
694 setattr(self, key, unicode(render_to_string('catalogue/fragment_short.html',
695 {'fragment': self})))
697 return mark_safe(getattr(self, key))
700 class FileRecord(models.Model):
701 slug = models.SlugField(_('slug'), max_length=120, db_index=True)
702 type = models.CharField(_('type'), max_length=20, db_index=True)
703 sha1 = models.CharField(_('sha-1 hash'), max_length=40)
704 time = models.DateTimeField(_('time'), auto_now_add=True)
707 ordering = ('-time','-slug', '-type')
708 verbose_name = _('file record')
709 verbose_name_plural = _('file records')
711 def __unicode__(self):
712 return "%s %s.%s" % (self.sha1, self.slug, self.type)
715 def _tags_updated_handler(sender, affected_tags, **kwargs):
716 # reset tag global counter
717 Tag.objects.filter(pk__in=[tag.pk for tag in affected_tags]).update(book_count=None)
719 # if book tags changed, reset book tag counter
720 if isinstance(sender, Book) and \
721 Tag.objects.filter(pk__in=(tag.pk for tag in affected_tags)).\
722 exclude(category__in=('book', 'theme', 'set')).count():
723 sender.reset_tag_counter()
724 # if fragment theme changed, reset book theme counter
725 elif isinstance(sender, Fragment) and \
726 Tag.objects.filter(pk__in=(tag.pk for tag in affected_tags)).\
727 filter(category='theme').count():
728 sender.book.reset_theme_counter()
729 tags_updated.connect(_tags_updated_handler)