X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/645219bfa68384dad930269b94e2ddf8b4cd3488..0098b333f963f74628c4fb0768b6dd56d9c6a07b:/apps/catalogue/models.py diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py index 455296217..272228627 100644 --- a/apps/catalogue/models.py +++ b/apps/catalogue/models.py @@ -2,6 +2,8 @@ # 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 _ @@ -17,11 +19,14 @@ from django.conf import settings from newtagging.models import TagBase, tags_updated from newtagging import managers -from catalogue.fields import JSONField +from catalogue.fields import JSONField, OverwritingFileField +from catalogue.utils import ExistingFile from librarian import dcparser, html, epub, NoDublinCore +import mutagen from mutagen import id3 from slughifi import slughifi +from sortify import sortify TAG_CATEGORIES = ( @@ -64,6 +69,12 @@ class Tag(TagBase): 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) + + class UrlDeprecationWarning(DeprecationWarning): + pass + categories_rev = { 'autor': 'author', 'epoka': 'epoch', @@ -122,19 +133,20 @@ class Tag(TagBase): real_tags = [] ambiguous_slugs = [] category = None + deprecated = False tags_splitted = tags.split('/') - for index, name in enumerate(tags_splitted): - if name in Tag.categories_rev: + for name in tags_splitted: + if category: + real_tags.append(Tag.objects.get(slug=name, category=category)) + category = None + elif name in Tag.categories_rev: category = Tag.categories_rev[name] else: - if category: - real_tags.append(Tag.objects.get(slug=name, category=category)) - category = None - else: - try: - real_tags.append(Tag.objects.exclude(category='book').get(slug=name)) - except Tag.MultipleObjectsReturned, e: - ambiguous_slugs.append(name) + try: + real_tags.append(Tag.objects.exclude(category='book').get(slug=name)) + deprecated = True + except Tag.MultipleObjectsReturned, e: + ambiguous_slugs.append(name) if category: # something strange left off @@ -145,8 +157,11 @@ class Tag(TagBase): e.tags = real_tags e.ambiguous_slugs = ambiguous_slugs raise e - else: - return real_tags + if deprecated: + e = Tag.UrlDeprecationWarning() + e.tags = real_tags + raise e + return real_tags else: return TagBase.get_tag_list(tags) @@ -168,24 +183,18 @@ def book_upload_path(ext=None, maxlen=100): name = slughifi(filename.split(".")[0]) else: name = slughifi(media.name) - return 'lektura/%s.%s' % (name[:maxlen-len('lektura/.%s' % ext)-4], ext) + return 'book/%s/%s.%s' % (ext, name[:maxlen-len('book/%s/.%s' % (ext, 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") - file = models.FileField(_('file'), upload_to=book_upload_path()) + file = OverwritingFileField(_('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('
'.join("%s" % (reverse('admin:catalogue_book_change', args=[b.id]), b.title) for b in self.book_set.all())) - books.short_description = _('books') + extra_info = JSONField(_('extra information'), default='{}', editable=False) + book = models.ForeignKey('Book', related_name='media') + source_sha1 = models.CharField(null=True, blank=True, max_length=40, editable=False) def __unicode__(self): return "%s (%s)" % (self.name, self.file.name.split("/")[-1]) @@ -195,35 +204,87 @@ class BookMedia(models.Model): 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.""" + def save(self, *args, **kwargs): 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} + old = BookMedia.objects.get(pk=self.pk) + except BookMedia.DoesNotExist, e: + pass + else: + # if name changed, change the file name, too + if slughifi(self.name) != slughifi(old.name): + self.file.save(None, ExistingFile(self.file.path), save=False, leave=True) + + super(BookMedia, self).save(*args, **kwargs) + extra_info = self.get_extra_info_value() + extra_info.update(self.read_meta()) + self.set_extra_info_value(extra_info) + self.source_sha1 = self.read_source_sha1(self.file.path, self.type) + return super(BookMedia, self).save(*args, **kwargs) + + def read_meta(self): + """ + Reads some metadata from the audiobook. + """ + + artist_name = director_name = project = funded_by = '' + if self.type == 'mp3': + 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')) + project = ", ".join([t.data for t in audio.getall('PRIV') + if t.owner=='wolnelektury.pl?project']) + funded_by = ", ".join([t.data for t in audio.getall('PRIV') + if t.owner=='wolnelektury.pl?funded_by']) + except: + pass + elif self.type == 'ogg': + try: + audio = mutagen.File(self.file.path) + artist_name = ', '.join(audio.get('artist', [])) + director_name = ', '.join(audio.get('conductor', [])) + project = ", ".join(audio.get('project', [])) + funded_by = ", ".join(audio.get('funded_by', [])) + except: + pass + else: + return {} + return {'artist_name': artist_name, 'director_name': director_name, + 'project': project, 'funded_by': funded_by} + + @staticmethod + def read_source_sha1(filepath, filetype): + """ + Reads source file SHA1 from audiobok metadata. + """ + + if filetype == 'mp3': + try: + audio = id3.ID3(filepath) + return [t.data for t in audio.getall('PRIV') + if t.owner=='wolnelektury.pl?flac_sha1'][0] + except: + return None + elif filetype == 'ogg': + try: + audio = mutagen.File(filepath) + return audio.get('flac_sha1', [None])[0] + except: + return None + else: + return None 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 @@ -232,9 +293,7 @@ class Book(models.Model): pdf_file = models.FileField(_('PDF file'), upload_to=book_upload_path('pdf'), blank=True) epub_file = models.FileField(_('EPUB file'), upload_to=book_upload_path('epub'), blank=True) txt_file = models.FileField(_('TXT file'), upload_to=book_upload_path('txt'), blank=True) - # other files - medias = models.ManyToManyField(BookMedia, blank=True) - + parent = models.ForeignKey('self', blank=True, null=True, related_name='children') objects = models.Manager() tagged = managers.ModelTaggedItemManager(Tag) @@ -247,7 +306,7 @@ class Book(models.Model): pass class Meta: - ordering = ('title',) + ordering = ('sort_key',) verbose_name = _('book') verbose_name_plural = _('books') @@ -255,6 +314,8 @@ class Book(models.Model): return self.title 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 = {} @@ -313,7 +374,7 @@ class Book(models.Model): else: return False else: - if self.medias.filter(book=self, type=type).count() > 0: + if self.media.filter(type=type).exists(): return True else: return False @@ -331,7 +392,7 @@ class Book(models.Model): elif type == "pdf": return self.pdf_file else: - return self.medias.filter(book=self, type=type) + return self.media.filter(type=type) else: return None @@ -365,7 +426,7 @@ class Book(models.Model): if self.has_media("txt"): formats.append(u'TXT' % self.get_media('txt').url) # other files - for m in self.medias.order_by('type'): + for m in self.media.order_by('type'): formats.append(u'%s' % (m.file.url, m.type.upper())) formats = [mark_safe(format) for format in formats] @@ -554,7 +615,7 @@ class Book(models.Model): 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) @@ -614,8 +675,8 @@ class Book(models.Model): new_fragment.save() new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags) - if not settings.NO_BUILD_TXT and build_txt: - book.build_txt() + 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() @@ -700,6 +761,24 @@ class Book(models.Model): 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() @@ -755,7 +834,8 @@ class FileRecord(models.Model): 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 \ @@ -770,23 +850,14 @@ def _tags_updated_handler(sender, affected_tags, **kwargs): 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 """ + """ refresh Book on BookMedia delete """ if sender == BookMedia: - instance.book_set.clear() + instance.book.save() 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() + instance.book.save() post_save.connect(_post_save_handler) -