from newtagging import managers
from catalogue.fields import JSONField
-from librarian import html, dcparser
+from librarian import dcparser, html, epub, NoDublinCore
from mutagen import id3
title = models.CharField(_('title'), max_length=120)
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=True)
+ created_at = models.DateTimeField(_('creation date'), auto_now_add=True)
_short_html = models.TextField(_('short HTML'), editable=False)
parent_number = models.IntegerField(_('parent number'), default=0)
extra_info = JSONField(_('extra information'))
txt_file = models.FileField(_('TXT file'), upload_to=book_upload_path('txt'), blank=True)
mp3_file = models.FileField(_('MP3 file'), upload_to=book_upload_path('mp3'), blank=True)
ogg_file = models.FileField(_('OGG file'), upload_to=book_upload_path('ogg'), blank=True)
+ daisy_file = models.FileField(_('DAISY file'), upload_to=book_upload_path('daisy.zip'), blank=True)
parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
def __unicode__(self):
return self.title
- def save(self, force_insert=False, force_update=False, reset_short_html=True, refresh_mp3=True):
+ def save(self, force_insert=False, force_update=False, reset_short_html=True, refresh_mp3=True, **kwargs):
if reset_short_html:
# Reset _short_html during save
update = {}
formats.append(u'<a href="%s">%s</a>' % (reverse('book_text', kwargs={'slug': self.slug}), _('Read online')))
if self.pdf_file:
formats.append(u'<a href="%s">PDF</a>' % self.pdf_file.url)
- if self.epub_file:
- formats.append(u'<a href="%s">EPUB</a>' % self.epub_file.url)
+ if self.root_ancestor.epub_file:
+ formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.epub_file.url)
if self.odt_file:
formats.append(u'<a href="%s">ODT</a>' % self.odt_file.url)
if self.txt_file:
formats.append(u'<a href="%s">MP3</a>' % self.mp3_file.url)
if self.ogg_file:
formats.append(u'<a href="%s">OGG</a>' % self.ogg_file.url)
+ if self.daisy_file:
+ formats.append(u'<a href="%s">DAISY</a>' % self.daisy_file.url)
formats = [mark_safe(format) for format in formats]
return mark_safe(getattr(self, key))
+ @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 get_mp3_info(self):
"""Retrieves artist and director names from audio ID3 tags."""
audio = id3.ID3(self.mp3_file.path)
has_html_file.short_description = 'HTML'
has_html_file.boolean = True
+ 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.
+ """
+
+ from StringIO import StringIO
+ from hashlib import sha1
+ from django.core.files.base import ContentFile
+ from librarian import DocProvider
+
+ class BookImportDocProvider(DocProvider):
+ """ used for joined EPUBs """
+
+ def __init__(self, book):
+ self.book = book
+
+ def by_slug(self, slug):
+ if slug == self.book.slug:
+ return self.book.xml_file
+ else:
+ return Book.objects.get(slug=slug).xml_file
+
+ if self.parent:
+ # don't need an epub
+ return
+
+ 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()
+ FileRecord(slug=self.slug, type='epub', sha1=sha1(epub_file.getvalue()).hexdigest()).save()
+ 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())
+
+
@classmethod
def from_xml_file(cls, xml_file, overwrite=False):
# use librarian to parse meta-data
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)
except Book.DoesNotExist, e:
raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug)
- book_descendants = list(book.children.all())
- while len(book_descendants) > 0:
- child_book = book_descendants.pop(0)
- 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())
-
# Save XML and HTML files
book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
+ # delete old fragments when overwriting
+ book.fragments.all().delete()
+
html_file = NamedTemporaryFile()
if html.transform(book.xml_file.path, html_file, parse_dublincore=False):
book.html_file.save('%s.html' % book.slug, File(html_file), save=False)
+ # get ancestor l-tags for adding to new fragments
+ ancestor_tags = []
+ p = book.parent
+ while p:
+ ancestor_tags.append(p.book_tag())
+ p = p.parent
+
# Extract fragments
closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
for fragment in closed_fragments.values():
- text = fragment.to_string()
- short_text = ''
- if (len(MarkupString(text)) > 240):
- short_text = unicode(MarkupString(text)[:160])
- new_fragment, created = Fragment.objects.get_or_create(anchor=fragment.id, book=book,
- defaults={'text': text, 'short_text': short_text})
-
try:
theme_names = [s.strip() for s in fragment.themes.split(',')]
except AttributeError:
continue
themes = []
for theme_name in theme_names:
+ if not theme_name:
+ continue
tag, created = Tag.objects.get_or_create(slug=slughifi(theme_name), category='theme')
if created:
tag.name = theme_name
tag.sort_key = slughifi(theme_name)
tag.save()
themes.append(tag)
+ if not themes:
+ continue
+
+ text = fragment.to_string()
+ short_text = ''
+ if (len(MarkupString(text)) > 240):
+ short_text = unicode(MarkupString(text)[:160])
+ new_fragment, created = Fragment.objects.get_or_create(anchor=fragment.id, book=book,
+ defaults={'text': text, 'short_text': short_text})
+
new_fragment.save()
- new_fragment.tags = set(book_tags + themes + [book_tag])
+ new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags)
+
+ if not book.parent:
+ book.build_epub(remove_descendants=False)
+
+ book_descendants = list(book.children.all())
+ # add l-tag to descendants and their fragments
+ # delete unnecessary EPUB files
+ while len(book_descendants) > 0:
+ child_book = book_descendants.pop(0)
+ child_book.tags = list(child_book.tags) + [book_tag]
+ if child_book.has_epub_file():
+ child_book.epub_file.delete()
+ 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())
# refresh cache
- book.tag_counter
- book.theme_counter
+ book.reset_tag_counter()
+ book.reset_theme_counter()
book.save()
return book
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)
+ sha1 = models.CharField(_('sha-1 hash'), max_length=40)
+ time = models.DateTimeField(_('time'), auto_now_add=True)
+
+ class Meta:
+ ordering = ('-time','-slug', '-type')
+ verbose_name = _('file record')
+ verbose_name_plural = _('file records')
+
+ def __unicode__(self):
+ return "%s %s.%s" % (self.sha1, self.slug, self.type)
+
+
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)
fnp / wolnelektury.git / blobdiff