opds: added search, moved to separate app
[wolnelektury.git] / apps / catalogue / models.py
index a7e04f1..d3b24d3 100644 (file)
@@ -17,7 +17,7 @@ from newtagging.models import TagBase, tags_updated
 from newtagging import managers
 from catalogue.fields import JSONField
 
 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
 
 
 from mutagen import id3
 
 
@@ -207,7 +207,7 @@ class Book(models.Model):
     def __unicode__(self):
         return self.title
 
     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 = {}
         if reset_short_html:
             # Reset _short_html during save
             update = {}
@@ -263,8 +263,8 @@ class Book(models.Model):
                 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)
                 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:
             if self.odt_file:
                 formats.append(u'<a href="%s">ODT</a>' % self.odt_file.url)
             if self.txt_file:
@@ -282,6 +282,18 @@ class Book(models.Model):
             return mark_safe(getattr(self, key))
 
 
             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)
     def get_mp3_info(self):
         """Retrieves artist and director names from audio ID3 tags."""
         audio = id3.ID3(self.mp3_file.path)
@@ -314,6 +326,52 @@ class Book(models.Model):
     has_html_file.short_description = 'HTML'
     has_html_file.boolean = True
 
     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
     @classmethod
     def from_xml_file(cls, xml_file, overwrite=False):
         # use librarian to parse meta-data
@@ -332,6 +390,7 @@ class Book(models.Model):
         from tempfile import NamedTemporaryFile
         from slughifi import slughifi
         from markupstring import MarkupString
         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)
 
         # Read book metadata
         book_base, book_slug = book_info.url.rsplit('/', 1)
@@ -384,15 +443,6 @@ class Book(models.Model):
                 except Book.DoesNotExist, e:
                     raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug)
 
                 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)
 
         # Save XML and HTML files
         book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)
 
@@ -403,28 +453,48 @@ class Book(models.Model):
             # Extract fragments
             closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
             for fragment in closed_fragments.values():
             # 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:
                 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)
                     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.save()
                 new_fragment.tags = set(book_tags + themes + [book_tag])
 
+        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
         # refresh cache
         book.tag_counter
         book.theme_counter
@@ -538,6 +608,21 @@ class BookStub(models.Model):
         return self.title
 
 
         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)
 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)