Test błędnego XMLa.
[wolnelektury.git] / apps / catalogue / models.py
index feccaed..58025fa 100644 (file)
@@ -6,11 +6,14 @@ from django.contrib.auth.models import User
 from django.core.files import File
 from django.template.loader import render_to_string
 from django.utils.safestring import mark_safe
+from django.core.urlresolvers import reverse
 
 from newtagging.models import TagBase
 from newtagging import managers
+from catalogue.fields import JSONField
 
 from librarian import html, dcparser
+from mutagen import id3
 
 
 TAG_CATEGORIES = (
@@ -20,6 +23,7 @@ TAG_CATEGORIES = (
     ('genre', _('genre')),
     ('theme', _('theme')),
     ('set', _('set')),
+    ('book', _('book')),
 )
 
 
@@ -33,14 +37,18 @@ class TagSubcategoryManager(models.Manager):
 
 
 class Tag(TagBase):
-    name = models.CharField(_('name'), max_length=50, unique=True, db_index=True)
+    name = models.CharField(_('name'), max_length=50, db_index=True)
     slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
     sort_key = models.SlugField(_('sort key'), max_length=120, db_index=True)
     category = models.CharField(_('category'), max_length=50, blank=False, null=False, 
         db_index=True, choices=TAG_CATEGORIES)
-    description = models.TextField(blank=True)
+    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'), default=0, blank=False, null=False)
+    gazeta_link = models.CharField(blank=True,  max_length=240)
+    wiki_link = models.CharField(blank=True,  max_length=240)
     
     def has_description(self):
         return len(self.description) > 0
@@ -68,45 +76,86 @@ class Tag(TagBase):
             return TagBase.get_tag_list(tags)
 
 
+def book_upload_path(ext):
+    def get_dynamic_path(book, filename):
+        return 'lektura/%s.%s' % (book.slug, ext)
+    return get_dynamic_path
+
+
 class Book(models.Model):
     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)
     _short_html = models.TextField(_('short HTML'), editable=False)
+    parent_number = models.IntegerField(_('parent number'), default=0)
+    extra_info = JSONField(_('extra information'))
+    gazeta_link = models.CharField(blank=True,  max_length=240)
+    wiki_link = models.CharField(blank=True,  max_length=240)
+
     
     # Formats
-    xml_file = models.FileField(_('XML file'), upload_to='books/xml', blank=True)
-    pdf_file = models.FileField(_('PDF file'), upload_to='books/pdf', blank=True)
-    odt_file = models.FileField(_('ODT file'), upload_to='books/odt', blank=True)
-    html_file = models.FileField(_('HTML file'), upload_to='books/html', blank=True)
+    xml_file = models.FileField(_('XML file'), upload_to=book_upload_path('xml'), blank=True)
+    html_file = models.FileField(_('HTML file'), upload_to=book_upload_path('html'), blank=True)
+    pdf_file = models.FileField(_('PDF file'), upload_to=book_upload_path('pdf'), blank=True)
+    odt_file = models.FileField(_('ODT file'), upload_to=book_upload_path('odt'), blank=True)
+    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)
     
     parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
     
     objects = models.Manager()
     tagged = managers.ModelTaggedItemManager(Tag)
     tags = managers.TagDescriptor(Tag)
+
+    
+    @property
+    def name(self):
+        return self.title
     
     def short_html(self):
         if len(self._short_html):
             return mark_safe(self._short_html)
         else:
-            tags = self.tags.filter(~Q(category__in=('set', 'theme')))
-            tags = [u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) for tag in tags]
+            tags = self.tags.filter(~Q(category__in=('set', 'theme', 'book')))
+            tags = [mark_safe(u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name)) for tag in tags]
 
             formats = []
             if self.html_file:
-                formats.append(u'<a href="%s">Czytaj online</a>' % self.html_file.url)
+                formats.append(u'<a href="%s">Czytaj online</a>' % reverse('book_text', kwargs={'slug': self.slug}))
             if self.pdf_file:
-                formats.append(u'<a href="%s">Plik PDF</a>' % self.pdf_file.url)
+                formats.append(u'<a href="%s">PDF</a>' % self.pdf_file.url)
             if self.odt_file:
-                formats.append(u'<a href="%s">Plik ODT</a>' % self.odt_file.url)
+                formats.append(u'<a href="%s">ODT</a>' % self.odt_file.url)
+            if self.txt_file:
+                formats.append(u'<a href="%s">TXT</a>' % self.txt_file.url)
+            if self.mp3_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)
+            
+            formats = [mark_safe(format) for format in formats]
             
-            self._short_html = render_to_string('catalogue/book_short.html',
-                {'book': self, 'tags': tags, 'formats': formats})
+            self._short_html = unicode(render_to_string('catalogue/book_short.html',
+                {'book': self, 'tags': tags, 'formats': formats}))
             self.save()
             return mark_safe(self._short_html)
     
+    def save(self, force_insert=False, force_update=False):
+        if self.mp3_file:
+            extra_info = self.get_extra_info_value()
+            extra_info.update(self.get_mp3_info())
+            self.set_extra_info_value(extra_info)
+        return super(Book, self).save(force_insert, force_update)
+    
+    def get_mp3_info(self):
+        """Retrieves artist and director names from audio ID3 tags."""
+        audio = id3.ID3(self.mp3_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'))
+        return {'artist_name': artist_name, 'director_name': director_name}
+        
     def has_description(self):
         return len(self.description) > 0
     has_description.short_description = _('description')
@@ -127,15 +176,31 @@ class Book(models.Model):
     has_html_file.short_description = 'HTML'
     has_html_file.boolean = True
 
+    class AlreadyExists(Exception):
+        pass
+    
     @staticmethod
-    def from_xml_file(xml_file):
+    def from_xml_file(xml_file, overwrite=False):
         from tempfile import NamedTemporaryFile
         from slughifi import slughifi
         from markupstring import MarkupString
         
         # Read book metadata
         book_info = dcparser.parse(xml_file)
-        book = Book(title=book_info.title, slug=slughifi(book_info.title))
+        book_base, book_slug = book_info.url.rsplit('/', 1)
+        book, created = Book.objects.get_or_create(slug=book_slug)
+        
+        if created:
+            book_shelves = []
+        else:
+            if not overwrite:
+                raise Book.AlreadyExists('Book %s already exists' % book_slug)
+            # Save shelves for this book
+            book_shelves = list(book.tags.filter(category='set'))
+        
+        book.title = book_info.title
+        book.set_extra_info_value(book_info.to_dict())
+        book._short_html = ''
         book.save()
         
         book_tags = []
@@ -145,50 +210,81 @@ class Book(models.Model):
             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(name=tag_name,
-                slug=slughifi(tag_name), sort_key=slughifi(tag_sort_key), category=category)
-            tag.save()
+            tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name))
+            if created:
+                tag.name = tag_name
+                tag.sort_key = slughifi(tag_sort_key)
+                tag.category = category
+                tag.save()
             book_tags.append(tag)
+            
+        book_tag, created = Tag.objects.get_or_create(slug=('l-' + book.slug)[:120])
+        if created:
+            book_tag.name = book.title[:50]
+            book_tag.sort_key = ('l-' + book.slug)[:120]
+            book_tag.category = 'book'
+            book_tag.save()
+        book_tags.append(book_tag)
+        
         book.tags = book_tags
         
         if hasattr(book_info, 'parts'):
-            for part_url in book_info.parts:
+            for n, part_url in enumerate(book_info.parts):
                 base, slug = part_url.rsplit('/', 1)
                 child_book = Book.objects.get(slug=slug)
                 child_book.parent = book
+                child_book.parent_number = n
                 child_book.save()
-        
+
+        book_descendants = list(book.children.all())
+        while len(book_descendants) > 0:
+            child_book = book_descendants.pop(0)
+            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, File(file(xml_file)), save=False)
+        if not isinstance(xml_file, File):
+            xml_file = File(file(xml_file))
+        book.xml_file.save('%s.xml' % book.slug, xml_file, save=False)
         
         html_file = NamedTemporaryFile()
-        html.transform(book.xml_file.path, html_file)
-        book.html_file.save('%s.html' % book.slug, File(html_file), save=False)
-        
-        # Extract fragments
-        closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
-        book_themes = []
-        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 = Fragment(text=text, short_text=short_text, anchor=fragment.id, book=book)
+        if html.transform(book.xml_file.path, html_file):
+            book.html_file.save('%s.html' % book.slug, File(html_file), save=False)
+            
+            # Extract fragments
+            closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
+            book_themes = []
+            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})
                 
-            theme_names = [s.strip() for s in fragment.themes.split(',')]
-            themes = []
-            for theme_name in theme_names:
-                tag, created = Tag.objects.get_or_create(name=theme_name,
-                    slug=slughifi(theme_name), sort_key=slughifi(theme_name), category='theme')
-                tag.save()
-                themes.append(tag)
-            new_fragment.save()
-            new_fragment.tags = list(book.tags) + themes
-            book_themes += themes
+                try:
+                    theme_names = [s.strip() for s in fragment.themes.split(',')]
+                except AttributeError:
+                    continue
+                themes = []
+                for theme_name in theme_names:
+                    tag, created = Tag.objects.get_or_create(slug=slughifi(theme_name))
+                    if created:
+                        tag.name = theme_name
+                        tag.sort_key = slughifi(theme_name)
+                        tag.category = 'theme'
+                        tag.save()
+                    themes.append(tag)
+                new_fragment.save()
+                new_fragment.tags = set(list(book.tags) + themes + [book_tag])
+                book_themes += themes
+            
+            book_themes = set(book_themes)
+            book.tags = list(book.tags) + list(book_themes) + book_shelves
         
-        book_themes = set(book_themes)
-        book.tags = list(book.tags) + list(book_themes)
-        return book.save()
+        book.save()
+        return book
     
     @permalink
     def get_absolute_url(self):
@@ -207,7 +303,7 @@ class Fragment(models.Model):
     text = models.TextField()
     short_text = models.TextField(editable=False)
     _short_html = models.TextField(editable=False)
-    anchor = models.IntegerField()
+    anchor = models.CharField(max_length=120)
     book = models.ForeignKey(Book, related_name='fragments')
 
     objects = models.Manager()
@@ -218,14 +314,17 @@ class Fragment(models.Model):
         if len(self._short_html):
             return mark_safe(self._short_html)
         else:
-            book_authors = [u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name
+            book_authors = [mark_safe(u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name)
                 for tag in self.book.tags if tag.category == 'author']
             
-            self._short_html = render_to_string('catalogue/fragment_short.html',
-                {'fragment': self, 'book': self.book, 'book_authors': book_authors})
+            self._short_html = unicode(render_to_string('catalogue/fragment_short.html',
+                {'fragment': self, 'book': self.book, 'book_authors': book_authors}))
             self.save()
             return mark_safe(self._short_html)
-        
+    
+    def get_absolute_url(self):
+        return '%s#m%s' % (reverse('book_text', kwargs={'slug': self.book.slug}), self.anchor)
+    
     class Meta:
         ordering = ('book', 'anchor',)
         verbose_name = _('fragment')