#968: don't import invalid slugs

[wolnelektury.git] / apps / catalogue / models.py

diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py
index 1d0b83b..cb1eff8 100644 (file)
--- a/apps/catalogue/models.py
+++ b/apps/catalogue/models.py
@@ -12,6 +12,7 @@ from django.template.defaultfilters import slugify
 from django.utils.safestring import mark_safe
 from django.utils.translation import get_language
 from django.core.urlresolvers import reverse
 from django.utils.safestring import mark_safe
 from django.utils.translation import get_language
 from django.core.urlresolvers import reverse

+from django.db.models.signals import post_save, m2m_changed, pre_delete

 
 from django.conf import settings
 
 
 from django.conf import settings
 


@@ -59,7 +60,7 @@ class Tag(TagBase):
     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)
     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'), blank=False, null=True)
+    book_count = models.IntegerField(_('book count'), blank=True, null=True)

     gazeta_link = models.CharField(blank=True, max_length=240)
     wiki_link = models.CharField(blank=True, max_length=240)
 
     gazeta_link = models.CharField(blank=True, max_length=240)
     wiki_link = models.CharField(blank=True, max_length=240)
 


@@ -171,8 +172,9 @@ def book_upload_path(ext=None):
 class BookMedia(models.Model):
     type        = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
     name        = models.CharField(_('name'), max_length="100", blank=True)
 class BookMedia(models.Model):
     type        = models.CharField(_('type'), choices=MEDIA_FORMATS, max_length="100")
     name        = models.CharField(_('name'), max_length="100", blank=True)

-    file        = models.FileField(_('file'), upload_to=book_upload_path(), blank=True)    
+    file        = models.FileField(_('file'), upload_to=book_upload_path(), blank=True)

     uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)
     uploaded_at = models.DateTimeField(_('creation date'), auto_now_add=True, editable=False)

+    extra_info  = JSONField(_('extra information'), default='{}')

 
     def __unicode__(self):
         return "%s (%s)" % (self.name, self.file.name.split("/")[-1])
 
     def __unicode__(self):
         return "%s (%s)" % (self.name, self.file.name.split("/")[-1])


@@ -182,6 +184,27 @@ class BookMedia(models.Model):
         verbose_name        = _('book media')
         verbose_name_plural = _('book media')
 
         verbose_name        = _('book media')
         verbose_name_plural = _('book media')
 

+    def save(self, force_insert=False, force_update=False):
+        media = super(BookMedia, self).save(force_insert, force_update)
+        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)
+        return media
+
+    def get_mp3_info(self):
+        """Retrieves artist and director names from audio ID3 tags."""
+        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}
+
+

 class Book(models.Model):
     title         = models.CharField(_('title'), max_length=120)
     slug          = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
 class Book(models.Model):
     title         = models.CharField(_('title'), max_length=120)
     slug          = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)


@@ -220,7 +243,7 @@ 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, **kwargs):
+    def save(self, force_insert=False, force_update=False, reset_short_html=True, **kwargs):

         if reset_short_html:
             # Reset _short_html during save
             update = {}
         if reset_short_html:
             # Reset _short_html during save
             update = {}


@@ -230,16 +253,7 @@ class Book(models.Model):
             # Fragment.short_html relies on book's tags, so reset it here too
             self.fragments.all().update(**update)
 
             # Fragment.short_html relies on book's tags, so reset it here too
             self.fragments.all().update(**update)
 

-        book = super(Book, self).save(force_insert, force_update)
-
-        if refresh_mp3 and self.has_media('mp3'):
-            file = self.get_media('mp3')[0]
-            #print file, file.path
-            extra_info = self.get_extra_info_value()
-            extra_info.update(self.get_mp3_info())
-            self.set_extra_info_value(extra_info)
-            book = super(Book, self).save(force_insert, force_update)
-        return book
+        return super(Book, self).save(force_insert, force_update)

 
     @permalink
     def get_absolute_url(self):
 
     @permalink
     def get_absolute_url(self):


@@ -300,7 +314,7 @@ class Book(models.Model):
             elif type == "html":
                 return self.html_file
             elif type == "epub":
             elif type == "html":
                 return self.html_file
             elif type == "epub":

-                return self.html_file                
+                return self.epub_file

             elif type == "txt":
                 return self.txt_file
             elif type == "pdf":
             elif type == "txt":
                 return self.txt_file
             elif type == "pdf":


@@ -337,13 +351,11 @@ class Book(models.Model):
                 formats.append(u'<a href="%s">PDF</a>' % self.get_media('pdf').url)
             if self.root_ancestor.has_media("epub"):
                 formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.get_media('epub').url)
                 formats.append(u'<a href="%s">PDF</a>' % self.get_media('pdf').url)
             if self.root_ancestor.has_media("epub"):
                 formats.append(u'<a href="%s">EPUB</a>' % self.root_ancestor.get_media('epub').url)

-            if self.has_media("odt"):
-                formats.append(u'<a href="%s">ODT</a>' % self.get_media('odt').url)

             if self.has_media("txt"):
                 formats.append(u'<a href="%s">TXT</a>' % self.get_media('txt').url)
             # other files
             for m in self.medias.order_by('type'):
             if self.has_media("txt"):
                 formats.append(u'<a href="%s">TXT</a>' % self.get_media('txt').url)
             # other files
             for m in self.medias.order_by('type'):

-                formats.append(u'<a href="%s">%s</a>' % m.type, m.file.url)
+                formats.append(u'<a href="%s">%s</a>' % (m.file.url, m.type.upper()))

 
             formats = [mark_safe(format) for format in formats]
 
 
             formats = [mark_safe(format) for format in formats]
 


@@ -365,13 +377,6 @@ class Book(models.Model):
         return self._root_ancestor
 
 
         return self._root_ancestor
 
 

-    def get_mp3_info(self):
-        """Retrieves artist and director names from audio ID3 tags."""
-        audio = id3.ID3(self.get_media('mp3')[0].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')
     def has_description(self):
         return len(self.description) > 0
     has_description.short_description = _('description')


@@ -388,6 +393,11 @@ class Book(models.Model):
     has_epub_file.short_description = 'EPUB'
     has_epub_file.boolean = True
 
     has_epub_file.short_description = 'EPUB'
     has_epub_file.boolean = True
 

+    def has_txt_file(self):
+        return bool(self.txt_file)
+    has_txt_file.short_description = 'HTML'
+    has_txt_file.boolean = True
+

     def has_html_file(self):
         return bool(self.html_file)
     has_html_file.short_description = 'HTML'
     def has_html_file(self):
         return bool(self.html_file)
     has_html_file.short_description = 'HTML'


@@ -442,9 +452,8 @@ class Book(models.Model):
 
         epub_file = StringIO()
         try:
 
         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(refresh_mp3=False)
+            epub.transform(BookImportDocProvider(self), self.slug, output_file=epub_file)
+            self.epub_file.save('%s.epub' % self.slug, ContentFile(epub_file.getvalue()))

             FileRecord(slug=self.slug, type='epub', sha1=sha1(epub_file.getvalue()).hexdigest()).save()
         except NoDublinCore:
             pass
             FileRecord(slug=self.slug, type='epub', sha1=sha1(epub_file.getvalue()).hexdigest()).save()
         except NoDublinCore:
             pass


@@ -455,12 +464,22 @@ class Book(models.Model):
             if remove_descendants and child_book.has_epub_file():
                 child_book.epub_file.delete()
             # save anyway, to refresh short_html
             if remove_descendants and child_book.has_epub_file():
                 child_book.epub_file.delete()
             # save anyway, to refresh short_html

-            child_book.save(refresh_mp3=False)
+            child_book.save()

             book_descendants += list(child_book.children.all())
 
             book_descendants += list(child_book.children.all())
 

+    def build_txt(self):
+        from StringIO import StringIO
+        from django.core.files.base import ContentFile
+        from librarian import text
+
+        out = StringIO()
+        text.transform(open(self.xml_file.path), out)
+        self.txt_file.save('%s.txt' % self.slug, ContentFile(out.getvalue()))
+        self.save()
+

 
     @classmethod
 
     @classmethod

-    def from_xml_file(cls, xml_file, overwrite=False):
+    def from_xml_file(cls, xml_file, **kwargs):

         # use librarian to parse meta-data
         book_info = dcparser.parse(xml_file)
 
         # use librarian to parse meta-data
         book_info = dcparser.parse(xml_file)
 


@@ -468,12 +487,13 @@ class Book(models.Model):
             xml_file = File(open(xml_file))
 
         try:
             xml_file = File(open(xml_file))
 
         try:

-            return cls.from_text_and_meta(xml_file, book_info, overwrite)
+            return cls.from_text_and_meta(xml_file, book_info, **kwargs)

         finally:
             xml_file.close()
 
     @classmethod
         finally:
             xml_file.close()
 
     @classmethod

-    def from_text_and_meta(cls, raw_file, book_info, overwrite=False):
+    def from_text_and_meta(cls, raw_file, book_info, overwrite=False, build_epub=True, build_txt=True):
+        import re

         from tempfile import NamedTemporaryFile
         from slughifi import slughifi
         from markupstring import MarkupString
         from tempfile import NamedTemporaryFile
         from slughifi import slughifi
         from markupstring import MarkupString


@@ -492,6 +512,8 @@ class Book(models.Model):
 
         # Read book metadata
         book_base, book_slug = book_info.url.rsplit('/', 1)
 
         # Read book metadata
         book_base, book_slug = book_info.url.rsplit('/', 1)

+        if re.search(r'[^a-zA-Z0-9-]', book_slug):
+            raise ValueError('Invalid characters in slug')

         book, created = Book.objects.get_or_create(slug=book_slug)
 
         if created:
         book, created = Book.objects.get_or_create(slug=book_slug)
 
         if created:


@@ -582,8 +604,11 @@ class Book(models.Model):
                 new_fragment.save()
                 new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags)
 
                 new_fragment.save()
                 new_fragment.tags = set(book_tags + themes + [book_tag] + ancestor_tags)
 

-        if not settings.NO_BUILD_EPUB:
-            book.root_ancestor().build_epub()
+        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()

 
         book_descendants = list(book.children.all())
         # add l-tag to descendants and their fragments
 
         book_descendants = list(book.children.all())
         # add l-tag to descendants and their fragments


@@ -612,12 +637,12 @@ class Book(models.Model):
         for tag in self.tags.exclude(category__in=('book', 'theme', 'set')).order_by():
             tags[tag.pk] = 1
         self.set__tag_counter_value(tags)
         for tag in self.tags.exclude(category__in=('book', 'theme', 'set')).order_by():
             tags[tag.pk] = 1
         self.set__tag_counter_value(tags)

-        self.save(reset_short_html=False, refresh_mp3=False)
+        self.save(reset_short_html=False)

         return tags
 
     def reset_tag_counter(self):
         self._tag_counter = None
         return tags
 
     def reset_tag_counter(self):
         self._tag_counter = None

-        self.save(reset_short_html=False, refresh_mp3=False)
+        self.save(reset_short_html=False)

         if self.parent:
             self.parent.reset_tag_counter()
 
         if self.parent:
             self.parent.reset_tag_counter()
 


@@ -633,12 +658,12 @@ class Book(models.Model):
             for tag in fragment.tags.filter(category='theme').order_by():
                 tags[tag.pk] = tags.get(tag.pk, 0) + 1
         self.set__theme_counter_value(tags)
             for tag in fragment.tags.filter(category='theme').order_by():
                 tags[tag.pk] = tags.get(tag.pk, 0) + 1
         self.set__theme_counter_value(tags)

-        self.save(reset_short_html=False, refresh_mp3=False)
+        self.save(reset_short_html=False)

         return tags
 
     def reset_theme_counter(self):
         self._theme_counter = None
         return tags
 
     def reset_theme_counter(self):
         self._theme_counter = None

-        self.save(reset_short_html=False, refresh_mp3=False)
+        self.save(reset_short_html=False)

         if self.parent:
             self.parent.reset_theme_counter()
 
         if self.parent:
             self.parent.reset_theme_counter()
 


@@ -711,6 +736,12 @@ class FileRecord(models.Model):
     def __unicode__(self):
         return "%s %s.%s" % (self.sha1,  self.slug, self.type)
 
     def __unicode__(self):
         return "%s %s.%s" % (self.sha1,  self.slug, self.type)
 

+###########
+#
+# SIGNALS
+#
+###########
+

 
 def _tags_updated_handler(sender, affected_tags, **kwargs):
     # reset tag global counter
 
 def _tags_updated_handler(sender, affected_tags, **kwargs):
     # reset tag global counter


@@ -728,3 +759,24 @@ def _tags_updated_handler(sender, affected_tags, **kwargs):
         sender.book.reset_theme_counter()
 tags_updated.connect(_tags_updated_handler)
 
         sender.book.reset_theme_counter()
 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 """
+    if sender == BookMedia:
+        instance.book_set.clear()
+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()
+post_save.connect(_post_save_handler)
+