Fixing flickr parsing
[redakcja.git] / apps / catalogue / models / book.py
index 098781d..b1b2730 100755 (executable)
@@ -12,11 +12,12 @@ from slughifi import slughifi
 
 
 import apiclient
 
 
 import apiclient
-from catalogue.helpers import cached_in_field
-from catalogue.models import BookPublishRecord, ChunkPublishRecord
+from catalogue.helpers import cached_in_field, GalleryMerger
+from catalogue.models import BookPublishRecord, ChunkPublishRecord, Project
 from catalogue.signals import post_publish
 from catalogue.tasks import refresh_instance, book_content_updated
 from catalogue.xml_tools import compile_text, split_xml
 from catalogue.signals import post_publish
 from catalogue.tasks import refresh_instance, book_content_updated
 from catalogue.xml_tools import compile_text, split_xml
+from cover.models import Image
 import os
 import shutil
 import re
 import os
 import shutil
 import re
@@ -28,6 +29,7 @@ class Book(models.Model):
     slug = models.SlugField(_('slug'), max_length=128, unique=True, db_index=True)
     public = models.BooleanField(_('public'), default=True, db_index=True)
     gallery = models.CharField(_('scan gallery name'), max_length=255, blank=True)
     slug = models.SlugField(_('slug'), max_length=128, unique=True, db_index=True)
     public = models.BooleanField(_('public'), default=True, db_index=True)
     gallery = models.CharField(_('scan gallery name'), max_length=255, blank=True)
+    project = models.ForeignKey(Project, null=True)
 
     #wl_slug = models.CharField(_('title'), max_length=255, null=True, db_index=True, editable=False)
     parent = models.ForeignKey('self', null=True, blank=True, verbose_name=_('parent'), related_name="children", editable=False)
 
     #wl_slug = models.CharField(_('title'), max_length=255, null=True, db_index=True, editable=False)
     parent = models.ForeignKey('self', null=True, blank=True, verbose_name=_('parent'), related_name="children", editable=False)
@@ -39,6 +41,8 @@ class Book(models.Model):
     _new_publishable = models.NullBooleanField(editable=False)
     _published = models.NullBooleanField(editable=False)
     _on_track = models.IntegerField(null=True, blank=True, db_index=True, editable=False)
     _new_publishable = models.NullBooleanField(editable=False)
     _published = models.NullBooleanField(editable=False)
     _on_track = models.IntegerField(null=True, blank=True, db_index=True, editable=False)
+    dc_cover_image = models.ForeignKey(Image, blank=True, null=True,
+        db_index=True, on_delete=models.SET_NULL, editable=False)
     dc_slug = models.CharField(max_length=128, null=True, blank=True,
             editable=False, db_index=True)
 
     dc_slug = models.CharField(max_length=128, null=True, blank=True,
             editable=False, db_index=True)
 
@@ -196,87 +200,17 @@ class Book(models.Model):
             number += 1
         assert not other.chunk_set.exists()
 
             number += 1
         assert not other.chunk_set.exists()
 
-        self.append_gallery(other, len_other)
-        
-        other.delete()
-
-
-    def append_gallery(self, other, len_other):
-        if self.gallery is None:
-            self.gallery = other.gallery
-            return
-        if other.gallery is None:
-            return
-        
-        def get_prefix(name):
-            m = re.match(r"^([0-9])-", name)
-            if m:
-                return int(m.groups()[0])
-            return None
-        
-        def set_prefix(name, prefix, always=False):
-            m = not always and re.match(r"^([0-9])-", name)
-            return "%1d-%s" % (prefix, m and name[2:] or name)
-
-        files = os.listdir(os.path.join(settings.MEDIA_ROOT,
-                                        settings.IMAGE_DIR, self.gallery))
-        files_other = os.listdir(os.path.join(settings.MEDIA_ROOT,
-                                              settings.IMAGE_DIR, other.gallery))
-
-        prefixes = {}
-        renamed_files = {}
-        renamed_files_other = {}
-        last_pfx = -1
-
-        # check if all elements of my files have a prefix
-        files_prefixed = True
-        for f in files:
-            p = get_prefix(f)
-            if p:
-                if p > last_pfx: last_pfx = p
-            else:
-                files_prefixed = False
-                break
-
-        # if not, add a 0 prefix to them
-        if not files_prefixed:
-            prefixes[0] = 0
-            for f in files:
-                renamed_files[f] = set_prefix(f, 0, True)
-
-        # two cases here - either all are prefixed or not.
-        files_other_prefixed = True
-        for f in files_other:
-            pfx = get_prefix(f)
-            if pfx is not None:
-                if not pfx in prefixes:
-                    last_pfx += 1
-                    prefixes[pfx] = last_pfx
-                renamed_files_other[f] = set_prefix(f, prefixes[pfx])
-            else:
-                # ops, not all files here were prefixed.
-                files_other_prefixed = False
-                break
-
-        # just set a 1- prefix to all of them
-        if not files_other_prefixed:
-            for f in files_other:
-                renamed_files_other[f] = set_prefix(f, 1, True)
-
-        # finally, move / rename files.
-        for frm, to in renamed_files.items():
-            shutil.move(os.path.join(settings.MEDIA_ROOT, settings.IMAGE_DIR, self.gallery, frm),
-                        os.path.join(settings.MEDIA_ROOT, settings.IMAGE_DIR, self.gallery, to))
-        for frm, to in renamed_files_other.items():
-            shutil.move(os.path.join(settings.MEDIA_ROOT, settings.IMAGE_DIR, other.gallery, frm),
-                        os.path.join(settings.MEDIA_ROOT, settings.IMAGE_DIR, self.gallery, to))            
+        gm = GalleryMerger(self.gallery, other.gallery)
+        self.gallery = gm.merge()
 
         # and move the gallery starts
 
         # and move the gallery starts
-        num_files = len(files)
-        for chunk in self[len(self) - len_other:]:
-            chunk.gallery_start += num_files
-            chunk.save()
-            
+        if gm.was_merged:
+                for chunk in self[len(self) - len_other:]:
+                        old_start = chunk.gallery_start or 1
+                        chunk.gallery_start = old_start + gm.dest_size - gm.num_deleted
+                        chunk.save()
+
+        other.delete()
 
 
     @transaction.commit_on_success
 
 
     @transaction.commit_on_success
@@ -323,13 +257,11 @@ class Book(models.Model):
             changes = self.get_current_changes(publishable=True)
         except self.NoTextError:
             raise AssertionError(_('Not all chunks have publishable revisions.'))
             changes = self.get_current_changes(publishable=True)
         except self.NoTextError:
             raise AssertionError(_('Not all chunks have publishable revisions.'))
-        book_xml = self.materialize(changes=changes)
 
 
-        from librarian.dcparser import BookInfo
         from librarian import NoDublinCore, ParseError, ValidationError
 
         try:
         from librarian import NoDublinCore, ParseError, ValidationError
 
         try:
-            bi = BookInfo.from_string(book_xml.encode('utf-8'), strict=True)
+            bi = self.wldocument(changes=changes, strict=True).book_info
         except ParseError, e:
             raise AssertionError(_('Invalid XML') + ': ' + unicode(e))
         except NoDublinCore:
         except ParseError, e:
             raise AssertionError(_('Invalid XML') + ': ' + unicode(e))
         except NoDublinCore:
@@ -340,6 +272,14 @@ class Book(models.Model):
         valid_about = self.correct_about()
         assert bi.about == valid_about, _("rdf:about is not") + " " + valid_about
 
         valid_about = self.correct_about()
         assert bi.about == valid_about, _("rdf:about is not") + " " + valid_about
 
+    def publishable_error(self):
+        try:
+            return self.assert_publishable()
+        except AssertionError, e:
+            return e
+        else:
+            return None
+
     def hidden(self):
         return self.slug.startswith('.')
 
     def hidden(self):
         return self.slug.startswith('.')
 
@@ -400,11 +340,20 @@ class Book(models.Model):
     def refresh_dc_cache(self):
         update = {
             'dc_slug': None,
     def refresh_dc_cache(self):
         update = {
             'dc_slug': None,
+            'dc_cover_image': None,
         }
 
         info = self.book_info()
         if info is not None:
             update['dc_slug'] = info.url.slug
         }
 
         info = self.book_info()
         if info is not None:
             update['dc_slug'] = info.url.slug
+            if info.cover_source:
+                try:
+                    image = Image.objects.get(pk=int(info.cover_source.rstrip('/').rsplit('/', 1)[-1]))
+                except:
+                    pass
+                else:
+                    if info.cover_source == image.get_full_url():
+                        update['dc_cover_image'] = image
         Book.objects.filter(pk=self.pk).update(**update)
 
     def touch(self):
         Book.objects.filter(pk=self.pk).update(**update)
 
     def touch(self):
@@ -456,14 +405,16 @@ class Book(models.Model):
             changes = self.get_current_changes(publishable)
         return compile_text(change.materialize() for change in changes)
 
             changes = self.get_current_changes(publishable)
         return compile_text(change.materialize() for change in changes)
 
-    def wldocument(self, publishable=True, changes=None, parse_dublincore=True):
+    def wldocument(self, publishable=True, changes=None, 
+            parse_dublincore=True, strict=False):
         from catalogue.ebook_utils import RedakcjaDocProvider
         from librarian.parser import WLDocument
 
         return WLDocument.from_string(
                 self.materialize(publishable=publishable, changes=changes),
                 provider=RedakcjaDocProvider(publishable=publishable),
         from catalogue.ebook_utils import RedakcjaDocProvider
         from librarian.parser import WLDocument
 
         return WLDocument.from_string(
                 self.materialize(publishable=publishable, changes=changes),
                 provider=RedakcjaDocProvider(publishable=publishable),
-                parse_dublincore=parse_dublincore)
+                parse_dublincore=parse_dublincore,
+                strict=strict)
 
     def publish(self, user):
         """
 
     def publish(self, user):
         """