Cleaning: timezone issues, deprecated urls.py imports, missing notes.

[wolnelektury.git] / apps / search / index.py

diff --git a/apps/search/index.py b/apps/search/index.py
index ea1a6c5..7fb60b5 100644 (file)
--- a/apps/search/index.py
+++ b/apps/search/index.py
@@ -1,5 +1,7 @@
 # -*- coding: utf-8 -*-
-
+# This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
+#
 from django.conf import settings
 
 import os
@@ -93,7 +95,8 @@ class Snippets(object):
 
     def close(self):
         """Close snippet file"""
-        self.file.close()
+        if self.file:
+            self.file.close()
 
     def remove(self):
         self.revision = None
@@ -133,7 +136,6 @@ class Index(SolrIndex):
                 for res in ids:
                     uids.add(res['uid'])
                 st += rows
-                #        print "Will delete %s" % ','.join([x for x in uids])
         if uids:
             self.index.delete(uids)
             return True
@@ -163,7 +165,7 @@ class Index(SolrIndex):
 
                 q_id_cat = self.index.Q(q_id & q_cat)
                 tag_qs.append(q_id_cat)
-            self.delete_query(tag_qs)
+            self.delete_query(*tag_qs)
         else:  # all
             q = self.index.Q(tag_id__any=True)
             self.delete_query(q)
@@ -464,7 +466,6 @@ class Index(SolrIndex):
                                        text=u''.join(footnote),
                                        is_footnote=True)
                         self.index.add(doc)
-                        #print "@ footnote text: %s" % footnote
                         footnote = []
 
                     # handle fragments and themes.
@@ -497,7 +498,6 @@ class Index(SolrIndex):
                                        fragment_anchor=fid,
                                        text=fix_format(frag['text']),
                                        themes=frag['themes'])
-                        #print '@ FRAG %s' % frag['content']
                         self.index.add(doc)
 
                         # Collect content.
@@ -510,7 +510,6 @@ class Index(SolrIndex):
                         # in the end, add a section text.
                 doc = add_part(snippets, header_index=position,
                                header_type=header.tag, text=fix_format(content))
-                #print '@ CONTENT: %s' % fix_format(content)
 
                 self.index.add(doc)
 
@@ -829,7 +828,7 @@ class Search(SolrIndex):
                 idx += 1
 
         except IOError, e:
-            log.error("Cannot open snippet file for book id = %d [rev=%d], %s" % (book_id, revision, e))
+            log.error("Cannot open snippet file for book id = %d [rev=%s], %s" % (book_id, revision, e))
             return []
         finally:
             snippets.close()
@@ -867,6 +866,8 @@ class Search(SolrIndex):
         res = self.apply_filters(query, filters).execute()
 
         tags = []
+        pd_tags = []
+
         for doc in res:
             is_pdcounter = doc.get('is_pdcounter', False)
             category = doc.get('tag_category')
@@ -878,17 +879,19 @@ class Search(SolrIndex):
                         tag = PDCounterBook.objects.get(id=doc.get('tag_id'))
                         tag.category = 'pd_book'  # make it look more lik a tag.
                     else:
-                        print "Warning. cannot get pdcounter tag_id=%d from db; cat=%s" % (int(doc.get('tag_id')), category)
+                        print ("Warning. cannot get pdcounter tag_id=%d from db; cat=%s" % (int(doc.get('tag_id')), category)).encode('utf-8')
+                    pd_tags.append(tag)
                 else:
                     tag = catalogue.models.Tag.objects.get(id=doc.get("tag_id"))
-                    # don't add the pdcounter tag if same tag already exists
-
-                tags.append(tag)
+                    tags.append(tag)
 
             except catalogue.models.Tag.DoesNotExist: pass
             except PDCounterAuthor.DoesNotExist: pass
             except PDCounterBook.DoesNotExist: pass
 
+        tags_slugs = set(map(lambda t: t.slug, tags))
+        tags = tags + filter(lambda t: not t.slug in tags_slugs, pd_tags)
+
         log.debug('search_tags: %s' % tags)
 
         return tags
@@ -914,10 +917,15 @@ class Search(SolrIndex):
         Searches for Book objects using query
         """
         bks = []
+        bks_found = set()
+        query = query.query(is_book=True)
         res = self.apply_filters(query, filters).field_limit(['book_id'])
         for r in res:
             try:
-                bks.append(catalogue.models.Book.objects.get(id=r['book_id']))
+                bid = r['book_id']
+                if not bid in bks_found:
+                    bks.append(catalogue.models.Book.objects.get(id=bid))
+                    bks_found.add(bid)
             except catalogue.models.Book.DoesNotExist: pass
         return bks