minor fixes in search

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

diff --git a/src/search/index.py b/src/search/index.py
index ffad257..f9fb4b2 100644 (file)
--- a/src/search/index.py
+++ b/src/search/index.py
@@ -139,7 +139,6 @@ class Index(SolrIndex):
         else:
             return False
 
-    # WTF
     def index_tags(self, *tags, **kw):
         """
         Re-index global tag list.
@@ -273,7 +272,7 @@ class Index(SolrIndex):
         ]
 
     ignore_content_tags = [
-        'uwaga', 'extra',
+        'uwaga', 'extra', 'nota_red',
         'zastepnik_tekstu', 'sekcja_asterysk', 'separator_linia', 'zastepnik_wersu',
         'didaskalia',
         'naglowek_aktu', 'naglowek_sceny', 'naglowek_czesc',
@@ -366,8 +365,8 @@ class Index(SolrIndex):
         if master is None:
             return []
 
-        def walker(node, ignore_tags=()):
-            if node.tag not in ignore_tags:
+        def walker(node):
+            if node.tag not in self.ignore_content_tags:
                 yield node, None, None
                 if node.text is not None:
                     yield None, node.text, None
@@ -418,17 +417,10 @@ class Index(SolrIndex):
 
             if 'themes' in fields:
                 doc['themes'] = fields['themes']
-            doc['uid'] = "part%s%s%s" % (doc['header_index'],
-                                         doc['header_span'],
-                                         doc.get('fragment_anchor', ''))
+            doc['uid'] = "part%s-%s-%s-%s" % (
+                book.id, doc['header_index'], doc['header_span'], doc.get('fragment_anchor', ''))
             return doc
 
-        def give_me_utf8(s):
-            if isinstance(s, unicode):
-                return s.encode('utf-8')
-            else:
-                return s
-
         fragments = {}
         snippets = Snippets(book.id).open('w')
         try:
@@ -449,7 +441,7 @@ class Index(SolrIndex):
                     content.append(text)
                 handle_text = [all_content]
 
-                for start, text, end in walker(header, ignore_tags=self.ignore_content_tags):
+                for start, text, end in walker(header):
                     # handle footnotes
                     if start is not None and start.tag in self.footnote_tags:
                         footnote = []
@@ -516,8 +508,7 @@ class Index(SolrIndex):
 
 
 class SearchResult(object):
-    def __init__(self, doc, how_found=None, query=None, query_terms=None):
-        #        self.search = search
+    def __init__(self, doc, how_found=None, query_terms=None):
         self.boost = 1.0
         self._hits = []
         self._processed_hits = None  # processed hits
@@ -745,7 +736,7 @@ class Search(SolrIndex):
 
         q = self.index.query(**{field: searched})
         q = self.apply_filters(q, filters).field_limit(score=True, all_fields=True)
-        res = q.execute()
+        res = q.paginate(rows=100).execute()
         return [SearchResult(found, how_found=u'search_phrase') for found in res]
 
     def search_some(self, searched, fields, book=True,
@@ -822,13 +813,18 @@ class Search(SolrIndex):
                 text = snippets.get((int(position),
                                      int(length)))
                 snip = self.index.highlight(text=text, field=field, q=query)
-                snips[idx] = snip
-                if snip:
-                    num -= 1
+                if snip not in snips:
+                    snips[idx] = snip
+                    if snip:
+                        num -= 1
                 idx += 1
 
         except IOError, e:
-            log.error("Cannot open snippet file for book id = %d [rev=%s], %s" % (book_id, revision, e))
+            book = catalogue.models.Book.objects.filter(id=book_id)
+            if not book:
+                log.error("Book does not exist for book id = %d" % book_id)
+            elif not book.get().children.exists():
+                log.error("Cannot open snippet file for book id = %d [rev=%s], %s" % (book_id, revision, e))
             return []
         finally:
             snippets.close()
@@ -912,8 +908,10 @@ class Search(SolrIndex):
         query = query.strip()
         if prefix:
             q |= self.index.Q(title=query + "*")
+            q |= self.index.Q(title_orig=query + "*")
         else:
             q |= self.make_term_query(query, field='title')
+            q |= self.make_term_query(query, field='title_orig')
         qu = self.index.query(q)
         only_books = self.index.Q(is_book=True)
         return self.search_books(qu, [only_books])