profiling

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

diff --git a/apps/search/index.py b/apps/search/index.py
index da54fe7..26da062 100644 (file)
--- a/apps/search/index.py
+++ b/apps/search/index.py
@@ -18,6 +18,7 @@ import sunburnt
 import custom
 import operator
 
 import custom
 import operator
 

+log = logging.getLogger('search')

 
 class SolrIndex(object):
     def __init__(self, mode=None):
 
 class SolrIndex(object):
     def __init__(self, mode=None):


@@ -145,6 +146,7 @@ class Index(SolrIndex):
         Removes all tags from index, then index them again.
         Indexed fields include: id, name (with and without polish stems), category
         """
         Removes all tags from index, then index them again.
         Indexed fields include: id, name (with and without polish stems), category
         """

+        log.debug("Indexing tags")

         remove_only = kw.get('remove_only', False)
         # first, remove tags from index.
         if tags:
         remove_only = kw.get('remove_only', False)
         # first, remove tags from index.
         if tags:


@@ -201,7 +203,6 @@ class Index(SolrIndex):
                         "is_pdcounter": False,
                         "uid": "tag%d" % tag.id
                         }
                         "is_pdcounter": False,
                         "uid": "tag%d" % tag.id
                         }

-                print "ADD 1 %s" % doc

                 self.index.add(doc)
 
     def create_book_doc(self, book):
                 self.index.add(doc)
 
     def create_book_doc(self, book):


@@ -241,7 +242,7 @@ class Index(SolrIndex):
             self.remove_book(book, remove_snippets=False)
 
         book_doc = self.create_book_doc(book)
             self.remove_book(book, remove_snippets=False)
 
         book_doc = self.create_book_doc(book)

-        meta_fields = self.extract_metadata(book, book_info, dc_only=['source_name', 'authors', 'title'])
+        meta_fields = self.extract_metadata(book, book_info, dc_only=['source_name', 'authors', 'translators', 'title'])

         # let's not index it - it's only used for extracting publish date
         if 'source_name' in meta_fields:
             del meta_fields['source_name']
         # let's not index it - it's only used for extracting publish date
         if 'source_name' in meta_fields:
             del meta_fields['source_name']


@@ -250,7 +251,6 @@ class Index(SolrIndex):
             book_doc[n] = f
 
         book_doc['uid'] = "book%s" % book_doc['book_id']
             book_doc[n] = f
 
         book_doc['uid'] = "book%s" % book_doc['book_id']

-        print "ADD 2 %s" % book_doc

         self.index.add(book_doc)
         del book_doc
         book_fields = {
         self.index.add(book_doc)
         del book_doc
         book_fields = {


@@ -258,6 +258,7 @@ class Index(SolrIndex):
             'authors': meta_fields['authors'],
             'published_date': meta_fields['published_date']
             }
             'authors': meta_fields['authors'],
             'published_date': meta_fields['published_date']
             }

+

         if 'translators' in meta_fields:
             book_fields['translators'] = meta_fields['translators']
 
         if 'translators' in meta_fields:
             book_fields['translators'] = meta_fields['translators']
 


@@ -462,7 +463,6 @@ class Index(SolrIndex):
                         doc = add_part(snippets, header_index=position, header_type=header.tag,
                                        text=u''.join(footnote),
                                        is_footnote=True)
                         doc = add_part(snippets, header_index=position, header_type=header.tag,
                                        text=u''.join(footnote),
                                        is_footnote=True)

-                        print "ADD 3 %s" % doc

                         self.index.add(doc)
                         #print "@ footnote text: %s" % footnote
                         footnote = []
                         self.index.add(doc)
                         #print "@ footnote text: %s" % footnote
                         footnote = []


@@ -498,7 +498,6 @@ class Index(SolrIndex):
                                        text=fix_format(frag['text']),
                                        themes=frag['themes'])
                         #print '@ FRAG %s' % frag['content']
                                        text=fix_format(frag['text']),
                                        themes=frag['themes'])
                         #print '@ FRAG %s' % frag['content']

-                        print "ADD 4 %s" % doc

                         self.index.add(doc)
 
                         # Collect content.
                         self.index.add(doc)
 
                         # Collect content.


@@ -513,7 +512,6 @@ class Index(SolrIndex):
                                header_type=header.tag, text=fix_format(content))
                 #print '@ CONTENT: %s' % fix_format(content)
 
                                header_type=header.tag, text=fix_format(content))
                 #print '@ CONTENT: %s' % fix_format(content)
 

-                print "ADD 5 %s" % doc

                 self.index.add(doc)
 
         finally:
                 self.index.add(doc)
 
         finally:


@@ -653,7 +651,6 @@ class SearchResult(object):
             except catalogue.models.Fragment.DoesNotExist:
                 # stale index
                 continue
             except catalogue.models.Fragment.DoesNotExist:
                 # stale index
                 continue

-            print f

             # Figure out if we were searching for a token matching some word in theme name.
             themes = frag.tags.filter(category='theme')
             themes_hit = set()
             # Figure out if we were searching for a token matching some word in theme name.
             themes = frag.tags.filter(category='theme')
             themes_hit = set()


@@ -797,6 +794,7 @@ class Search(SolrIndex):
         modal - applies to boolean query
         fuzzy - should the query by fuzzy.
         """
         modal - applies to boolean query
         fuzzy - should the query by fuzzy.
         """

+        if query is None: query = ''

         q = self.index.Q()
         q = reduce(modal, map(lambda s: self.index.Q(**{field: s}),
                         query.split(r" ")), q)
         q = self.index.Q()
         q = reduce(modal, map(lambda s: self.index.Q(**{field: s}),
                         query.split(r" ")), q)


@@ -958,7 +956,6 @@ class Search(SolrIndex):
                     continue
                 text = snippets.get((int(position),
                                      int(length)))
                     continue
                 text = snippets.get((int(position),
                                      int(length)))

-                print "== %s -- %s ==" % (query, text)

                 snip = self.index.highlight(text=text, field=field, q=query)
                 snips[idx] = snip
                 if snip:
                 snip = self.index.highlight(text=text, field=field, q=query)
                 snips[idx] = snip
                 if snip:


@@ -975,6 +972,7 @@ class Search(SolrIndex):
         snips = map(lambda s: s and s.replace("/\n", "\n"), snips)
 
         searchresult.snippets = snips
         snips = map(lambda s: s and s.replace("/\n", "\n"), snips)
 
         searchresult.snippets = snips

+

         return snips
 
     def hint_tags(self, query, pdcounter=True, prefix=True):
         return snips
 
     def hint_tags(self, query, pdcounter=True, prefix=True):