Changeset - c0ddc86b4654
[Not reviewed]
beta
0 1 0
Marcin Kuzminski - 13 years ago 2012-09-15 21:08:53
marcin@python-works.com
Fix possible exception about repo_name not defined, on whoosh indexer when using index-only option
1 file changed with 1 insertions and 0 deletions:
0 comments (0 inline, 0 general)
rhodecode/lib/indexers/daemon.py
Show inline comments
 
@@ -193,96 +193,97 @@ class WhooshIndexingDaemon(object):
 
        indexed = 0
 
        for cs in repo.get_changesets(start=start_rev):
 
            log.debug('    >> %s' % cs)
 
            writer.add_document(
 
                raw_id=unicode(cs.raw_id),
 
                owner=unicode(repo.contact),
 
                date=cs._timestamp,
 
                repository=safe_unicode(repo_name),
 
                author=cs.author,
 
                message=cs.message,
 
                last=cs.last,
 
                added=u' '.join([safe_unicode(node.path) for node in cs.added]).lower(),
 
                removed=u' '.join([safe_unicode(node.path) for node in cs.removed]).lower(),
 
                changed=u' '.join([safe_unicode(node.path) for node in cs.changed]).lower(),
 
                parents=u' '.join([cs.raw_id for cs in cs.parents]),
 
            )
 
            indexed += 1
 

	
 
        log.debug('indexed %d changesets for repo %s' % (indexed, repo_name))
 
        return indexed
 

	
 
    def index_files(self, file_idx_writer, repo_name, repo):
 
        """
 
        Index files for given repo_name
 

	
 
        :param file_idx_writer: the whoosh index writer to add to
 
        :param repo_name: name of the repository we're indexing
 
        :param repo: instance of vcs repo
 
        """
 
        i_cnt = iwc_cnt = 0
 
        log.debug('building index for [%s]' % repo.path)
 
        for idx_path in self.get_paths(repo):
 
            i, iwc = self.add_doc(file_idx_writer, idx_path, repo, repo_name)
 
            i_cnt += i
 
            iwc_cnt += iwc
 

	
 
        log.debug('added %s files %s with content for repo %s' %
 
                  (i_cnt + iwc_cnt, iwc_cnt, repo.path))
 
        return i_cnt, iwc_cnt
 

	
 
    def update_changeset_index(self):
 
        idx = open_dir(self.index_location, indexname=CHGSET_IDX_NAME)
 

	
 
        with idx.searcher() as searcher:
 
            writer = idx.writer()
 
            writer_is_dirty = False
 
            try:
 
                indexed_total = 0
 
                repo_name = None
 
                for repo_name, repo in self.repo_paths.items():
 
                    # skip indexing if there aren't any revs in the repo
 
                    num_of_revs = len(repo)
 
                    if num_of_revs < 1:
 
                        continue
 

	
 
                    qp = QueryParser('repository', schema=CHGSETS_SCHEMA)
 
                    q = qp.parse(u"last:t AND %s" % repo_name)
 

	
 
                    results = searcher.search(q)
 

	
 
                    # default to scanning the entire repo
 
                    last_rev = 0
 
                    start_id = None
 

	
 
                    if len(results) > 0:
 
                        # assuming that there is only one result, if not this
 
                        # may require a full re-index.
 
                        start_id = results[0]['raw_id']
 
                        last_rev = repo.get_changeset(revision=start_id).revision
 

	
 
                    # there are new changesets to index or a new repo to index
 
                    if last_rev == 0 or num_of_revs > last_rev + 1:
 
                        # delete the docs in the index for the previous
 
                        # last changeset(s)
 
                        for hit in results:
 
                            q = qp.parse(u"last:t AND %s AND raw_id:%s" %
 
                                            (repo_name, hit['raw_id']))
 
                            writer.delete_by_query(q)
 

	
 
                        # index from the previous last changeset + all new ones
 
                        indexed_total += self.index_changesets(writer,
 
                                                repo_name, repo, start_id)
 
                        writer_is_dirty = True
 
                log.debug('indexed %s changesets for repo %s' % (
 
                             indexed_total, repo_name)
 
                )
 
            finally:
 
                if writer_is_dirty:
 
                    log.debug('>> COMMITING CHANGES TO CHANGESET INDEX<<')
 
                    writer.commit(merge=True)
 
                    log.debug('>> COMMITTED CHANGES TO CHANGESET INDEX<<')
 
                else:
 
                    writer.cancel
 
                    log.debug('>> NOTHING TO COMMIT<<')
 

	
 
    def update_file_index(self):
 
        log.debug((u'STARTING INCREMENTAL INDEXING UPDATE FOR EXTENSIONS %s '
0 comments (0 inline, 0 general)