log.debug("couldn't add doc - %s did not have %r at %s", repo, path, index_rev)

            return 0, 0

        indexed = indexed_w_content = 0

        if self.is_indexable_node(node):

            u_content = node.content

            if not isinstance(u_content, unicode):

                log.warning('  >> %s Could not get this content as unicode '

                            'replacing with empty content' % path)

                u_content = u''

            else:

                log.debug('    >> %s [WITH CONTENT]', path)

                indexed_w_content += 1

        else:

            log.debug('    >> %s', path)

            # just index file name without it's content

            u_content = u''

            indexed += 1

        p = safe_unicode(path)

        writer.add_document(

            fileid=p,

            owner=unicode(repo.contact),

            repository=safe_unicode(repo_name),

            path=p,

            content=u_content,

            modtime=self.get_node_mtime(node),

            extension=node.extension

        )

        return indexed, indexed_w_content

    def index_changesets(self, writer, repo_name, repo, start_rev=None):

        """

        Add all changeset in the vcs repo starting at start_rev

        to the index writer

        :param writer: the whoosh index writer to add to

        :param repo_name: name of the repository from whence the

          changeset originates including the repository group

        :param repo: the vcs repository instance to index changesets for,

          the presumption is the repo has changesets to index

        :param start_rev=None: the full sha id to start indexing from

          if start_rev is None then index from the first changeset in

          the repo

        """

        if start_rev is None:

            start_rev = repo[0].raw_id

        log.debug('indexing changesets in %s starting at rev: %s',

                  repo_name, start_rev)

        indexed = 0

        cs_iter = repo.get_changesets(start=start_rev)

        total = len(cs_iter)

        for cs in cs_iter:

            log.debug('    >> %s/%s', cs, total)

            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 @revision:%s', repo.path,

    def __str__(self):

        return '<%s at %s>' % (self.__class__.__name__, self.path)

    def __repr__(self):

        return self.__str__()

    def __len__(self):

        return self.count()

    def __eq__(self, other):

        same_instance = isinstance(other, self.__class__)

        return same_instance and getattr(other, 'path', None) == self.path

    def __ne__(self, other):

        return not self.__eq__(other)

    @LazyProperty

    def alias(self):

        for k, v in settings.BACKENDS.items():

            if v.split('.')[-1] == str(self.__class__.__name__):

                return k

    @LazyProperty

    def name(self):

        raise NotImplementedError

    @property

    def name_unicode(self):

        return safe_unicode(self.name)

    @LazyProperty

    def owner(self):

        raise NotImplementedError

    @LazyProperty

    def description(self):

        raise NotImplementedError

    @LazyProperty

    def size(self):

        """

        Returns combined size in bytes for all repository files

        """

        size = 0

        try:

            tip = self.get_changeset()

            for topnode, dirs, files in tip.walk('/'):

        rebuild_index(full_index=True) # rebuild fully for subsequent tests

    @parametrize('reponame', [

        (u'indexing_test'),

        (u'indexing_test-fork'),

        (u'group/indexing_test'),

        (u'this-is-it'),

        (u'*-fork'),

        (u'group/*'),

    ])

    @parametrize('searchtype,query,hit', [

        ('content', 'this_should_be_unique_content', 1),

        ('commit', 'this_should_be_unique_commit_log', 1),

        ('path', 'this_should_be_unique_filename.txt', 1),

    ])

    def test_repository_tokenization(self, reponame, searchtype, query, hit):

        self.log_user()

        q = 'repository:%s %s' % (reponame, query)

        response = self.app.get(url(controller='search', action='index'),

                                {'q': q, 'type': searchtype})

        response.mustcontain('>%d results' % hit)

    @parametrize('searchtype,query,hit', [

        ('content', 'this_should_be_unique_content', 1),

        ('commit', 'this_should_be_unique_commit_log', 1),

        ('path', 'this_should_be_unique_filename.txt', 1),

    ])

    def test_repository_case_sensitivity(self, searchtype, query, hit):

        self.log_user()

        lname = u'indexing_test-foo'

        uname = u'indexing_test-FOO'

        # (1) "repository:REPONAME" condition should match against

        # repositories case-insensitively

        q = 'repository:%s %s' % (lname, query)

        response = self.app.get(url(controller='search', action='index'),

                                {'q': q, 'type': searchtype})

        response.mustcontain('>%d results' % (hit * 2))

        # (2) on the other hand, searching under the specific

        # repository should return results only for that repository,

        # even if specified name matches against another repository

        # case-insensitively.

        response = self.app.get(url(controller='search', action='index',