@@ -54,33 +54,42 @@ You are ready to use rhodecode, to run i
Setting up Whoosh full text search
----------------------------------
Index for whoosh can be build starting from version 1.1 using paster command
passing repo locations to index, as well as Your config file that stores
whoosh index files locations. There is possible to pass `-f` to the options
Starting from version 1.1 whoosh index can be build using paster command.
You have to specify the config file that stores location of index, and
location of repositories (`--repo-location`). Starting from version 1.2 it is
also possible to specify a comma separated list of repositories (`--index-only`)
to build index only on chooses repositories skipping any other found in repos
location
There is possible also to pass `-f` to the options
to enable full index rebuild. Without that indexing will run always in in
incremental mode.
::
incremental mode::
paster make-index production.ini --repo-location=<location for repos>
for full index rebuild You can use
for full index rebuild You can use::
paster make-index production.ini -f --repo-location=<location for repos>
- For full text search You can either put crontab entry for
building index just for chosen repositories is possible with such command::
paster make-index production.ini --repo-location=<location for repos> --index-only=vcs,rhodecode
This command can be run even from crontab in order to do periodical
index builds and keep Your index always up to date. An example entry might
look like this
In order to do periodical index builds and keep Your index always up to date.
It's recommended to do a crontab entry for incremental indexing.
An example entry might look like this
/path/to/python/bin/paster /path/to/rhodecode/production.ini --repo-location=<location for repos>
When using incremental(default) mode whoosh will check last modification date
When using incremental (default) mode whoosh will check last modification date
of each file and add it to reindex if newer file is available. Also indexing
daemon checks for removed files and removes them from index.
@@ -6,6 +6,8 @@ from os.path import dirname as dn, join
#to get the rhodecode import
sys.path.append(dn(dn(dn(os.path.realpath(__file__)))))
from string import strip
from rhodecode.model import init_model
from rhodecode.model.scm import ScmModel
from rhodecode.config.environment import load_environment
@@ -71,6 +73,7 @@ class MakeIndex(BasePasterCommand):
index_location = config['index_dir']
repo_location = self.options.repo_location
repo_list = map(strip, self.options.repo_list.split(','))
#======================================================================
# WHOOSH DAEMON
@@ -80,7 +83,8 @@ class MakeIndex(BasePasterCommand):
try:
l = DaemonLock()
WhooshIndexingDaemon(index_location=index_location,
repo_location=repo_location)\
repo_location=repo_location,
repo_list=repo_list)\
.run(full_index=self.options.full_index)
l.release()
except LockHeld:
@@ -92,6 +96,12 @@ class MakeIndex(BasePasterCommand):
dest='repo_location',
help="Specifies repositories location to index REQUIRED",
)
self.parser.add_option('--index-only',
action='store',
dest='repo_list',
help="Specifies a comma separated list of repositores "
"to build index on OPTIONAL",
self.parser.add_option('-f',
action='store_true',
dest='full_index',
@@ -70,7 +70,7 @@ class WhooshIndexingDaemon(object):
"""
def __init__(self, indexname='HG_INDEX', index_location=None,
repo_location=None, sa=None):
repo_location=None, sa=None, repo_list=None):
self.indexname = indexname
self.index_location = index_location
@@ -82,6 +82,16 @@ class WhooshIndexingDaemon(object):
raise Exception('You have to provide repositories location')
self.repo_paths = ScmModel(sa).repo_scan(self.repo_location, None)
if repo_list:
filtered_repo_paths = {}
for repo_name, repo in self.repo_paths.items():
if repo_name in repo_list:
filtered_repo_paths[repo.name] = repo
self.repo_paths = filtered_repo_paths
self.initial = False
if not os.path.isdir(self.index_location):
os.makedirs(self.index_location)
@@ -154,8 +164,8 @@ class WhooshIndexingDaemon(object):
idx = create_in(self.index_location, SCHEMA, indexname=IDX_NAME)
writer = idx.writer()
print self.repo_paths.values()
for cnt, repo in enumerate(self.repo_paths.values()):
for repo in self.repo_paths.values():
log.debug('building index @ %s' % repo.path)
for idx_path in self.get_paths(repo):
Status change: