super(KallitheaAppConfig, self).__init__()

        self['package'] = kallithea

        self['prefer_toscawidgets2'] = False

        self['use_toscawidgets'] = False

        self['renderers'] = []

        # Enable json in expose

        self['renderers'].append('json')

        # Configure template rendering

        self['renderers'].append('mako')

        self['default_renderer'] = 'mako'

        self['use_dotted_templatenames'] = False

        # Configure Sessions, store data as JSON to avoid pickle security issues

        self['session.enabled'] = True

        self['session.data_serializer'] = 'json'

        # Configure the base SQLALchemy Setup

        self['use_sqlalchemy'] = True

        self['model'] = kallithea.model.base

        self['DBSession'] = kallithea.model.meta.Session

        # Configure App without an authentication backend.

        self['auth_backend'] = None

        # Use custom error page for these errors. By default, Turbogears2 does not add

        # 400 in this list.

        # Explicitly listing all is considered more robust than appending to defaults,

        # in light of possible future framework changes.

        self['errorpage.status_codes'] = [400, 401, 403, 404]

        # Disable transaction manager -- currently Kallithea takes care of transactions itself

        self['tm.enabled'] = False

base_config = KallitheaAppConfig()

# TODO still needed as long as we use pylonslib

sys.modules['pylons'] = tg

# DebugBar, a debug toolbar for TurboGears2.

# (https://github.com/TurboGears/tgext.debugbar)

# To enable it, install 'tgext.debugbar' and 'kajiki', and run Kallithea with

# 'debug = true' (not in production!)

# See the Kallithea documentation for more information.

try:

    from tgext.debugbar import enable_debugbar

    import kajiki # only to check its existence

except ImportError:

    pass

else:

    base_config['renderers'].append('kajiki')

    enable_debugbar(base_config)

def setup_configuration(app):

    config = app.config

    if not kallithea.lib.locale.current_locale_is_valid():

        log.error("Terminating ...")

        sys.exit(1)

    # Mercurial sets encoding at module import time, so we have to monkey patch it

    hgencoding = config.get('hgencoding')

    if hgencoding:

        mercurial.encoding.encoding = hgencoding

    if config.get('ignore_alembic_revision', False):

        log.warn('database alembic revision checking is disabled')

    else:

        dbconf = config['sqlalchemy.url']

        alembic_cfg = alembic.config.Config()

        alembic_cfg.set_main_option('script_location', 'kallithea:alembic')

        alembic_cfg.set_main_option('sqlalchemy.url', dbconf)

        script_dir = ScriptDirectory.from_config(alembic_cfg)

        available_heads = sorted(script_dir.get_heads())

        engine = create_engine(dbconf)

        with engine.connect() as conn:

            context = MigrationContext.configure(conn)

            current_heads = sorted(str(s) for s in context.get_current_heads())

        if current_heads != available_heads:

            log.error('Failed to run Kallithea:\n\n'

                      'The database version does not match the Kallithea version.\n'

                      'Please read the documentation on how to upgrade or downgrade the database.\n'

                      'Current database version id(s): %s\n'

                      'Expected database version id(s): %s\n'

                      'If you are a developer and you know what you are doing, you can add `ignore_alembic_revision = True` '

                      'to your .ini file to skip the check.\n' % (' '.join(current_heads), ' '.join(available_heads)))

            sys.exit(1)

    # store some globals into kallithea

    kallithea.CONFIG = config

    load_rcextensions(root_path=config['here'])

    set_available_permissions(config)

    repos_path = make_ui().configitems('paths')[0][1]

    config['base_path'] = repos_path

    set_app_settings(config)

    instance_id = kallithea.CONFIG.get('instance_id', '*')

    if instance_id == '*':

        instance_id = '%s-%s' % (platform.uname()[1], os.getpid())

        kallithea.CONFIG['instance_id'] = instance_id

    # update kallithea.CONFIG with the meanwhile changed 'config'

    kallithea.CONFIG.update(config)

    # configure vcs and indexer libraries (they are supposed to be independent

    # as much as possible and thus avoid importing tg.config or

    # kallithea.CONFIG).

    set_vcs_config(kallithea.CONFIG)

    set_indexer_config(kallithea.CONFIG)

    check_git_version()

hooks.register('configure_new_app', setup_configuration)

def setup_application(app):

    config = app.config

    # we want our low level middleware to get to the request ASAP. We don't

    # need any stack middleware in them - especially no StatusCodeRedirect buffering

    app = SimpleHg(app, config)

    app = SimpleGit(app, config)

    # Enable https redirects based on HTTP_X_URL_SCHEME set by proxy

    if any(asbool(config.get(x)) for x in ['https_fixup', 'force_https', 'use_htsts']):

        app = HttpsFixup(app, config)

    app = PermanentRepoUrl(app, config)

    return app

hooks.register('before_config', setup_application)

# -*- coding: utf-8 -*-

# This program is free software: you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation, either version 3 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program.  If not, see <http://www.gnu.org/licenses/>.

"""

kallithea.controllers.search

~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Search controller for Kallithea

This file was forked by the Kallithea project in July 2014.

Original author and date, and relevant copyright and licensing information is below:

:created_on: Aug 7, 2010

:author: marcink

:copyright: (c) 2013 RhodeCode GmbH, and others.

:license: GPLv3, see LICENSE.md for more details.

"""

import logging

import traceback

import urllib

from tg.i18n import ugettext as _

from tg import request, config, tmpl_context as c

from whoosh.index import open_dir, exists_in, EmptyIndexError

from whoosh.qparser import QueryParser, QueryParserError

from whoosh.query import Phrase, Prefix

from webhelpers2.html.tools import update_params

from kallithea.lib.auth import LoginRequired

from kallithea.lib.base import BaseRepoController, render

from kallithea.lib.indexers import CHGSETS_SCHEMA, SCHEMA, CHGSET_IDX_NAME, \

    IDX_NAME, WhooshResultWrapper

from kallithea.lib.page import Page

from kallithea.lib.utils2 import safe_str, safe_int

from kallithea.model.repo import RepoModel

log = logging.getLogger(__name__)

class SearchController(BaseRepoController):

    @LoginRequired(allow_default_user=True)

    def index(self, repo_name=None):

        c.repo_name = repo_name

        c.formated_results = []

        c.runtime = ''

        c.cur_query = request.GET.get('q', None)

        c.cur_type = request.GET.get('type', 'content')

        c.cur_search = search_type = {'content': 'content',

                                      'commit': 'message',

                                      'path': 'path',

                                      'repository': 'repository'

                                      }.get(c.cur_type, 'content')

        index_name = {

            'content': IDX_NAME,

            'commit': CHGSET_IDX_NAME,

            'path': IDX_NAME

        }.get(c.cur_type, IDX_NAME)

        schema_defn = {

            'content': SCHEMA,

            'commit': CHGSETS_SCHEMA,

            'path': SCHEMA

        }.get(c.cur_type, SCHEMA)

        log.debug('IDX: %s', index_name)

        log.debug('SCHEMA: %s', schema_defn)

        if c.cur_query:

            cur_query = c.cur_query.lower()

            log.debug(cur_query)

        if c.cur_query:

            p = safe_int(request.GET.get('page'), 1)

            highlight_items = set()

            try:

                if not exists_in(index_dir, index_name):

                    raise EmptyIndexError

                idx = open_dir(index_dir, indexname=index_name)

                searcher = idx.searcher()

                qp = QueryParser(search_type, schema=schema_defn)

                if c.repo_name:

                    # use "repository_rawname:" instead of "repository:"

                    # for case-sensitive matching

                    cur_query = u'repository_rawname:%s %s' % (c.repo_name, cur_query)

                try:

                    query = qp.parse(unicode(cur_query))

                    # extract words for highlight

                    if isinstance(query, Phrase):

                        highlight_items.update(query.words)

                    elif isinstance(query, Prefix):

                        highlight_items.add(query.text)

                    else:

                        for i in query.all_terms():

                            if i[0] in ['content', 'message']:

                                highlight_items.add(i[1])

                    matcher = query.matcher(searcher)

                    log.debug('query: %s', query)

                    log.debug('hl terms: %s', highlight_items)

                    results = searcher.search(query)

                    res_ln = len(results)

                    c.runtime = '%s results (%.3f seconds)' % (

                        res_ln, results.runtime

                    )

                    def url_generator(**kw):

                        q = urllib.quote(safe_str(c.cur_query))

                        return update_params("?q=%s&type=%s" % (q, safe_str(c.cur_type)), **kw)

                    repo_location = RepoModel().repos_path

                    c.formated_results = Page(

                        WhooshResultWrapper(search_type, searcher, matcher,

                                            highlight_items, repo_location),

                        page=p,

                        item_count=res_ln,

                        items_per_page=10,

                        url=url_generator

                    )

                except QueryParserError:

                    c.runtime = _('Invalid search query. Try quoting it.')

                searcher.close()

            except EmptyIndexError:

                log.error("Empty search index - run 'kallithea-cli index-create' regularly")

                c.runtime = _('The server has no search index.')

            except Exception:

                log.error(traceback.format_exc())

                c.runtime = _('An error occurred during search operation.')

        # Return a rendered template

        return render('/search/search.html')

#

# You should have received a copy of the GNU General Public License

# along with this program.  If not, see <http://www.gnu.org/licenses/>.

"""

kallithea.lib.celerylib

~~~~~~~~~~~~~~~~~~~~~~~

celery libs for Kallithea

This file was forked by the Kallithea project in July 2014.

Original author and date, and relevant copyright and licensing information is below:

:created_on: Nov 27, 2010

:author: marcink

:copyright: (c) 2013 RhodeCode GmbH, and others.

:license: GPLv3, see LICENSE.md for more details.

"""

import os

import logging

from tg import config

from hashlib import md5

from decorator import decorator

from kallithea import CELERY_ON, CELERY_EAGER

from kallithea.lib.utils2 import safe_str

from kallithea.lib.pidlock import DaemonLock, LockHeld

from kallithea.model.base import init_model

from kallithea.model import meta

log = logging.getLogger(__name__)

class FakeTask(object):

    """Fake a sync result to make it look like a finished task"""

    def __init__(self, result):

        self.result = result

    def failed(self):

        return False

    traceback = None # if failed

    task_id = None

def task(f_org):

    """Wrapper of celery.task.task, running async if CELERY_ON

    """

    if CELERY_ON:

        def f_async(*args, **kwargs):

            log.info('executing %s task', f_org.__name__)

            try:

                f_org(*args, **kwargs)

            finally:

                log.info('executed %s task', f_org.__name__)

        f_async.__name__ = f_org.__name__

        from kallithea.lib import celerypylons

        runner = celerypylons.task(ignore_result=True)(f_async)

        def f_wrapped(*args, **kwargs):

            t = runner.apply_async(args=args, kwargs=kwargs)

            log.info('executing task %s in async mode - id %s', f_org, t.task_id)

            return t

    else:

        def f_wrapped(*args, **kwargs):

            log.info('executing task %s in sync', f_org.__name__)

            try:

                result = f_org(*args, **kwargs)

            except Exception as e:

                log.error('exception executing sync task %s in sync: %r', f_org.__name__, e)

                raise # TODO: return this in FakeTask as with async tasks?

            return FakeTask(result)

    return f_wrapped

def __get_lockkey(func, *fargs, **fkwargs):

    params = list(fargs)

    params.extend(['%s-%s' % ar for ar in fkwargs.items()])

    func_name = str(func.__name__) if hasattr(func, '__name__') else str(func)

    lockkey = 'task_%s.lock' % \

        md5(func_name + '-' + '-'.join(map(safe_str, params))).hexdigest()

    return lockkey

def locked_task(func):

    def __wrapper(func, *fargs, **fkwargs):

        lockkey = __get_lockkey(func, *fargs, **fkwargs)

        log.info('running task with lockkey %s', lockkey)

        try:

            l = DaemonLock(os.path.join(lockkey_path, lockkey))

            ret = func(*fargs, **fkwargs)

            l.release()

            return ret

        except LockHeld:

            log.info('LockHeld')

            return 'Task with key %s already running' % lockkey

    return decorator(__wrapper, func)

def get_session():

    sa = meta.Session()

    return sa

def dbsession(func):

    def __wrapper(func, *fargs, **fkwargs):

        try:

            ret = func(*fargs, **fkwargs)

            return ret

        finally:

            if CELERY_ON and not CELERY_EAGER:

                meta.Session.remove()

    return decorator(__wrapper, func)

# -*- coding: utf-8 -*-

# This program is free software: you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation, either version 3 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program.  If not, see <http://www.gnu.org/licenses/>.

"""

kallithea.lib.celerylib.tasks

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Kallithea task modules, containing all task that suppose to be run

by celery daemon

This file was forked by the Kallithea project in July 2014.

Original author and date, and relevant copyright and licensing information is below:

:created_on: Oct 6, 2010

:author: marcink

:copyright: (c) 2013 RhodeCode GmbH, and others.

:license: GPLv3, see LICENSE.md for more details.

"""

import os

import traceback

import logging

import rfc822

from collections import OrderedDict

from time import mktime

from operator import itemgetter

from tg import config

from kallithea import CELERY_ON

from kallithea.lib import celerylib

from kallithea.lib.helpers import person

from kallithea.lib.rcmail.smtp_mailer import SmtpMailer

from kallithea.lib.utils import action_logger

from kallithea.lib.utils2 import str2bool

from kallithea.lib.vcs.utils import author_email

from kallithea.lib.compat import json

from kallithea.lib.hooks import log_create_repository

from kallithea.model.db import Statistics, RepoGroup, Repository, User

__all__ = ['whoosh_index', 'get_commits_stats', 'send_email']

log = logging.getLogger(__name__)

@celerylib.task

@celerylib.locked_task

@celerylib.dbsession

def whoosh_index(repo_location, full_index):

    from kallithea.lib.indexers.daemon import WhooshIndexingDaemon

    celerylib.get_session() # initialize database connection

    index_location = config['index_dir']

    WhooshIndexingDaemon(index_location=index_location,

                         repo_location=repo_location) \

                         .run(full_index=full_index)

@celerylib.task

@celerylib.dbsession

def get_commits_stats(repo_name, ts_min_y, ts_max_y, recurse_limit=100):

    DBS = celerylib.get_session()

    lockkey = celerylib.__get_lockkey('get_commits_stats', repo_name, ts_min_y,

                            ts_max_y)

    log.info('running task with lockkey %s', lockkey)

    try:

        lock = celerylib.DaemonLock(os.path.join(lockkey_path, lockkey))

        # for js data compatibility cleans the key for person from '

        akc = lambda k: person(k).replace('"', "")

        co_day_auth_aggr = {}

        commits_by_day_aggregate = {}

        repo = Repository.get_by_repo_name(repo_name)

        if repo is None:

            return True

        repo = repo.scm_instance

        repo_size = repo.count()

        # return if repo have no revisions

        if repo_size < 1:

            lock.release()

            return True

        skip_date_limit = True

        last_rev = None

        last_cs = None

        timegetter = itemgetter('time')

        dbrepo = DBS.query(Repository) \

            .filter(Repository.repo_name == repo_name).scalar()

        cur_stats = DBS.query(Statistics) \

            .filter(Statistics.repository == dbrepo).scalar()

        if cur_stats is not None:

            last_rev = cur_stats.stat_on_revision

        if last_rev == repo.get_changeset().revision and repo_size > 1:

            # pass silently without any work if we're not on first revision or

            # current state of parsing revision(from db marker) is the

            # last revision

            lock.release()

            return True

        if cur_stats:

            commits_by_day_aggregate = OrderedDict(json.loads(

                                        cur_stats.commit_activity_combined))

            co_day_auth_aggr = json.loads(cur_stats.commit_activity)

        log.debug('starting parsing %s', parse_limit)

        lmktime = mktime

        last_rev = last_rev + 1 if last_rev >= 0 else 0

        log.debug('Getting revisions from %s to %s',

             last_rev, last_rev + parse_limit

        )

        for cs in repo[last_rev:last_rev + parse_limit]:

            log.debug('parsing %s', cs)

            last_cs = cs  # remember last parsed changeset

            k = lmktime([cs.date.timetuple()[0], cs.date.timetuple()[1],

                          cs.date.timetuple()[2], 0, 0, 0, 0, 0, 0])

            if akc(cs.author) in co_day_auth_aggr:

                try:

                    l = [timegetter(x) for x in

                         co_day_auth_aggr[akc(cs.author)]['data']]

                    time_pos = l.index(k)

                except ValueError:

                    time_pos = None

                if time_pos >= 0 and time_pos is not None:

                    datadict = \

                        co_day_auth_aggr[akc(cs.author)]['data'][time_pos]

                    datadict["commits"] += 1

                    datadict["added"] += len(cs.added)

                    datadict["changed"] += len(cs.changed)

                    datadict["removed"] += len(cs.removed)

                else:

                    if k >= ts_min_y and k <= ts_max_y or skip_date_limit:

                        datadict = {"time": k,

                                    "commits": 1,

                                    "added": len(cs.added),

                                    "changed": len(cs.changed),

                                    "removed": len(cs.removed),

                                   }

                        co_day_auth_aggr[akc(cs.author)]['data'] \

                            .append(datadict)

            else:

                if k >= ts_min_y and k <= ts_max_y or skip_date_limit:

                    co_day_auth_aggr[akc(cs.author)] = {

                                        "label": akc(cs.author),

                                        "data": [{"time":k,

                                                 "commits":1,

                                                 "added":len(cs.added),

                                                 "changed":len(cs.changed),

                                                 "removed":len(cs.removed),

                                                 }],

                                        "schema": ["commits"],

                                        }

            # gather all data by day

            if k in commits_by_day_aggregate:

                commits_by_day_aggregate[k] += 1

            else:

                commits_by_day_aggregate[k] = 1

        overview_data = sorted(commits_by_day_aggregate.items(),

                               key=itemgetter(0))

        if not co_day_auth_aggr:

            co_day_auth_aggr[akc(repo.contact)] = {

                "label": akc(repo.contact),

                "data": [0, 1],

                "schema": ["commits"],

            }

import mock

import os

from kallithea.tests.base import *

class TestSearchController(TestController):

    def test_index(self):

        self.log_user()

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

        response.mustcontain('class="form-control" id="q" name="q" type="text"')

        # Test response...

    def test_empty_search(self, tmpdir):

        self.log_user()

        config_mock = {

        }

        with mock.patch('kallithea.controllers.search.config', config_mock):

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

                                    {'q': HG_REPO})

            response.mustcontain('The server has no search index.')

    def test_normal_search(self):

        self.log_user()

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

                                {'q': 'def repo'})

        response.mustcontain('58 results')

    def test_repo_search(self):

        self.log_user()

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

                                {'q': 'repository:%s def test' % HG_REPO})

        response.mustcontain('18 results')

    def test_search_last(self):

        self.log_user()

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

                                {'q': 'last:t', 'type': 'commit'})

        response.mustcontain('2 results')

    def test_search_commit_message(self):

        self.log_user()

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

                    {'q': 'bother to ask where to fetch repo during tests',

                     'type': 'commit'})

        response.mustcontain('2 results')

        response.mustcontain('a00c1b6f5d7a6ae678fd553a8b81d92367f7ecf1')

        response.mustcontain('c6eb379775c578a95dad8ddab53f963b80894850')

    def test_search_commit_message_hg_repo(self):

        self.log_user()

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

                                    repo_name=HG_REPO),

                    {'q': 'bother to ask where to fetch repo during tests',

                     'type': 'commit'})

        response.mustcontain('1 results')

        response.mustcontain('a00c1b6f5d7a6ae678fd553a8b81d92367f7ecf1')

    def test_search_commit_changed_file(self):

        self.log_user()

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

                                {'q': 'changed:tests/utils.py',

                                 'type': 'commit'})

        response.mustcontain('29 results')

    def test_search_commit_changed_files_get_commit(self):

        self.log_user()

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

                                {'q': 'changed:vcs/utils/archivers.py',

                                 'type': 'commit'})

        response.mustcontain('8 results')

        response.mustcontain('25213a5fbb048dff8ba65d21e466a835536e5b70')

        response.mustcontain('47aedd538bf616eedcb0e7d630ea476df0e159c7')

        response.mustcontain('f5d23247fad4856a1dabd5838afade1e0eed24fb')

        response.mustcontain('04ad456aefd6461aea24f90b63954b6b1ce07b3e')

        response.mustcontain('c994f0de03b2a0aa848a04fc2c0d7e737dba31fc')

        response.mustcontain('d1f898326327e20524fe22417c22d71064fe54a1')

        response.mustcontain('fe568b4081755c12abf6ba673ba777fc02a415f3')

        response.mustcontain('bafe786f0d8c2ff7da5c1dcfcfa577de0b5e92f1')

    def test_search_commit_added_file(self):

        self.log_user()

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

                                {'q': 'added:README.rst',

                                 'type': 'commit'})

        response.mustcontain('2 results')

        # Hg

        response.mustcontain('3803844fdbd3b711175fc3da9bdacfcd6d29a6fb')

        # Git

        response.mustcontain('ff7ca51e58c505fec0dd2491de52c622bb7a806b')

    def test_search_author(self):

        self.log_user()

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

                    {'q': 'author:marcin@python-blog.com raw_id:b986218ba1c9b0d6a259fac9b050b1724ed8e545',

                     'type': 'commit'})

        response.mustcontain('1 results')

    def test_search_file_name(self):

        self.log_user()

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

                    {'q': 'README.rst', 'type': 'path'})

        response.mustcontain('2 results')