from celery.decorators import task

from operator import itemgetter

from pylons.i18n.translation import _

from rhodecode.lib.celerylib import run_task, locked_task

from rhodecode.lib.helpers import person

from rhodecode.lib.smtp_mailer import SmtpMailer

from rhodecode.lib.utils import OrderedDict

from time import mktime

from vcs.backends.hg import MercurialRepository

import traceback

try:

    import json

except ImportError:

    #python 2.5 compatibility

    import simplejson as json

try:

    from celeryconfig import PYLONS_CONFIG as config

    celery_on = True

except ImportError:

    #if celeryconfig is not present let's just load our pylons

    #config instead

    from pylons import config

    celery_on = False

__all__ = ['whoosh_index', 'get_commits_stats',

           'reset_user_password', 'send_email']

def get_session():

    if celery_on:

        from sqlalchemy import engine_from_config

        from sqlalchemy.orm import sessionmaker, scoped_session

        engine = engine_from_config(dict(config.items('app:main')), 'sqlalchemy.db1.')

        sa = scoped_session(sessionmaker(bind=engine))

    else:

        #If we don't use celery reuse our current application Session

        from rhodecode.model.meta import Session

        sa = Session

    return sa

def get_hg_settings():

    from rhodecode.model.db import RhodeCodeSettings

    sa = get_session()

    ret = sa.query(RhodeCodeSettings).all()

    if not ret:

        raise Exception('Could not get application settings !')

    settings = {}

    for each in ret:

        settings['rhodecode_' + each.app_settings_name] = each.app_settings_value    

    return settings

def get_hg_ui_settings():

    from rhodecode.model.db import RhodeCodeUi

    sa = get_session()

    ret = sa.query(RhodeCodeUi).all()

    if not ret:

        raise Exception('Could not get application ui settings !')

    settings = {}

    for each in ret:

        k = each.ui_key

        v = each.ui_value

        if k == '/':

            k = 'root_path'

        if k.find('.') != -1:

            k = k.replace('.', '_')

        if each.ui_section == 'hooks':

            v = each.ui_active

        settings[each.ui_section + '_' + k] = v  

    return settings   

@task

@locked_task

def whoosh_index(repo_location, full_index):

    log = whoosh_index.get_logger()

    from rhodecode.lib.indexers.daemon import WhooshIndexingDaemon

    WhooshIndexingDaemon(repo_location=repo_location).run(full_index=full_index)

@task

@locked_task

def get_commits_stats(repo_name, ts_min_y, ts_max_y):

    from rhodecode.model.db import Statistics, Repository

    log = get_commits_stats.get_logger()

    author_key_cleaner = lambda k: person(k).replace('"', "") #for js data compatibilty

    commits_by_day_author_aggregate = {}

    commits_by_day_aggregate = {}

    repos_path = get_hg_ui_settings()['paths_root_path'].replace('*', '')

    repo = MercurialRepository(repos_path + repo_name)

    skip_date_limit = True

    parse_limit = 350 #limit for single task changeset parsing optimal for

    last_rev = 0

    last_cs = None

    timegetter = itemgetter('time')

    sa = get_session()

    dbrepo = sa.query(Repository)\

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

    cur_stats = sa.query(Statistics)\

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

    if cur_stats:

        last_rev = cur_stats.stat_on_revision

    if not repo.revisions:

        return True

    if last_rev == repo.revisions[-1] and len(repo.revisions) > 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

        return True

    if cur_stats:

        commits_by_day_aggregate = OrderedDict(

                                       json.loads(

                                        cur_stats.commit_activity_combined))

        commits_by_day_author_aggregate = json.loads(cur_stats.commit_activity)

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

    for cnt, rev in enumerate(repo.revisions[last_rev:]):

        last_cs = cs = repo.get_changeset(rev)

        k = '%s-%s-%s' % (cs.date.timetuple()[0], cs.date.timetuple()[1],

                          cs.date.timetuple()[2])

        timetupple = [int(x) for x in k.split('-')]

        timetupple.extend([0 for _ in xrange(6)])

        k = mktime(timetupple)

        if commits_by_day_author_aggregate.has_key(author_key_cleaner(cs.author)):

            try:

                l = [timegetter(x) for x in commits_by_day_author_aggregate\

                        [author_key_cleaner(cs.author)]['data']]

                time_pos = l.index(k)

            except ValueError:

                time_pos = False

            if time_pos >= 0 and time_pos is not False:

                datadict = commits_by_day_author_aggregate\

                    [author_key_cleaner(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),

                               }

                    commits_by_day_author_aggregate\

                        [author_key_cleaner(cs.author)]['data'].append(datadict)

        else:

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

                commits_by_day_author_aggregate[author_key_cleaner(cs.author)] = {

                                    "label":author_key_cleaner(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 commits_by_day_aggregate.has_key(k):

            commits_by_day_aggregate[k] += 1

        else:

            commits_by_day_aggregate[k] = 1

        if cnt >= parse_limit:

            #don't fetch to much data since we can freeze application

            break

    overview_data = []

    for k, v in commits_by_day_aggregate.items():

        overview_data.append([k, v])

    overview_data = sorted(overview_data, key=itemgetter(0))

    if not commits_by_day_author_aggregate:

        commits_by_day_author_aggregate[author_key_cleaner(repo.contact)] = {

            "label":author_key_cleaner(repo.contact),

            "data":[0, 1],

            "schema":["commits"],

        }

    stats = cur_stats if cur_stats else Statistics()

    stats.commit_activity = json.dumps(commits_by_day_author_aggregate)

    stats.commit_activity_combined = json.dumps(overview_data)

    log.debug('last revison %s', last_rev)

    leftovers = len(repo.revisions[last_rev:])

    log.debug('revisions to parse %s', leftovers)

    if last_rev == 0 or leftovers < parse_limit:    

        stats.languages = json.dumps(__get_codes_stats(repo_name))

    stats.repository = dbrepo

    stats.stat_on_revision = last_cs.revision

    try:

        sa.add(stats)

        sa.commit()    

    except:

        log.error(traceback.format_exc())

        sa.rollback()

        return False

    if len(repo.revisions) > 1:

        run_task(get_commits_stats, repo_name, ts_min_y, ts_max_y)

    return True

@task

def reset_user_password(user_email):

    log = reset_user_password.get_logger()

    from rhodecode.lib import auth

    from rhodecode.model.db import User

    try:

        try:

            sa = get_session()

            user = sa.query(User).filter(User.email == user_email).scalar()

            new_passwd = auth.PasswordGenerator().gen_password(8,

                             auth.PasswordGenerator.ALPHABETS_BIG_SMALL)

            if user:

                user.password = auth.get_crypt_password(new_passwd)

                sa.add(user)

                sa.commit()

                log.info('change password for %s', user_email)

            if new_passwd is None:

                raise Exception('unable to generate new password')

        except:

            log.error(traceback.format_exc())

            sa.rollback()

        run_task(send_email, user_email,

                 "Your new rhodecode password",

                 'Your new rhodecode password:%s' % (new_passwd))

        log.info('send new password mail to %s', user_email)

    except:

        log.error('Failed to update user password')

        log.error(traceback.format_exc())

    return True

@task    

def send_email(recipients, subject, body):

    log = send_email.get_logger()

    email_config = dict(config.items('DEFAULT')) 

    mail_from = email_config.get('app_email_from')

    user = email_config.get('smtp_username')

    passwd = email_config.get('smtp_password')

    mail_server = email_config.get('smtp_server')

    mail_port = email_config.get('smtp_port')

    tls = email_config.get('smtp_use_tls')

    ssl = False

    try:

        m = SmtpMailer(mail_from, user, passwd, mail_server,

                       mail_port, ssl, tls)

        m.send(recipients, subject, body)  

    except:

        log.error('Mail sending failed')

        log.error(traceback.format_exc())

        return False

    return True

@task

def create_repo_fork(form_data, cur_user):

    import os

    from rhodecode.model.repo_model import RepoModel

    sa = get_session()

    rm = RepoModel(sa)

    rm.create(form_data, cur_user, just_db=True, fork=True)

    repos_path = get_hg_ui_settings()['paths_root_path'].replace('*', '')

    repo_path = os.path.join(repos_path, form_data['repo_name'])

    repo_fork_path = os.path.join(repos_path, form_data['fork_name'])

    MercurialRepository(str(repo_fork_path), True, clone_url=str(repo_path))

def __get_codes_stats(repo_name):

    LANGUAGES_EXTENSIONS = ['action', 'adp', 'ashx', 'asmx', 'aspx', 'asx', 'axd', 'c',

                    'cfg', 'cfm', 'cpp', 'cs', 'diff', 'do', 'el', 'erl',

                    'h', 'java', 'js', 'jsp', 'jspx', 'lisp',

                    'lua', 'm', 'mako', 'ml', 'pas', 'patch', 'php', 'php3',

                    'php4', 'phtml', 'pm', 'py', 'rb', 'rst', 's', 'sh',

                    'tpl', 'txt', 'vim', 'wss', 'xhtml', 'xml', 'xsl', 'xslt',

                    'yaws']

    repos_path = get_hg_ui_settings()['paths_root_path'].replace('*', '')

    repo = MercurialRepository(repos_path + repo_name)

    code_stats = {}

        for f in files:

            k = f.mimetype

            if f.extension in LANGUAGES_EXTENSIONS:

                if code_stats.has_key(k):

                    code_stats[k] += 1

                else:

                    code_stats[k] = 1

    return code_stats or {}

#!/usr/bin/env python

# encoding: utf-8

# whoosh indexer daemon for rhodecode

# Copyright (C) 2009-2010 Marcin Kuzminski <marcin@python-works.com>

#

# 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; version 2

# of the License or (at your opinion) any later version of the license.

# 

# 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, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,

# MA  02110-1301, USA.

"""

Created on Jan 26, 2010

@author: marcink

A deamon will read from task table and run tasks

"""

import sys

import os

from os.path import dirname as dn

from os.path import join as jn

#to get the rhodecode import

project_path = dn(dn(dn(dn(os.path.realpath(__file__)))))

sys.path.append(project_path)

from rhodecode.lib.pidlock import LockHeld, DaemonLock

from rhodecode.model.hg_model import HgModel

from rhodecode.lib.helpers import safe_unicode

from whoosh.index import create_in, open_dir

from shutil import rmtree

from rhodecode.lib.indexers import INDEX_EXTENSIONS, IDX_LOCATION, SCHEMA, IDX_NAME

from time import mktime

from vcs.exceptions import ChangesetError, RepositoryError

import logging

log = logging.getLogger('whooshIndexer')

# create logger

log.setLevel(logging.DEBUG)

log.propagate = False

# create console handler and set level to debug

ch = logging.StreamHandler()

ch.setLevel(logging.DEBUG)

# create formatter

formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")

# add formatter to ch

ch.setFormatter(formatter)

# add ch to logger

log.addHandler(ch)

def scan_paths(root_location):

    return HgModel.repo_scan('/', root_location, None, True)

class WhooshIndexingDaemon(object):

    """

    Deamon for atomic jobs

    """

    def __init__(self, indexname='HG_INDEX', repo_location=None):

        self.indexname = indexname

        self.repo_location = repo_location

        self.repo_paths = scan_paths(self.repo_location)

        self.initial = False

        if not os.path.isdir(IDX_LOCATION):

            os.mkdir(IDX_LOCATION)

            log.info('Cannot run incremental index since it does not'

                     ' yet exist running full build')

            self.initial = True

    def get_paths(self, repo):

        """

        recursive walk in root dir and return a set of all path in that dir

        based on repository walk function

        """

        index_paths_ = set()

        try:

                for f in files:

                    index_paths_.add(jn(repo.path, f.path))

                for dir in dirs:

                    for f in files:

                        index_paths_.add(jn(repo.path, f.path))

        except RepositoryError:

            pass

        return index_paths_        

    def get_node(self, repo, path):

        n_path = path[len(repo.path) + 1:]

        node = repo.get_changeset().get_node(n_path)

        return node

    def get_node_mtime(self, node):

        return mktime(node.last_changeset.date.timetuple())

    def add_doc(self, writer, path, repo):

        """Adding doc to writer"""

        node = self.get_node(repo, path)

        #we just index the content of chosen files

        if node.extension in INDEX_EXTENSIONS:

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

            u_content = node.content

        else:

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

            #just index file name without it's content

            u_content = u''

        writer.add_document(owner=unicode(repo.contact),

                        repository=safe_unicode(repo.name),

                        path=safe_unicode(path),

                        content=u_content,

                        modtime=self.get_node_mtime(node),

                        extension=node.extension)             

    def build_index(self):

        if os.path.exists(IDX_LOCATION):

            log.debug('removing previous index')

            rmtree(IDX_LOCATION)

        if not os.path.exists(IDX_LOCATION):

            os.mkdir(IDX_LOCATION)

        idx = create_in(IDX_LOCATION, SCHEMA, indexname=IDX_NAME)

        writer = idx.writer()

        for cnt, repo in enumerate(self.repo_paths.values()):

            log.debug('building index @ %s' % repo.path)

            for idx_path in self.get_paths(repo):

                self.add_doc(writer, idx_path, repo)

        log.debug('>> COMMITING CHANGES <<')

        writer.commit(merge=True)

        log.debug('>>> FINISHED BUILDING INDEX <<<')

    def update_index(self):

        log.debug('STARTING INCREMENTAL INDEXING UPDATE')

        idx = open_dir(IDX_LOCATION, indexname=self.indexname)

        # The set of all paths in the index

        indexed_paths = set()

        # The set of all paths we need to re-index

        to_index = set()

        reader = idx.reader()

        writer = idx.writer()

        # Loop over the stored fields in the index

        for fields in reader.all_stored_fields():

            indexed_path = fields['path']

            indexed_paths.add(indexed_path)

            repo = self.repo_paths[fields['repository']]

            try:

                node = self.get_node(repo, indexed_path)

            except ChangesetError:

                # This file was deleted since it was indexed

                log.debug('removing from index %s' % indexed_path)

                writer.delete_by_term('path', indexed_path)

            else:

                # Check if this file was changed since it was indexed

                indexed_time = fields['modtime']

                mtime = self.get_node_mtime(node)

                if mtime > indexed_time:

                    # The file has changed, delete it and add it to the list of

                    # files to reindex

                    log.debug('adding to reindex list %s' % indexed_path)

                    writer.delete_by_term('path', indexed_path)

                    to_index.add(indexed_path)

        # Loop over the files in the filesystem

        # Assume we have a function that gathers the filenames of the

        # documents to be indexed

        for repo in self.repo_paths.values():

            for path in self.get_paths(repo):

                if path in to_index or path not in indexed_paths:

                    # This is either a file that's changed, or a new file

                    # that wasn't indexed before. So index it!

                    self.add_doc(writer, path, repo)

                    log.debug('re indexing %s' % path)

        log.debug('>> COMMITING CHANGES <<')

        writer.commit(merge=True)

        log.debug('>>> FINISHED REBUILDING INDEX <<<')

    def run(self, full_index=False):

        """Run daemon"""

        if full_index or self.initial:

            self.build_index()

        else:

            self.update_index()

if __name__ == "__main__":

    arg = sys.argv[1:]

    if len(arg) != 2:

        sys.stderr.write('Please specify indexing type [full|incremental]' 

                         'and path to repositories as script args \n')

        sys.exit()

    if arg[0] == 'full':

        full_index = True

    elif arg[0] == 'incremental':

        # False means looking just for changes

        full_index = False

    else:

        sys.stdout.write('Please use [full|incremental]' 

                         ' as script first arg \n')

        sys.exit()

    if not os.path.isdir(arg[1]):

        sys.stderr.write('%s is not a valid path \n' % arg[1])

        sys.exit()

    else:

        if arg[1].endswith('/'):

            repo_location = arg[1] + '*'

        else:

            repo_location = arg[1] + '/*'

    try:

        l = DaemonLock()

        WhooshIndexingDaemon(repo_location=repo_location)\

            .run(full_index=full_index)

        l.release()

        reload(logging)

    except LockHeld:

        sys.exit(1)