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

"""

    rhodecode.lib.indexers.__init__

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

    Whoosh indexing module for RhodeCode

    :created_on: Aug 17, 2010

    :author: marcink

    :copyright: (C) 2010-2012 Marcin Kuzminski <marcin@python-works.com>

    :license: GPLv3, see COPYING for more details.

"""

# 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/>.

import os

import sys

import traceback

import logging

from os.path import dirname as dn, join as jn

#to get the rhodecode import

sys.path.append(dn(dn(dn(os.path.realpath(__file__)))))

from string import strip

from shutil import rmtree

from whoosh.analysis import RegexTokenizer, LowercaseFilter, StopFilter

from whoosh.fields import TEXT, ID, STORED, Schema, FieldType

from whoosh.index import create_in, open_dir

from whoosh.formats import Characters

from whoosh.highlight import highlight, HtmlFormatter, ContextFragmenter

from webhelpers.html.builder import escape

from sqlalchemy import engine_from_config

from rhodecode.model import init_model

from rhodecode.model.scm import ScmModel

from rhodecode.model.repo import RepoModel

from rhodecode.config.environment import load_environment

from rhodecode.lib.utils2 import LazyProperty

from rhodecode.lib.utils import BasePasterCommand, Command, add_cache,\

    load_rcextensions

# CUSTOM ANALYZER wordsplit + lowercase filter

ANALYZER = RegexTokenizer(expression=r"\w+") | LowercaseFilter()

#INDEX SCHEMA DEFINITION

SCHEMA = Schema(

    owner=TEXT(),

    repository=TEXT(stored=True),

    path=TEXT(stored=True),

    content=FieldType(format=Characters(), analyzer=ANALYZER,

                      scorable=True, stored=True),

    modtime=STORED(),

    extension=TEXT(stored=True)

)

IDX_NAME = 'HG_INDEX'

FORMATTER = HtmlFormatter('span', between='\n<span class="break">...</span>\n')

FRAGMENTER = ContextFragmenter(200)

class MakeIndex(BasePasterCommand):

    max_args = 1

    min_args = 1

    usage = "CONFIG_FILE"

    summary = "Creates index for full text search given configuration file"

    group_name = "RhodeCode"

    takes_config_file = -1

    parser = Command.standard_parser(verbose=True)

    def command(self):

        logging.config.fileConfig(self.path_to_ini_file)

        from pylons import config

        add_cache(config)

        engine = engine_from_config(config, 'sqlalchemy.db1.')

        init_model(engine)

        index_location = config['index_dir']

        repo_location = self.options.repo_location \

            if self.options.repo_location else RepoModel().repos_path

        repo_list = map(strip, self.options.repo_list.split(',')) \

            if self.options.repo_list else None

        repo_update_list = map(strip, self.options.repo_update_list.split(',')) \

            if self.options.repo_update_list else None

        load_rcextensions(config['here'])

        #======================================================================

        # WHOOSH DAEMON

        #======================================================================

        from rhodecode.lib.pidlock import LockHeld, DaemonLock

        from rhodecode.lib.indexers.daemon import WhooshIndexingDaemon

        try:

            l = DaemonLock(file_=jn(dn(dn(index_location)), 'make_index.lock'))

            WhooshIndexingDaemon(index_location=index_location,

                                 repo_location=repo_location,

                                 repo_list=repo_list,

                                 repo_update_list=repo_update_list)\

                .run(full_index=self.options.full_index)

            l.release()

        except LockHeld:

            sys.exit(1)

    def update_parser(self):

        self.parser.add_option('--repo-location',

                          action='store',

                          dest='repo_location',

                          help="Specifies repositories location to index OPTIONAL",

                          )

        self.parser.add_option('--index-only',

                          action='store',

                          dest='repo_list',

                          help="Specifies a comma separated list of repositores "

                                "to build index on. If not given all repositories "

                                "are scanned for indexing. OPTIONAL",

                          )

        self.parser.add_option('--update-only',

                          action='store',

                          dest='repo_update_list',

                          help="Specifies a comma separated list of repositores "

                                "to re-build index on. OPTIONAL",

                          )

        self.parser.add_option('-f',

                          action='store_true',

                          dest='full_index',

                          help="Specifies that index should be made full i.e"

                                " destroy old and build from scratch",

                          default=False)

class WhooshResultWrapper(object):

    def __init__(self, search_type, searcher, matcher, highlight_items,

                 repo_location):

        self.search_type = search_type

        self.searcher = searcher

        self.matcher = matcher

        self.highlight_items = highlight_items

        self.fragment_size = 200

        self.repo_location = repo_location

    @LazyProperty

    def doc_ids(self):

        docs_id = []

        while self.matcher.is_active():

class WhooshIndexingDaemon(object):

    """

    Daemon for atomic indexing jobs

    """

    def __init__(self, indexname=IDX_NAME, index_location=None,

                 repo_location=None, sa=None, repo_list=None,

                 repo_update_list=None):

        self.indexname = indexname

        self.index_location = index_location

        if not index_location:

            raise Exception('You have to provide index location')

        self.repo_location = repo_location

        if not repo_location:

            raise Exception('You have to provide repositories location')

        self.repo_paths = ScmModel(sa).repo_scan(self.repo_location)

        #filter repo list

        if repo_list:

            self.filtered_repo_paths = {}

            for repo_name, repo in self.repo_paths.items():

                if repo_name in repo_list:

                    self.filtered_repo_paths[repo_name] = repo

            self.repo_paths = self.filtered_repo_paths

        #filter update repo list

        self.filtered_repo_update_paths = {}

        if repo_update_list:

            self.filtered_repo_update_paths = {}

            for repo_name, repo in self.repo_paths.items():

                if repo_name in repo_update_list:

                    self.filtered_repo_update_paths[repo_name] = repo

            self.repo_paths = self.filtered_repo_update_paths

        self.initial = False

        if not os.path.isdir(self.index_location):

            os.makedirs(self.index_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:

            tip = repo.get_changeset('tip')

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

                for f in files:

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

        except RepositoryError, e:

            log.debug(traceback.format_exc())

            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, repo_name):

        """

        Adding doc to writer this function itself fetches data from

        the instance of vcs backend

        """

        node = self.get_node(repo, path)

        indexed = indexed_w_content = 0

        # we just index the content of chosen files, and skip binary files

        if node.extension in INDEX_EXTENSIONS and not node.is_binary:

            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

        writer.add_document(

            owner=unicode(repo.contact),

            repository=safe_unicode(repo_name),

            content=u_content,

            modtime=self.get_node_mtime(node),

            extension=node.extension

        )

        return indexed, indexed_w_content

    def build_index(self):

        if os.path.exists(self.index_location):

            log.debug('removing previous index')

            rmtree(self.index_location)

        if not os.path.exists(self.index_location):

            os.mkdir(self.index_location)

        idx = create_in(self.index_location, SCHEMA, indexname=IDX_NAME)

        writer = idx.writer()

        log.debug('BUILDIN INDEX FOR EXTENSIONS %s' % INDEX_EXTENSIONS)

        for repo_name, repo in self.repo_paths.items():

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

            i_cnt = iwc_cnt = 0

            for idx_path in self.get_paths(repo):

                i, iwc = self.add_doc(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)

            )

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

        writer.commit(merge=True)

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

    def update_index(self):

        log.debug((u'STARTING INCREMENTAL INDEXING UPDATE FOR EXTENSIONS %s '

                   'AND REPOS %s') % (INDEX_EXTENSIONS, self.repo_paths.keys()))

        idx = open_dir(self.index_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_repo_path = fields['repository']

            indexed_paths.add(indexed_path)

            if not indexed_repo_path in self.filtered_repo_update_paths:

                continue

            repo = self.repo_paths[indexed_repo_path]

            try:

                node = self.get_node(repo, indexed_path)

                # 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

                    to_index.add(indexed_path)

            except (ChangesetError, NodeDoesNotExistError):

                # This file was deleted since it was indexed

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

                writer.delete_by_term('path', indexed_path)

        # Loop over the files in the filesystem

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

        # documents to be indexed

        ri_cnt = riwc_cnt = 0

        for repo_name, repo in self.repo_paths.items():

            for path in self.get_paths(repo):

                path = safe_unicode(path)

                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!

                    i, iwc = self.add_doc(writer, path, repo, repo_name)

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

                    ri_cnt += i

                    riwc_cnt += iwc

        log.debug('added %s files %s with content for repo %s' % (

                     ri_cnt + riwc_cnt, riwc_cnt, repo.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()