from pidlock import LockHeld, DaemonLock

import traceback

from pylons_app.config.environment import load_environment

from pylons_app.model.hg_model import HgModel

from pylons_app.lib.helpers import safe_unicode

from whoosh.index import create_in, open_dir

from shutil import rmtree

from pylons_app.lib.indexers import ANALYZER, INDEX_EXTENSIONS, IDX_LOCATION, \

SCHEMA, IDX_NAME

import logging

import logging.config

logging.config.fileConfig(jn(project_path, 'development.ini'))

log = logging.getLogger('whooshIndexer')

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

    def get_paths(self, root_dir):

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

        excluding files in .hg dir"""

        index_paths_ = set()

        for path, dirs, files in os.walk(root_dir):

            if path.find('.hg') == -1:

                for f in files:

                    index_paths_.add(jn(path, f))

        return index_paths_

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

        """Adding doc to writer"""

        ext = unicode(path.split('/')[-1].split('.')[-1].lower())

        #we just index the content of choosen files

        if ext in INDEX_EXTENSIONS:

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

            fobj = open(path, 'rb')

            content = fobj.read()

            fobj.close()

            u_content = safe_unicode(content)

        else:

                    # 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)

                    #writer.commit()

        # 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 scan_paths(self.repo_location).values():

            for path in self.get_paths(repo.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!

                    self.add_doc(writer, path, repo)

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

        writer.commit(merge=True)

        #idx.optimize()

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

    def run(self, full_index=False):

        """Run daemon"""

            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])