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)

        """

        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

        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()