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

log = logging.getLogger(__name__)

# CUSTOM ANALYZER wordsplit + lowercase filter

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

#INDEX SCHEMA DEFINITION

SCHEMA = Schema(

    fileid=ID(unique=True),

    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)

CHGSETS_SCHEMA = Schema(

    raw_id=ID(unique=True, stored=True),

    date=NUMERIC(stored=True),

    last=BOOLEAN(),

    owner=TEXT(),

    repository=ID(unique=True, stored=True),

    author=TEXT(stored=True),

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

                      scorable=True, stored=True),

    parents=TEXT(),

    added=TEXT(),

    removed=TEXT(),

    changed=TEXT(),

)

CHGSET_IDX_NAME = 'CHGSET_INDEX'

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',

        while self.matcher.is_active():

            docnum = self.matcher.id()

            chunks = [offsets for offsets in self.get_chunks()]

            docs_id.append([docnum, chunks])

            self.matcher.next()

        return docs_id

    def __str__(self):

        return '<%s at %s>' % (self.__class__.__name__, len(self.doc_ids))

    def __repr__(self):

        return self.__str__()

    def __len__(self):

        return len(self.doc_ids)

    def __iter__(self):

        """

        Allows Iteration over results,and lazy generate content

        *Requires* implementation of ``__getitem__`` method.

        """

        for docid in self.doc_ids:

            yield self.get_full_content(docid)

    def __getitem__(self, key):

        """

        Slicing of resultWrapper

        """

        i, j = key.start, key.stop

        slices = []

        for docid in self.doc_ids[i:j]:

            slices.append(self.get_full_content(docid))

        return slices

    def get_full_content(self, docid):

        res = self.searcher.stored_fields(docid[0])

        log.debug('result: %s' % res)

        if self.search_type == 'content':

            full_repo_path = jn(self.repo_location, res['repository'])

            f_path = res['path'].split(full_repo_path)[-1]

            f_path = f_path.lstrip(os.sep)

            content_short = self.get_short_content(res, docid[1])

            res.update({'content_short': content_short,

                        'content_short_hl': self.highlight(content_short),

                        'f_path': f_path

                      })

        elif self.search_type == 'message':

            res.update({'message_hl': self.highlight(res['message'])})

        log.debug('result: %s' % res)

        return res

    def get_short_content(self, res, chunks):

        return ''.join([res['content'][chunk[0]:chunk[1]] for chunk in chunks])

    def get_chunks(self):

        """

        Smart function that implements chunking the content

        but not overlap chunks so it doesn't highlight the same

        close occurrences twice.

        :param matcher:

        :param size:

        """

        memory = [(0, 0)]

        if self.matcher.supports('positions'):

            for span in self.matcher.spans():

                start = span.startchar or 0

                end = span.endchar or 0

                start_offseted = max(0, start - self.fragment_size)

                end_offseted = end + self.fragment_size

                if start_offseted < memory[-1][1]:

                    start_offseted = memory[-1][1]

                memory.append((start_offseted, end_offseted,))

                yield (start_offseted, end_offseted,)

    def highlight(self, content, top=5):

        if self.search_type not in ['content', 'message']:

            return ''

        hl = highlight(

            text=content,

            terms=self.highlight_items,

            analyzer=ANALYZER,

            fragmenter=FRAGMENTER,

            formatter=FORMATTER,

            top=top

        )

        return hl

    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',

                                    search_repo=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('a00c1b6f5d7a6ae678fd553a8b81d92367f7ecf1')

    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')