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

# 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/>.

"""

kallithea.config.conf

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

Various config settings for Kallithea

This file was forked by the Kallithea project in July 2014.

Original author and date, and relevant copyright and licensing information is below:

:created_on: Mar 7, 2012

:author: marcink

:copyright: (c) 2013 RhodeCode GmbH, and others.

:license: GPLv3, see LICENSE.md for more details.

"""

from kallithea.lib import pygmentsutils

# language map is also used by whoosh indexer, which for those specified

# extensions will index it's content

LANGUAGES_EXTENSIONS_MAP = pygmentsutils.get_extension_descriptions()

# Whoosh index targets

# Extensions we want to index content of using whoosh

# Filenames we want to index content of using whoosh

INDEX_FILENAMES = pygmentsutils.get_index_filenames()

# list of readme files to search in file tree and display in summary

# attached weights defines the search  order lower is first

ALL_READMES = [

    ('readme', 0), ('README', 0), ('Readme', 0),

    ('doc/readme', 1), ('doc/README', 1), ('doc/Readme', 1),

    ('Docs/readme', 2), ('Docs/README', 2), ('Docs/Readme', 2),

    ('DOCS/readme', 2), ('DOCS/README', 2), ('DOCS/Readme', 2),

    ('docs/readme', 2), ('docs/README', 2), ('docs/Readme', 2),

]

# extension together with weights to search lower is first

RST_EXTS = [

    ('', 0), ('.rst', 1), ('.rest', 1),

    ('.RST', 2), ('.REST', 2),

    ('.txt', 3), ('.TXT', 3)

]

MARKDOWN_EXTS = [

    ('.md', 1), ('.MD', 1),

    ('.mkdn', 2), ('.MKDN', 2),

    ('.mdown', 3), ('.MDOWN', 3),

    ('.markdown', 4), ('.MARKDOWN', 4)

]

PLAIN_EXTS = [('.text', 2), ('.TEXT', 2)]

ALL_EXTS = MARKDOWN_EXTS + RST_EXTS + PLAIN_EXTS

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    def index(self, repo_name, revision, f_path, annotate=False):

        # redirect to given revision from form if given

        post_revision = request.POST.get('at_rev', None)

        if post_revision:

            cs = self.__get_cs(post_revision) # FIXME - unused!

        c.revision = revision

        c.changeset = self.__get_cs(revision)

        c.branch = request.GET.get('branch', None)

        c.f_path = f_path

        c.annotate = annotate

        cur_rev = c.changeset.revision

        # used in files_source.html:

        c.cut_off_limit = self.cut_off_limit

        c.fulldiff = request.GET.get('fulldiff')

        # prev link

        try:

            prev_rev = c.db_repo_scm_instance.get_changeset(cur_rev).prev(c.branch)

            c.url_prev = url('files_home', repo_name=c.repo_name,

                         revision=prev_rev.raw_id, f_path=f_path)

            if c.branch:

                c.url_prev += '?branch=%s' % c.branch

        except (ChangesetDoesNotExistError, VCSError):

            c.url_prev = '#'

        # next link

        try:

            next_rev = c.db_repo_scm_instance.get_changeset(cur_rev).next(c.branch)

            c.url_next = url('files_home', repo_name=c.repo_name,

                     revision=next_rev.raw_id, f_path=f_path)

            if c.branch:

                c.url_next += '?branch=%s' % c.branch

        except (ChangesetDoesNotExistError, VCSError):

            c.url_next = '#'

        # files or dirs

        try:

            c.file = c.changeset.get_node(f_path)

            if c.file.is_submodule():

                raise HTTPFound(location=c.file.url)

            elif c.file.is_file():

                c.load_full_history = False

                # determine if we're on branch head

                _branches = c.db_repo_scm_instance.branches

                _hist = []

                c.file_history = []

                if c.load_full_history:

                    c.file_history, _hist = self._get_node_history(c.changeset, f_path)

                c.authors = []

                for a in set([x.author for x in _hist]):

                    c.authors.append((h.email(a), h.person(a)))

            else:

                c.authors = c.file_history = []

        except RepositoryError as e:

            h.flash(safe_str(e), category='error')

            raise HTTPNotFound()

        if request.environ.get('HTTP_X_PARTIAL_XHR'):

            return render('files/files_ypjax.html')

        # TODO: tags and bookmarks?

        c.revision_options = [(c.changeset.raw_id,

                              _('%s at %s') % (b, h.short_id(c.changeset.raw_id))) for b in c.changeset.branches] + \

            [(n, b) for b, n in c.db_repo_scm_instance.branches.items()]

        if c.db_repo_scm_instance.closed_branches:

            prefix = _('(closed)') + ' '

            c.revision_options += [('-', '-')] + \

                [(n, prefix + b) for b, n in c.db_repo_scm_instance.closed_branches.items()]

        return render('files/files.html')

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    @jsonify

    def history(self, repo_name, revision, f_path):

        changeset = self.__get_cs(revision)

        _file = changeset.get_node(f_path)

        if _file.is_file():

            file_history, _hist = self._get_node_history(changeset, f_path)

            res = []

            for obj in file_history:

                res.append({

                    'text': obj[1],

                    'children': [{'id': o[0], 'text': o[1]} for o in obj[0]]

                })

            data = {

                'more': False,

                'results': res

            }

        raw_mimetype_mapping = {

            # map original mimetype to a mimetype used for "show as raw"

            # you can also provide a content-disposition to override the

            # default "attachment" disposition.

            # orig_type: (new_type, new_dispo)

            # show images inline:

            'image/x-icon': ('image/x-icon', 'inline'),

            'image/png': ('image/png', 'inline'),

            'image/gif': ('image/gif', 'inline'),

            'image/jpeg': ('image/jpeg', 'inline'),

            'image/svg+xml': ('image/svg+xml', 'inline'),

        }

        mimetype = file_node.mimetype

        try:

            mimetype, dispo = raw_mimetype_mapping[mimetype]

        except KeyError:

            # we don't know anything special about this, handle it safely

            if file_node.is_binary:

                # do same as download raw for binary files

                mimetype, dispo = 'application/octet-stream', 'attachment'

            else:

                # do not just use the original mimetype, but force text/plain,

                # otherwise it would serve text/html and that might be unsafe.

                # Note: underlying vcs library fakes text/plain mimetype if the

                # mimetype can not be determined and it thinks it is not

                # binary.This might lead to erroneous text display in some

                # cases, but helps in other cases, like with text files

                # without extension.

                mimetype, dispo = 'text/plain', 'inline'

        if dispo == 'attachment':

            dispo = 'attachment; filename=%s' % \

                        safe_str(f_path.split(os.sep)[-1])

        response.content_disposition = dispo

        response.content_type = mimetype

        return file_node.content

    @LoginRequired()

    @HasRepoPermissionLevelDecorator('write')

    def delete(self, repo_name, revision, f_path):

        repo = c.db_repo

        # check if revision is a branch identifier- basically we cannot

        # create multiple heads via file editing

        _branches = repo.scm_instance.branches

        # check if revision is a branch name or branch hash

            h.flash(_('You can only delete files with revision '

                      'being a valid branch'), category='warning')

            raise HTTPFound(location=h.url('files_home',

                                  repo_name=repo_name, revision='tip',

                                  f_path=f_path))

        r_post = request.POST

        c.cs = self.__get_cs(revision)

        c.file = self.__get_filenode(c.cs, f_path)

        c.default_message = _('Deleted file %s via Kallithea') % (f_path)

        c.f_path = f_path

        node_path = f_path

        author = request.authuser.full_contact

        if r_post:

            message = r_post.get('message') or c.default_message

            try:

                nodes = {

                    node_path: {

                        'content': ''

                    }

                }

                self.scm_model.delete_nodes(

                    user=request.authuser.user_id,

                    ip_addr=request.ip_addr,

                    repo=c.db_repo,

                    message=message,

                    nodes=nodes,

                    parent_cs=c.cs,

                    author=author,

                )

                h.flash(_('Successfully deleted file %s') % f_path,

                        category='success')

            except Exception:

                log.error(traceback.format_exc())

                h.flash(_('Error occurred during commit'), category='error')

            raise HTTPFound(location=url('changeset_home',

                                repo_name=c.repo_name, revision='tip'))

        return render('files/files_delete.html')

    @LoginRequired()

    @HasRepoPermissionLevelDecorator('write')

    def edit(self, repo_name, revision, f_path):

        repo = c.db_repo

        # check if revision is a branch identifier- basically we cannot

        # create multiple heads via file editing

        _branches = repo.scm_instance.branches

        # check if revision is a branch name or branch hash

            h.flash(_('You can only edit files with revision '

                      'being a valid branch'), category='warning')

            raise HTTPFound(location=h.url('files_home',

                                  repo_name=repo_name, revision='tip',

                                  f_path=f_path))

        r_post = request.POST

        c.cs = self.__get_cs(revision)

        c.file = self.__get_filenode(c.cs, f_path)

        if c.file.is_binary:

            raise HTTPFound(location=url('files_home', repo_name=c.repo_name,

                            revision=c.cs.raw_id, f_path=f_path))

        c.default_message = _('Edited file %s via Kallithea') % (f_path)

        c.f_path = f_path

        if r_post:

            old_content = c.file.content

            sl = old_content.splitlines(1)

            first_line = sl[0] if sl else ''

            # modes:  0 - Unix, 1 - Mac, 2 - DOS

            mode = detect_mode(first_line, 0)

            content = convert_line_endings(r_post.get('content', ''), mode)

            message = r_post.get('message') or c.default_message

            author = request.authuser.full_contact

            if content == old_content:

                h.flash(_('No changes'), category='warning')

                raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,

                                    revision='tip'))

            try:

                self.scm_model.commit_change(repo=c.db_repo_scm_instance,

                                             repo_name=repo_name, cs=c.cs,

                                             user=request.authuser.user_id,

                                             ip_addr=request.ip_addr,

                                             author=author, message=message,

                                             content=content, f_path=f_path)

                h.flash(_('Successfully committed to %s') % f_path,

                        category='success')

            except Exception:

                log.error(traceback.format_exc())

                h.flash(_('Error occurred during commit'), category='error')

            raise HTTPFound(location=url('changeset_home',

                                repo_name=c.repo_name, revision='tip'))

            branchrev = repo.closed_branches.get(branch)

            if branchrev:

                n = 'branch:%s:%s' % (branch, branchrev)

                branches.append((n, _('%s (closed)') % branch))

                selected = n

                branch = None

            if branch:

                log.debug('branch %r not found in %s', branch, repo)

        bookmarks = []

        for bookmark, bookmarkrev in repo.bookmarks.iteritems():

            n = 'book:%s:%s' % (bookmark, bookmarkrev)

            bookmarks.append((n, bookmark))

            if rev == bookmarkrev:

                selected = n

        tags = []

        for tag, tagrev in repo.tags.iteritems():

            if tag == 'tip':

                continue

            n = 'tag:%s:%s' % (tag, tagrev)

            tags.append((n, tag))

            # note: even if rev == tagrev, don't select the static tag - it must be chosen explicitly

        # prio 1: rev was selected as existing entry above

        # prio 2: create special entry for rev; rev _must_ be used

        specials = []

        if rev and selected is None:

            selected = 'rev:%s:%s' % (rev, rev)

            specials = [(selected, '%s: %s' % (_("Changeset"), rev[:12]))]

        # prio 3: most recent peer branch

        if peers and not selected:

            selected = peers[0][0]

        # prio 4: tip revision

        if not selected:

            if h.is_hg(repo):

                if tipbranch:

                    selected = 'branch:%s:%s' % (tipbranch, tiprev)

                else:

                    selected = 'tag:null:' + repo.EMPTY_CHANGESET

                    tags.append((selected, 'null'))

            else:

                if 'master' in repo.branches:

                    selected = 'branch:master:%s' % repo.branches['master']

                else:

                    selected = 'branch:%s:%s' % (k, v)

        groups = [(specials, _("Special")),

                  (peers, _("Peer branches")),

                  (bookmarks, _("Bookmarks")),

                  (branches, _("Branches")),

                  (tags, _("Tags")),

                  ]

        return [g for g in groups if g[0]], selected

    def _is_allowed_to_change_status(self, pull_request):

        if pull_request.is_closed():

            return False

        owner = request.authuser.user_id == pull_request.owner_id

        reviewer = PullRequestReviewer.query() \

            .filter(PullRequestReviewer.pull_request == pull_request) \

            .filter(PullRequestReviewer.user_id == request.authuser.user_id) \

            .count() != 0

        return request.authuser.admin or owner or reviewer

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    def show_all(self, repo_name):

        c.from_ = request.GET.get('from_') or ''

        c.closed = request.GET.get('closed') or ''

        url_params = {}

        if c.from_:

            url_params['from_'] = 1

        if c.closed:

            url_params['closed'] = 1

        p = safe_int(request.GET.get('page'), 1)

        q = PullRequest.query(include_closed=c.closed, sorted=True)

        if c.from_:

            q = q.filter_by(org_repo=c.db_repo)

        else:

            q = q.filter_by(other_repo=c.db_repo)

        c.pull_requests = q.all()

        c.pullrequests_pager = Page(c.pull_requests, page=p, items_per_page=100, **url_params)

        return render('/pullrequests/pullrequest_show_all.html')

    @LoginRequired()

    def show_my(self):

        c.closed = request.GET.get('closed') or ''

                log.error('CSRF check failed')

                raise webob.exc.HTTPForbidden()

        c.kallithea_version = __version__

        rc_config = Setting.get_app_settings()

        # Visual options

        c.visual = AttributeDict({})

        ## DB stored

        c.visual.show_public_icon = str2bool(rc_config.get('show_public_icon'))

        c.visual.show_private_icon = str2bool(rc_config.get('show_private_icon'))

        c.visual.stylify_metalabels = str2bool(rc_config.get('stylify_metalabels'))

        c.visual.page_size = safe_int(rc_config.get('dashboard_items', 100))

        c.visual.admin_grid_items = safe_int(rc_config.get('admin_grid_items', 100))

        c.visual.repository_fields = str2bool(rc_config.get('repository_fields'))

        c.visual.show_version = str2bool(rc_config.get('show_version'))

        c.visual.use_gravatar = str2bool(rc_config.get('use_gravatar'))

        c.visual.gravatar_url = rc_config.get('gravatar_url')

        c.ga_code = rc_config.get('ga_code')

        # TODO: replace undocumented backwards compatibility hack with db upgrade and rename ga_code

        if c.ga_code and '<' not in c.ga_code:

            c.ga_code = '''<script type="text/javascript">

                var _gaq = _gaq || [];

                _gaq.push(['_setAccount', '%s']);

                _gaq.push(['_trackPageview']);

                (function() {

                    var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;

                    ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';

                    var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);

                    })();

            </script>''' % c.ga_code

        c.site_name = rc_config.get('title')

        c.clone_uri_tmpl = rc_config.get('clone_uri_tmpl') or Repository.DEFAULT_CLONE_URI

        c.clone_ssh_tmpl = rc_config.get('clone_ssh_tmpl') or Repository.DEFAULT_CLONE_SSH

        ## INI stored

        c.visual.allow_repo_location_change = str2bool(config.get('allow_repo_location_change', True))

        c.visual.allow_custom_hooks_settings = str2bool(config.get('allow_custom_hooks_settings', True))

        c.ssh_enabled = str2bool(config.get('ssh_enabled', False))

        c.instance_id = config.get('instance_id')

        c.issues_url = config.get('bugtracker', url('issues_url'))

        # END CONFIG VARS

        c.repo_name = get_repo_slug(request)  # can be empty

        self.cut_off_limit = safe_int(config.get('cut_off_limit'))

        c.my_pr_count = PullRequest.query(reviewer_id=request.authuser.user_id, include_closed=False).count()

        self.scm_model = ScmModel()

    @staticmethod

    def _determine_auth_user(session_authuser, ip_addr):

        """

        Create an `AuthUser` object given the API key/bearer token

        (if any) and the value of the authuser session cookie.

        Returns None if no valid user is found (like not active or no access for IP).

        """

        # Authenticate by session cookie

        # In ancient login sessions, 'authuser' may not be a dict.

        # In that case, the user will have to log in again.

        # v0.3 and earlier included an 'is_authenticated' key; if present,

        # this must be True.

        if isinstance(session_authuser, dict) and session_authuser.get('is_authenticated', True):

            return AuthUser.from_cookie(session_authuser, ip_addr=ip_addr)

        # Authenticate by auth_container plugin (if enabled)

        if any(

            plugin.is_container_auth

            for plugin in auth_modules.get_auth_plugins()

        ):

            try:

                user_info = auth_modules.authenticate('', '', request.environ)

            except UserCreationError as e:

                from kallithea.lib import helpers as h

                h.flash(e, 'error', logf=log.error)

            else:

                if user_info is not None:

                    username = user_info['username']

                    user = User.get_by_username(username, case_insensitive=True)

                    return log_in_user(user, remember=False, is_external_auth=True, ip_addr=ip_addr)

        # User is default user (if active) or anonymous

        default_user = User.get_default_user(cache=True)

        authuser = AuthUser.make(dbuser=default_user, ip_addr=ip_addr)

        if authuser is None: # fall back to anonymous

            authuser = AuthUser(dbuser=default_user) # TODO: somehow use .make?

        return authuser

    @staticmethod

    def _basic_security_checks():

            copy_fork_permissions=copy_fork_permissions

        )

        action_logger(cur_user, 'user_forked_repo:%s' % repo_name_full,

                      fork_of.repo_name, '')

        DBS.commit()

        source_repo_path = os.path.join(base_path, fork_of.repo_name)

        # now create this repo on Filesystem

        RepoModel()._create_filesystem_repo(

            repo_name=repo_name,

            repo_type=repo_type,

            repo_group=RepoGroup.guess_instance(repo_group),

            clone_uri=source_repo_path,

        )

        repo = Repository.get_by_repo_name(repo_name_full)

        log_create_repository(repo.get_dict(), created_by=owner.username)

        # update repo changeset caches initially

        repo.update_changeset_cache()

        # set new created state

        repo.set_state(Repository.STATE_CREATED)

        DBS.commit()

    except Exception as e:

        log.warning('Exception %s occurred when forking repository, '

                    'doing cleanup...' % e)

        # rollback things manually !

        repo = Repository.get_by_repo_name(repo_name_full)

        if repo:

            Repository.delete(repo.repo_id)

            DBS.commit()

            RepoModel()._delete_filesystem_repo(repo)

        raise

    return True

def __get_codes_stats(repo_name):

    from kallithea.config.conf import LANGUAGES_EXTENSIONS_MAP

    repo = Repository.get_by_repo_name(repo_name).scm_instance

    tip = repo.get_changeset()

    code_stats = {}

    for _topnode, _dirnodes, filenodes in tip.walk('/'):

        for filenode in filenodes:

            ext = filenode.extension.lower()

                if ext in code_stats:

                    code_stats[ext] += 1

                else:

                    code_stats[ext] = 1

    return code_stats or {}

    s = html_escape(s)

    s = _urlify(s)

    if repo_name is not None:

        s = urlify_issues(s, repo_name)

    if link_ is not None:

        # make href around everything that isn't a href already

        s = linkify_others(s, link_)

    s = s.replace('\r\n', '<br/>').replace('\n', '<br/>')

    # Turn HTML5 into more valid HTML4 as required by some mail readers.

    # (This is not done in one step in html_escape, because character codes like

    # { risk to be seen as an issue reference due to the presence of '#'.)

    s = s.replace("'", "'")

    return literal(s)

def linkify_others(t, l):

    """Add a default link to html with links.

    HTML doesn't allow nesting of links, so the outer link must be broken up

    in pieces and give space for other links.

    """

    urls = re.compile(r'(\<a.*?\<\/a\>)',)

    links = []

    for e in urls.split(t):

        if e.strip() and not urls.match(e):

            links.append('<a class="message-link" href="%s">%s</a>' % (l, e))

        else:

            links.append(e)

    return ''.join(links)

# Global variable that will hold the actual urlify_issues function body.

# Will be set on first use when the global configuration has been read.

_urlify_issues_f = None

def urlify_issues(newtext, repo_name):

    """Urlify issue references according to .ini configuration"""

    global _urlify_issues_f

    if _urlify_issues_f is None:

        from kallithea import CONFIG

        from kallithea.model.db import URL_SEP

        assert CONFIG['sqlalchemy.url'] # make sure config has been loaded

        # Build chain of urlify functions, starting with not doing any transformation

        tmp_urlify_issues_f = lambda s: s

        issue_pat_re = re.compile(r'issue_pat(.*)')

            # Find all issue_pat* settings that also have corresponding server_link and prefix configuration

            m = issue_pat_re.match(k)

            if m is None:

                continue

            suffix = m.group(1)

            issue_pat = CONFIG.get(k)

            issue_server_link = CONFIG.get('issue_server_link%s' % suffix)

            issue_sub = CONFIG.get('issue_sub%s' % suffix)

            if not issue_pat or not issue_server_link or issue_sub is None: # issue_sub can be empty but should be present

                log.error('skipping incomplete issue pattern %r: %r -> %r %r', suffix, issue_pat, issue_server_link, issue_sub)

                continue

            # Wrap tmp_urlify_issues_f with substitution of this pattern, while making sure all loop variables (and compiled regexpes) are bound

            try:

                issue_re = re.compile(issue_pat)

            except re.error as e:

                log.error('skipping invalid issue pattern %r: %r -> %r %r. Error: %s', suffix, issue_pat, issue_server_link, issue_sub, str(e))

                continue

            log.debug('issue pattern %r: %r -> %r %r', suffix, issue_pat, issue_server_link, issue_sub)

            def issues_replace(match_obj,

                               issue_server_link=issue_server_link, issue_sub=issue_sub):

                try:

                    issue_url = match_obj.expand(issue_server_link)

                except (IndexError, re.error) as e:

                    log.error('invalid issue_url setting %r -> %r %r. Error: %s', issue_pat, issue_server_link, issue_sub, str(e))

                    issue_url = issue_server_link

                issue_url = issue_url.replace('{repo}', repo_name)

                issue_url = issue_url.replace('{repo_name}', repo_name.split(URL_SEP)[-1])

                # if issue_sub is empty use the matched issue reference verbatim

                if not issue_sub:

                    issue_text = match_obj.group()

                else:

                    try:

                        issue_text = match_obj.expand(issue_sub)

                    except (IndexError, re.error) as e:

                        log.error('invalid issue_sub setting %r -> %r %r. Error: %s', issue_pat, issue_server_link, issue_sub, str(e))

                        issue_text = match_obj.group()

                return (

                    '<a class="issue-tracker-link" href="%(url)s">'

                    '%(text)s'

                    '</a>'

                    ) % {

                     'url': issue_url,

                     'text': issue_text,

                    }

        Returns history of file as reversed list of ``Changeset`` objects for

        which file at given ``path`` has been modified.

        """

        self._get_filectx(path)

        from dulwich.walk import Walker

        include = [self.id]

        walker = Walker(self.repository._repo.object_store, include,

                        paths=[path], max_entries=1)

        return [self.repository.get_changeset(sha)

                for sha in (x.commit.id for x in walker)]

    def get_file_annotate(self, path):

        """

        Returns a generator of four element tuples with

            lineno, sha, changeset lazy loader and line

        TODO: This function now uses os underlying 'git' command which is

        generally not good. Should be replaced with algorithm iterating

        commits.

        """

        cmd = ['blame', '-l', '--root', '-r', self.id, '--', path]

        # -l     ==> outputs long shas (and we need all 40 characters)

        # --root ==> doesn't put '^' character for boundaries

        # -r sha ==> blames for the given revision

        so, se = self.repository.run_git_command(cmd)

        for i, blame_line in enumerate(so.split('\n')[:-1]):

            ln_no = i + 1

            sha, line = re.split(r' ', blame_line, 1)

            yield (ln_no, sha, lambda: self.repository.get_changeset(sha), line)

    def fill_archive(self, stream=None, kind='tgz', prefix=None,

                     subrepos=False):

        """

        Fills up given stream.

        :param stream: file like object.

        :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``.

            Default: ``tgz``.

        :param prefix: name of root directory in archive.

            Default is repository name and changeset's raw_id joined with dash

            (``repo-tip.<KIND>``).

        :param subrepos: include subrepos in this archive.

        :raise ImproperArchiveTypeError: If given kind is wrong.

        :raise VcsError: If given stream is None

        """

        if kind not in allowed_kinds:

            raise ImproperArchiveTypeError('Archive kind not supported use one'

        if stream is None:

            raise VCSError('You need to pass in a valid stream for filling'

                           ' with archival data')

        if prefix is None:

            prefix = '%s-%s' % (self.repository.name, self.short_id)

        elif prefix.startswith('/'):

            raise VCSError("Prefix cannot start with leading slash")

        elif prefix.strip() == '':

            raise VCSError("Prefix cannot be empty")

        if kind == 'zip':

            frmt = 'zip'

        else:

            frmt = 'tar'

        _git_path = settings.GIT_EXECUTABLE_PATH

        cmd = '%s archive --format=%s --prefix=%s/ %s' % (_git_path,

                                                frmt, prefix, self.raw_id)

        if kind == 'tgz':

            cmd += ' | gzip -9'

        elif kind == 'tbz2':

            cmd += ' | bzip2 -9'

        if stream is None:

            raise VCSError('You need to pass in a valid stream for filling'

                           ' with archival data')

        popen = Popen(cmd, stdout=PIPE, stderr=PIPE, shell=True,

                      cwd=self.repository.path)

        buffer_size = 1024 * 8

        chunk = popen.stdout.read(buffer_size)

        while chunk:

            stream.write(chunk)

            chunk = popen.stdout.read(buffer_size)

        # Make sure all descriptors would be read

        popen.communicate()

    def get_nodes(self, path):

        """

        Returns combined ``DirNode`` and ``FileNode`` objects list representing

        state of changeset at the given ``path``. If node at the given ``path``

        is not instance of ``DirNode``, ChangesetError would be raised.

        """

        if self._get_kind(path) != NodeKind.DIR:

            raise ChangesetError("Directory does not exist for revision %s at "

                " '%s'" % (self.revision, path))

        return self.get_file_history(path, limit=1)[0]

    def get_file_history(self, path, limit=None):

        """

        Returns history of file as reversed list of ``Changeset`` objects for

        which file at given ``path`` has been modified.

        """

        fctx = self._get_filectx(path)

        hist = []

        cnt = 0

        for cs in reversed([x for x in fctx.filelog()]):

            cnt += 1

            hist.append(hex(fctx.filectx(cs).node()))

            if limit is not None and cnt == limit:

                break

        return [self.repository.get_changeset(node) for node in hist]

    def get_file_annotate(self, path):

        """

        Returns a generator of four element tuples with

            lineno, sha, changeset lazy loader and line

        """

        annotations = self._get_filectx(path).annotate()

        try:

            annotation_lines = [(annotateline.fctx, annotateline.text) for annotateline in annotations]

        except AttributeError: # annotateline was introduced in Mercurial 4.6 (b33b91ca2ec2)

            annotation_lines = [(aline.fctx, l) for aline, l in annotations]

        for i, (fctx, l) in enumerate(annotation_lines):

            sha = fctx.hex()

            yield (i + 1, sha, lambda sha=sha, l=l: self.repository.get_changeset(sha), l)

    def fill_archive(self, stream=None, kind='tgz', prefix=None,

                     subrepos=False):

        """

        Fills up given stream.

        :param stream: file like object.

        :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``.

            Default: ``tgz``.

        :param prefix: name of root directory in archive.

            Default is repository name and changeset's raw_id joined with dash

            (``repo-tip.<KIND>``).

        :param subrepos: include subrepos in this archive.

        :raise ImproperArchiveTypeError: If given kind is wrong.

        :raise VcsError: If given stream is None

        """

        if kind not in allowed_kinds:

            raise ImproperArchiveTypeError('Archive kind not supported use one'

        if stream is None:

            raise VCSError('You need to pass in a valid stream for filling'

                           ' with archival data')

        if prefix is None:

            prefix = '%s-%s' % (self.repository.name, self.short_id)

        elif prefix.startswith('/'):