# 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.controllers.compare

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

compare controller showing differences between two

repos, branches, bookmarks or tips

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: May 6, 2012

:author: marcink

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

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

"""

import logging

import re

from tg import request

from tg import tmpl_context as c

from tg.i18n import ugettext as _

from webob.exc import HTTPBadRequest, HTTPFound, HTTPNotFound

from kallithea.config.routing import url

from kallithea.controllers.changeset import _context_url, _ignorews_url

from kallithea.lib import diffs

from kallithea.lib import helpers as h

from kallithea.lib.auth import HasRepoPermissionLevelDecorator, LoginRequired

from kallithea.lib.base import BaseRepoController, render

from kallithea.lib.graphmod import graph_data

from kallithea.lib.utils2 import safe_int, safe_str

from kallithea.lib.vcs.utils.hgcompat import unionrepo

from kallithea.model.db import Repository

log = logging.getLogger(__name__)

class CompareController(BaseRepoController):

    def _before(self, *args, **kwargs):

        super(CompareController, self)._before(*args, **kwargs)

        # The base repository has already been retrieved.

        c.a_repo = c.db_repo

        # Retrieve the "changeset" repository (default: same as base).

        other_repo = request.GET.get('other_repo', None)

        if other_repo is None:

            c.cs_repo = c.a_repo

        else:

            c.cs_repo = Repository.get_by_repo_name(other_repo)

            if c.cs_repo is None:

                msg = _('Could not find other repository %s') % other_repo

                h.flash(msg, category='error')

                raise HTTPFound(location=url('compare_home', repo_name=c.a_repo.repo_name))

        # Verify that it's even possible to compare these two repositories.

        if c.a_repo.scm_instance.alias != c.cs_repo.scm_instance.alias:

            msg = _('Cannot compare repositories of different types')

            h.flash(msg, category='error')

            raise HTTPFound(location=url('compare_home', repo_name=c.a_repo.repo_name))

    @staticmethod

    def _get_changesets(alias, org_repo, org_rev, other_repo, other_rev):

        """

        Returns lists of changesets that can be merged from org_repo@org_rev

        to other_repo@other_rev

        ... and the other way

        ... and the ancestors that would be used for merge

        :param org_repo: repo object, that is most likely the original repo we forked from

        :param org_rev: the revision we want our compare to be made

        :param other_repo: repo object, most likely the fork of org_repo. It has

            all changesets that we need to obtain

        :param other_rev: revision we want out compare to be made on other_repo

        """

        ancestors = None

        if org_rev == other_rev:

            org_changesets = []

            other_changesets = []

        elif alias == 'hg':

            # case two independent repos

            if org_repo != other_repo:

                                                       other_repo.path,

                                                       org_repo.path)

                # all ancestors of other_rev will be in other_repo and

                # rev numbers from hgrepo can be used in other_repo - org_rev ancestors cannot

            # no remote compare do it on the same repository

            else:

                hgrepo = other_repo._repo

            ancestors = [hgrepo[ancestor].hex() for ancestor in

                         hgrepo.revs("id(%s) & ::id(%s)", other_rev, org_rev)]

            if ancestors:

                log.debug("shortcut found: %s is already an ancestor of %s", other_rev, org_rev)

            else:

                log.debug("no shortcut found: %s is not an ancestor of %s", other_rev, org_rev)

                ancestors = [hgrepo[ancestor].hex() for ancestor in

                             hgrepo.revs("heads(::id(%s) & ::id(%s))", org_rev, other_rev)] # FIXME: expensive!

            other_revs = hgrepo.revs("ancestors(id(%s)) and not ancestors(id(%s)) and not id(%s)",

                                     other_rev, org_rev, org_rev)

            other_changesets = [other_repo.get_changeset(rev) for rev in other_revs]

            org_revs = hgrepo.revs("ancestors(id(%s)) and not ancestors(id(%s)) and not id(%s)",

                                   org_rev, other_rev, other_rev)

            org_changesets = [org_repo.get_changeset(hgrepo[rev].hex()) for rev in org_revs]

        elif alias == 'git':

            if org_repo != other_repo:

                from dulwich.repo import Repo

                from dulwich.client import SubprocessGitClient

                gitrepo = Repo(org_repo.path)

                SubprocessGitClient(thin_packs=False).fetch(safe_str(other_repo.path), gitrepo)

                gitrepo_remote = Repo(other_repo.path)

                SubprocessGitClient(thin_packs=False).fetch(safe_str(org_repo.path), gitrepo_remote)

                revs = [

                    x.commit.id

                    for x in gitrepo_remote.get_walker(include=[other_rev],

                                                       exclude=[org_rev])

                ]

                other_changesets = [other_repo.get_changeset(rev) for rev in reversed(revs)]

                if other_changesets:

                    ancestors = [other_changesets[0].parents[0].raw_id]

                else:

                    # no changesets from other repo, ancestor is the other_rev

                    ancestors = [other_rev]

                gitrepo.close()

                gitrepo_remote.close()

            else:

                so, se = org_repo.run_git_command(

                    ['log', '--reverse', '--pretty=format:%H',

                     '-s', '%s..%s' % (org_rev, other_rev)]

                )

                other_changesets = [org_repo.get_changeset(cs)

                              for cs in re.findall(r'[0-9a-fA-F]{40}', so)]

                so, se = org_repo.run_git_command(

                    ['merge-base', org_rev, other_rev]

                )

                ancestors = [re.findall(r'[0-9a-fA-F]{40}', so)[0]]

            org_changesets = []

        else:

            raise Exception('Bad alias only git and hg is allowed')

        return other_changesets, org_changesets, ancestors

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    def index(self, repo_name):

        c.compare_home = True

        c.a_ref_name = c.cs_ref_name = None

        return render('compare/compare_diff.html')

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    def compare(self, repo_name, org_ref_type, org_ref_name, other_ref_type, other_ref_name):

        org_ref_name = org_ref_name.strip()

        other_ref_name = other_ref_name.strip()

        # If merge is True:

        #   Show what org would get if merged with other:

        #   List changesets that are ancestors of other but not of org.

        #   New changesets in org is thus ignored.

        #   Diff will be from common ancestor, and merges of org to other will thus be ignored.

        # If merge is False:

        #   Make a raw diff from org to other, no matter if related or not.

        #   Changesets in one and not in the other will be ignored

        merge = bool(request.GET.get('merge'))

        # fulldiff disables cut_off_limit

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

        # partial uses compare_cs.html template directly

        partial = request.environ.get('HTTP_X_PARTIAL_XHR')

        # is_ajax_preview puts hidden input field with changeset revisions

    def delete(self, repo_name, pull_request_id):

        pull_request = PullRequest.get_or_404(pull_request_id)

        # only owner can delete it !

        if pull_request.owner_id == request.authuser.user_id:

            PullRequestModel().delete(pull_request)

            Session().commit()

            h.flash(_('Successfully deleted pull request'),

                    category='success')

            raise HTTPFound(location=url('my_pullrequests'))

        raise HTTPForbidden()

    @LoginRequired(allow_default_user=True)

    @HasRepoPermissionLevelDecorator('read')

    def show(self, repo_name, pull_request_id, extra=None):

        c.pull_request = PullRequest.get_or_404(pull_request_id)

        c.allowed_to_change_status = self._is_allowed_to_change_status(c.pull_request)

        cc_model = ChangesetCommentsModel()

        cs_model = ChangesetStatusModel()

        # pull_requests repo_name we opened it against

        # ie. other_repo must match

        if repo_name != c.pull_request.other_repo.repo_name:

            raise HTTPNotFound

        # load compare data into template context

        c.cs_repo = c.pull_request.org_repo

        (c.cs_ref_type,

         c.cs_ref_name,

         c.cs_rev) = c.pull_request.org_ref.split(':')

        c.a_repo = c.pull_request.other_repo

        (c.a_ref_type,

         c.a_ref_name,

         c.a_rev) = c.pull_request.other_ref.split(':') # a_rev is ancestor

        org_scm_instance = c.cs_repo.scm_instance # property with expensive cache invalidation check!!!

        try:

            c.cs_ranges = []

            for x in c.pull_request.revisions:

                c.cs_ranges.append(org_scm_instance.get_changeset(x))

        except ChangesetDoesNotExistError:

            c.cs_ranges = []

            h.flash(_('Revision %s not found in %s') % (x, c.cs_repo.repo_name),

                'error')

        c.cs_ranges_org = None # not stored and not important and moving target - could be calculated ...

        revs = [ctx.revision for ctx in reversed(c.cs_ranges)]

        c.jsdata = graph_data(org_scm_instance, revs)

        c.is_range = False

        try:

            if c.a_ref_type == 'rev': # this looks like a free range where target is ancestor

                cs_a = org_scm_instance.get_changeset(c.a_rev)

                root_parents = c.cs_ranges[0].parents

                c.is_range = cs_a in root_parents

                #c.merge_root = len(root_parents) > 1 # a range starting with a merge might deserve a warning

        except ChangesetDoesNotExistError: # probably because c.a_rev not found

            pass

        except IndexError: # probably because c.cs_ranges is empty, probably because revisions are missing

            pass

        avail_revs = set()

        avail_show = []

        c.cs_branch_name = c.cs_ref_name

        c.a_branch_name = None

        other_scm_instance = c.a_repo.scm_instance

        c.update_msg = ""

        c.update_msg_other = ""

        try:

            if not c.cs_ranges:

                c.update_msg = _('Error: changesets not found when displaying pull request from %s.') % c.cs_rev

            elif org_scm_instance.alias == 'hg' and c.a_ref_name != 'ancestor':

                if c.cs_ref_type != 'branch':

                    c.cs_branch_name = org_scm_instance.get_changeset(c.cs_ref_name).branch # use ref_type ?

                c.a_branch_name = c.a_ref_name

                if c.a_ref_type != 'branch':

                    try:

                        c.a_branch_name = other_scm_instance.get_changeset(c.a_ref_name).branch # use ref_type ?

                    except EmptyRepositoryError:

                        c.a_branch_name = 'null' # not a branch name ... but close enough

                # candidates: descendants of old head that are on the right branch

                #             and not are the old head itself ...

                #             and nothing at all if old head is a descendant of target ref name

                if not c.is_range and other_scm_instance._repo.revs('present(%s)::&%s', c.cs_ranges[-1].raw_id, c.a_branch_name):

                    c.update_msg = _('This pull request has already been merged to %s.') % c.a_branch_name

                elif c.pull_request.is_closed():

                    c.update_msg = _('This pull request has been closed and can not be updated.')

                else: # look for descendants of PR head on source branch in org repo

                    avail_revs = org_scm_instance._repo.revs('%s:: & branch(%s)',

                                                             revs[0], c.cs_branch_name)

                    if len(avail_revs) > 1: # more than just revs[0]

                        # also show changesets that not are descendants but would be merged in

                        targethead = other_scm_instance.get_changeset(c.a_branch_name).raw_id

                        if org_scm_instance.path != other_scm_instance.path:

                            # Note: org_scm_instance.path must come first so all

                            # valid revision numbers are 100% org_scm compatible

                            # - both for avail_revs and for revset results

                                                                   org_scm_instance.path,

                                                                   other_scm_instance.path)

                        else:

                            hgrepo = org_scm_instance._repo

                        show = set(hgrepo.revs('::%ld & !::parents(%s) & !::%s',

                                               avail_revs, revs[0], targethead))

                        if show:

                            c.update_msg = _('The following additional changes are available on %s:') % c.cs_branch_name

                        else:

                            c.update_msg = _('No additional changesets found for iterating on this pull request.')

                    else:

                        show = set()

                        avail_revs = set() # drop revs[0]

                        c.update_msg = _('No additional changesets found for iterating on this pull request.')

                    # TODO: handle branch heads that not are tip-most

                    brevs = org_scm_instance._repo.revs('%s - %ld - %s', c.cs_branch_name, avail_revs, revs[0])

                    if brevs:

                        # also show changesets that are on branch but neither ancestors nor descendants

                        show.update(org_scm_instance._repo.revs('::%ld - ::%ld - ::%s', brevs, avail_revs, c.a_branch_name))

                        show.add(revs[0]) # make sure graph shows this so we can see how they relate

                        c.update_msg_other = _('Note: Branch %s has another head: %s.') % (c.cs_branch_name,

                            h.short_id(org_scm_instance.get_changeset((max(brevs))).raw_id))

                    avail_show = sorted(show, reverse=True)

            elif org_scm_instance.alias == 'git':

                c.cs_repo.scm_instance.get_changeset(c.cs_rev) # check it exists - raise ChangesetDoesNotExistError if not

                c.update_msg = _("Git pull requests don't support iterating yet.")

        except ChangesetDoesNotExistError:

            c.update_msg = _('Error: some changesets not found when displaying pull request from %s.') % c.cs_rev

        c.avail_revs = avail_revs

        c.avail_cs = [org_scm_instance.get_changeset(r) for r in avail_show]

        c.avail_jsdata = graph_data(org_scm_instance, avail_show)

        raw_ids = [x.raw_id for x in c.cs_ranges]

        c.cs_comments = c.cs_repo.get_comments(raw_ids)

        c.cs_statuses = c.cs_repo.statuses(raw_ids)

        ignore_whitespace = request.GET.get('ignorews') == '1'

        line_context = safe_int(request.GET.get('context'), 3)

        c.ignorews_url = _ignorews_url

        c.context_url = _context_url

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

        diff_limit = None if fulldiff else self.cut_off_limit

        # we swap org/other ref since we run a simple diff on one repo

        log.debug('running diff between %s and %s in %s',

                  c.a_rev, c.cs_rev, org_scm_instance.path)

        try:

            raw_diff = diffs.get_diff(org_scm_instance, rev1=safe_str(c.a_rev), rev2=safe_str(c.cs_rev),

                                      ignore_whitespace=ignore_whitespace, context=line_context)

        except ChangesetDoesNotExistError:

            raw_diff = _("The diff can't be shown - the PR revisions could not be found.")

        diff_processor = diffs.DiffProcessor(raw_diff or '', diff_limit=diff_limit)

        c.limited_diff = diff_processor.limited_diff

        c.file_diff_data = []

        c.lines_added = 0

        c.lines_deleted = 0

        for f in diff_processor.parsed:

            st = f['stats']

            c.lines_added += st['added']

            c.lines_deleted += st['deleted']

            filename = f['filename']

            fid = h.FID('', filename)

            html_diff = diffs.as_html(enable_comments=True, parsed_lines=[f])

            c.file_diff_data.append((fid, None, f['operation'], f['old_filename'], filename, html_diff, st))

        # inline comments

        c.inline_cnt = 0

        c.inline_comments = cc_model.get_inline_comments(

                                c.db_repo.repo_id,

                                pull_request=pull_request_id)

        # count inline comments

        for __, lines in c.inline_comments:

            for comments in lines.values():

                c.inline_cnt += len(comments)

        # comments

        c.comments = cc_model.get_comments(c.db_repo.repo_id, pull_request=pull_request_id)

        # (badly named) pull-request status calculation based on reviewer votes

        (c.pull_request_reviewers,

         c.pull_request_pending_reviewers,

         c.current_voting_result,

         ) = cs_model.calculate_pull_request_result(c.pull_request)

        c.changeset_statuses = ChangesetStatus.STATUSES

        c.is_ajax_preview = False

        c.ancestors = None # [c.a_rev] ... but that is shown in an other way

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

    @LoginRequired()

    @HasRepoPermissionLevelDecorator('read')

    @jsonify

import os

import posixpath

from kallithea.lib.vcs.backends.base import BaseChangeset

from kallithea.lib.vcs.conf import settings

from kallithea.lib.vcs.exceptions import ChangesetDoesNotExistError, ChangesetError, ImproperArchiveTypeError, NodeDoesNotExistError, VCSError

from kallithea.lib.vcs.nodes import (

    AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode, SubModuleNode)

from kallithea.lib.vcs.utils import date_fromtimestamp, safe_str, safe_unicode

from kallithea.lib.vcs.utils.hgcompat import archival, hex, obsutil

from kallithea.lib.vcs.utils.lazy import LazyProperty

from kallithea.lib.vcs.utils.paths import get_dirs_for_path

class MercurialChangeset(BaseChangeset):

    """

    Represents state of the repository at a revision.

    """

    def __init__(self, repository, revision):

        self.repository = repository

        assert isinstance(revision, basestring), repr(revision)

        self.raw_id = revision

        self._ctx = repository._repo[revision]

        self.revision = self._ctx._rev

        self.nodes = {}

    @LazyProperty

    def tags(self):

        return [safe_unicode(tag) for tag in self._ctx.tags()]

    @LazyProperty

    def branch(self):

        return safe_unicode(self._ctx.branch())

    @LazyProperty

    def branches(self):

        return [safe_unicode(self._ctx.branch())]

    @LazyProperty

    def closesbranch(self):

        return self._ctx.closesbranch()

    @LazyProperty

    def obsolete(self):

        return self._ctx.obsolete()

    @LazyProperty

    def bumped(self):

    @LazyProperty

    def divergent(self):

    @LazyProperty

    def extinct(self):

        return self._ctx.extinct()

    @LazyProperty

    def unstable(self):

    @LazyProperty

    def phase(self):

        if(self._ctx.phase() == 1):

            return 'Draft'

        elif(self._ctx.phase() == 2):

            return 'Secret'

        else:

            return ''

    @LazyProperty

    def successors(self):

        successors = obsutil.successorssets(self._ctx._repo, self._ctx.node(), closest=True)

        if successors:

            # flatten the list here handles both divergent (len > 1)

            # and the usual case (len = 1)

            successors = [hex(n)[:12] for sub in successors for n in sub if n != self._ctx.node()]

        return successors

    @LazyProperty

    def predecessors(self):

        return [hex(n)[:12] for n in obsutil.closestpredecessors(self._ctx._repo, self._ctx.node())]

    @LazyProperty

    def bookmarks(self):

        return [safe_unicode(bookmark) for bookmark in self._ctx.bookmarks()]

    @LazyProperty

    def message(self):

        return safe_unicode(self._ctx.description())

    @LazyProperty

    def committer(self):

        return safe_unicode(self.author)

    @LazyProperty

    def author(self):

        return safe_unicode(self._ctx.user())

    @LazyProperty

    def date(self):

        return date_fromtimestamp(*self._ctx.date())

    @LazyProperty

    def _timestamp(self):

        return self._ctx.date()[0]

    @LazyProperty

    def status(self):

        """

        Returns modified, added, removed, deleted files for current changeset

        """

        return self.repository._repo.status(self._ctx.p1().node(),

                                            self._ctx.node())

    @LazyProperty

    def _file_paths(self):

        return list(self._ctx)

    @LazyProperty

    def _dir_paths(self):

        p = list(set(get_dirs_for_path(*self._file_paths)))

        p.insert(0, '')

        return p

    @LazyProperty

    def _paths(self):

        return self._dir_paths + self._file_paths

    @LazyProperty

    def id(self):

        if self.last:

            return u'tip'

        return self.short_id

    @LazyProperty

    def short_id(self):

        return self.raw_id[:12]

    @LazyProperty

    def parents(self):

        """

        Returns list of parents changesets.

        """

        return [self.repository.get_changeset(parent.rev())

                for parent in self._ctx.parents() if parent.rev() >= 0]

    @LazyProperty

    def children(self):

        """

        Returns list of children changesets.

        """

        return [self.repository.get_changeset(child.rev())

                for child in self._ctx.children() if child.rev() >= 0]

            cs = cs.repository.get_changeset(prev_rev)

            if not branch or branch == cs.branch:

                return cs

    def diff(self):

        # Only used to feed diffstat

        return ''.join(self._ctx.diff())

    def _fix_path(self, path):

        """

        Paths are stored without trailing slash so we need to get rid off it if

        needed. Also mercurial keeps filenodes as str so we need to decode

        from unicode to str

        """

        if path.endswith('/'):

            path = path.rstrip('/')

        return safe_str(path)

    def _get_kind(self, path):

        path = self._fix_path(path)

        if path in self._file_paths:

            return NodeKind.FILE

        elif path in self._dir_paths:

            return NodeKind.DIR

        else:

            raise ChangesetError("Node does not exist at the given path '%s'"

                % (path))

    def _get_filectx(self, path):

        path = self._fix_path(path)

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

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

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

        return self._ctx.filectx(path)

    def _extract_submodules(self):

        """

        returns a dictionary with submodule information from substate file

        of hg repository

        """

        return self._ctx.substate

    def get_file_mode(self, path):

        """

        Returns stat mode of the file at the given ``path``.

        """

        fctx = self._get_filectx(path)

        if 'x' in fctx.flags():

            return 0o100755

        else:

            return 0o100644

    def get_file_content(self, path):

        """

        Returns content of the file at given ``path``.

        """

        fctx = self._get_filectx(path)

        return fctx.data()

    def get_file_size(self, path):

        """

        Returns size of the file at given ``path``.

        """

        fctx = self._get_filectx(path)

        return fctx.size()

    def get_file_changeset(self, path):

        """

        Returns last commit of the file at the given ``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()

        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

        """

        allowed_kinds = settings.ARCHIVE_SPECS

        if kind not in allowed_kinds:

            raise ImproperArchiveTypeError('Archive kind not supported use one'

                'of %s' % ' '.join(allowed_kinds))

        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")

        archival.archive(self.repository._repo, stream, self.raw_id,

                         kind, prefix=prefix, subrepos=subrepos)

    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))

        path = self._fix_path(path)

        filenodes = [FileNode(f, changeset=self) for f in self._file_paths

            if os.path.dirname(f) == path]

        dirs = path == '' and '' or [d for d in self._dir_paths

            if d and posixpath.dirname(d) == path]

        dirnodes = [DirNode(d, changeset=self) for d in dirs

            if os.path.dirname(d) == path]

        als = self.repository.alias

        for k, vals in self._extract_submodules().iteritems():

            #vals = url,rev,type

            loc = vals[0]

            cs = vals[1]

            dirnodes.append(SubModuleNode(k, url=loc, changeset=cs,

                                          alias=als))

        nodes = dirnodes + filenodes

        for node in nodes:

            self.nodes[node.path] = node

        nodes.sort()

        return nodes

    def get_node(self, path):

        """

        Returns ``Node`` object from the given ``path``. If there is no node at

        the given ``path``, ``ChangesetError`` would be raised.

        """

        path = self._fix_path(path)

        if path not in self.nodes:

            if path in self._file_paths:

                node = FileNode(path, changeset=self)

            elif path in self._dir_paths or path in self._dir_paths:

                if path == '':

                    node = RootNode(changeset=self)

                else:

                    node = DirNode(path, changeset=self)

            else:

                raise NodeDoesNotExistError("There is no file nor directory "

                    "at the given path: '%s' at revision %s"

                    % (path, self.short_id))

            # cache node

            self.nodes[path] = node

        return self.nodes[path]

    @LazyProperty

    def affected_files(self):

        """

            # now check if it's a proper hg repo

            try:

                httppeer.instance(repoui or ui.ui(), url, False).lookup('tip')

            except Exception as e:

                raise urllib2.URLError(

                    "url [%s] does not look like an hg repo org_exc: %s"

                    % (cleaned_uri, e))

        return True

    def _get_repo(self, create, src_url=None, update_after_clone=False):

        """

        Function will check for mercurial repository in given path and return

        a localrepo object. If there is no repository in that path it will

        raise an exception unless ``create`` parameter is set to True - in

        that case repository would be created and returned.

        If ``src_url`` is given, would try to clone repository from the

        location at given clone_point. Additionally it'll make update to

        working copy accordingly to ``update_after_clone`` flag

        """

        try:

            if src_url:

                url = safe_str(self._get_url(src_url))

                opts = {}

                if not update_after_clone:

                    opts.update({'noupdate': True})

                MercurialRepository._check_url(url, self.baseui)

                clone(self.baseui, url, self.path, **opts)

                # Don't try to create if we've already cloned repo

                create = False

            return localrepo.instance(self.baseui, self.path, create=create)

        except (Abort, RepoError) as err:

            if create:

                msg = "Cannot create repository at %s. Original error was %s" \

                    % (self.path, err)

            else:

                msg = "Not valid repository at %s. Original error was %s" \

                    % (self.path, err)

            raise RepositoryError(msg)

    @LazyProperty

    def in_memory_changeset(self):

        return MercurialInMemoryChangeset(self)

    @LazyProperty

    def description(self):

        undefined_description = u'unknown'

        _desc = self._repo.ui.config('web', 'description', None, untrusted=True)

        return safe_unicode(_desc or undefined_description)

    @LazyProperty

    def contact(self):

        undefined_contact = u'Unknown'

        return safe_unicode(get_contact(self._repo.ui.config)

                            or undefined_contact)

    @LazyProperty

    def last_change(self):

        """

        Returns last change made on this repository as datetime object

        """

        return date_fromtimestamp(self._get_mtime(), makedate()[1])

    def _get_mtime(self):

        try:

            return time.mktime(self.get_changeset().date.timetuple())

        except RepositoryError:

            # fallback to filesystem

            cl_path = os.path.join(self.path, '.hg', "00changelog.i")

            st_path = os.path.join(self.path, '.hg', "store")

            if os.path.exists(cl_path):

                return os.stat(cl_path).st_mtime

            else:

                return os.stat(st_path).st_mtime

    def _get_revision(self, revision):

        """

        Gets an ID revision given as str. This will always return a full

        40 char revision number

        :param revision: str or int or None

        """

        if isinstance(revision, unicode):

            revision = safe_str(revision)

        if self._empty:

            raise EmptyRepositoryError("There are no changesets yet")

        if revision in [-1, None]:

            revision = 'tip'

        try:

            if isinstance(revision, int):