# -*- 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.controllers.compare

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

compare controller for pylons 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 pylons import request, tmpl_context as c, url

from pylons.i18n.translation import _

from webob.exc import HTTPFound, HTTPBadRequest

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

from kallithea.lib import helpers as h

from kallithea.lib.base import BaseRepoController, render

from kallithea.lib.auth import LoginRequired, HasRepoPermissionAnyDecorator

from kallithea.lib import diffs

from kallithea.model.db import Repository

from kallithea.lib.diffs import LimitedDiffContainer

from kallithea.controllers.changeset import _ignorews_url, _context_url

from kallithea.lib.graphmod import graph_data

from kallithea.lib.compat import json

log = logging.getLogger(__name__)

class CompareController(BaseRepoController):

    def __before__(self):

        super(CompareController, self).__before__()

    @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 ancestor 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

        """

        ancestor = None

        if org_rev == other_rev:

            org_changesets = []

            other_changesets = []

            ancestor = org_rev

        elif alias == 'hg':

            #case two independent repos

            if org_repo != other_repo:

                hgrepo = unionrepo.unionrepository(other_repo.baseui,

                                                   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

    @LoginRequired()

    @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',

                                   'repository.admin')

    def index(self, repo_name):

        c.compare_home = True

        org_repo = c.db_repo.repo_name

        other_repo = request.GET.get('other_repo', org_repo)

        c.a_repo = Repository.get_by_repo_name(org_repo)

        c.cs_repo = Repository.get_by_repo_name(other_repo)

        c.a_ref_name = c.cs_ref_name = _('Select changeset')

        return render('compare/compare_diff.html')

    @LoginRequired()

    @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',

                                   'repository.admin')

    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()

        org_repo = c.db_repo.repo_name

        other_repo = request.GET.get('other_repo', org_repo)

        # 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

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

        # partial uses compare_cs.html template directly

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

        # as_form puts hidden input field with changeset revisions

        c.as_form = partial and request.GET.get('as_form')

        # swap url for compare_diff page - never partial and never as_form

        c.swap_url = h.url('compare_url',

            repo_name=other_repo,

            org_ref_type=other_ref_type, org_ref_name=other_ref_name,

            other_repo=org_repo,

            other_ref_type=org_ref_type, other_ref_name=org_ref_name,

            merge=merge or '')

        # set callbacks for generating markup for icons

        c.ignorews_url = _ignorews_url

        c.context_url = _context_url

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

        org_repo = Repository.get_by_repo_name(org_repo)

        other_repo = Repository.get_by_repo_name(other_repo)

        if org_repo is None:

            msg = 'Could not find org repo %s' % org_repo

            log.error(msg)

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

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

        if other_repo is None:

            msg = 'Could not find other repo %s' % other_repo

            log.error(msg)

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

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

        if org_repo.scm_instance.alias != other_repo.scm_instance.alias:

            msg = 'compare of two different kind of remote repos not available'

            log.error(msg)

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

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

        c.a_rev = self._get_ref_rev(org_repo, org_ref_type, org_ref_name,

            returnempty=True)

        c.cs_rev = self._get_ref_rev(other_repo, other_ref_type, other_ref_name)

        c.compare_home = False

        c.a_repo = org_repo

        c.a_ref_name = org_ref_name

        c.a_ref_type = org_ref_type

        c.cs_repo = other_repo

        c.cs_ref_name = other_ref_name

        c.cs_ref_type = other_ref_type

        c.cs_ranges, c.cs_ranges_org, c.ancestor = self._get_changesets(

            org_repo.scm_instance.alias, org_repo.scm_instance, c.a_rev,

            other_repo.scm_instance, c.cs_rev)

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

        c.cs_comments = other_repo.get_comments(raw_ids)

        c.statuses = other_repo.statuses(raw_ids)

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

        c.jsdata = json.dumps(graph_data(c.cs_repo.scm_instance, revs))

        if partial:

            return render('compare/compare_cs.html')

        if merge and c.ancestor:

            # case we want a simple diff without incoming changesets,

# -*- 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.controllers.files

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

Files controller 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: Apr 21, 2010

:author: marcink

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

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

"""

import os

import posixpath

import logging

import traceback

import tempfile

import shutil

from pylons import request, response, tmpl_context as c, url

from pylons.i18n.translation import _

from webob.exc import HTTPFound

from kallithea.lib.utils import jsonify, action_logger

from kallithea.lib import diffs

from kallithea.lib import helpers as h

from kallithea.lib.compat import OrderedDict

from kallithea.lib.utils2 import convert_line_endings, detect_mode, safe_str, \

from kallithea.lib.auth import LoginRequired, HasRepoPermissionAnyDecorator

from kallithea.lib.base import BaseRepoController, render

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

from kallithea.lib.vcs.conf import settings

from kallithea.lib.vcs.exceptions import RepositoryError, \

    ChangesetDoesNotExistError, EmptyRepositoryError, \

    ImproperArchiveTypeError, VCSError, NodeAlreadyExistsError, \

    NodeDoesNotExistError, ChangesetError, NodeError

from kallithea.lib.vcs.nodes import FileNode

from kallithea.model.repo import RepoModel

from kallithea.model.scm import ScmModel

from kallithea.model.db import Repository

from kallithea.controllers.changeset import anchor_url, _ignorews_url, \

    _context_url, get_line_ctx, get_ignore_ws

from webob.exc import HTTPNotFound

from kallithea.lib.exceptions import NonRelativePathError

log = logging.getLogger(__name__)

class FilesController(BaseRepoController):

    def __before__(self):

        super(FilesController, self).__before__()

        c.cut_off_limit = self.cut_off_limit

    def __get_cs(self, rev, silent_empty=False):

        """

        Safe way to get changeset if error occur it redirects to tip with

        proper message

        :param rev: revision to fetch

        :silent_empty: return None if repository is empty

        """

        try:

            return c.db_repo_scm_instance.get_changeset(rev)

        except EmptyRepositoryError as e:

            if silent_empty:

                return None

            url_ = url('files_add_home',

                       repo_name=c.repo_name,

                       revision=0, f_path='', anchor='edit')

            add_new = h.link_to(_('Click here to add new file'), url_, class_="alert-link")

            h.flash(h.literal(_('There are no files yet. %s') % add_new),

        if archive_cache_dir and not subrepos: # TODO: subrepo caching?

            if not os.path.isdir(archive_cache_dir):

                os.makedirs(archive_cache_dir)

            cached_archive_path = os.path.join(archive_cache_dir, archive_name)

            if os.path.isfile(cached_archive_path):

                log.debug('Found cached archive in %s', cached_archive_path)

                archive_path = cached_archive_path

            else:

                log.debug('Archive %s is not yet cached', archive_name)

        if archive_path is None:

            # generate new archive

            fd, archive_path = tempfile.mkstemp()

            log.debug('Creating new temp archive in %s', archive_path)

            with os.fdopen(fd, 'wb') as stream:

                cs.fill_archive(stream=stream, kind=fileformat, subrepos=subrepos)

                # stream (and thus fd) has been closed by cs.fill_archive

            if cached_archive_path is not None:

                # we generated the archive - move it to cache

                log.debug('Storing new archive in %s', cached_archive_path)

                shutil.move(archive_path, cached_archive_path)

                archive_path = cached_archive_path

        def get_chunked_archive(archive_path):

            stream = open(archive_path, 'rb')

            while True:

                data = stream.read(16 * 1024)

                if not data:

                    break

                yield data

            stream.close()

            if archive_path != cached_archive_path:

                log.debug('Destroying temp archive %s', archive_path)

                os.remove(archive_path)

        action_logger(user=c.authuser,

                      action='user_downloaded_archive:%s' % (archive_name),

                      repo=repo_name, ipaddr=self.ip_addr, commit=True)

        response.content_disposition = str('attachment; filename=%s' % (archive_name))

        response.content_type = str(content_type)

        return get_chunked_archive(archive_path)

    @LoginRequired()

    @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',

                                   'repository.admin')

    def diff(self, repo_name, f_path):

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

        diff2 = request.GET.get('diff2', '')

        diff1 = request.GET.get('diff1', '') or diff2

        c.action = request.GET.get('diff')

        c.no_changes = diff1 == diff2

        c.f_path = f_path

        c.big_diff = False

        c.anchor_url = anchor_url

        c.ignorews_url = _ignorews_url

        c.context_url = _context_url

        c.changes = OrderedDict()

        c.changes[diff2] = []

        #special case if we want a show rev only, it's impl here

        #to reduce JS and callbacks

        if request.GET.get('show_rev'):

            if str2bool(request.GET.get('annotate', 'False')):

                _url = url('files_annotate_home', repo_name=c.repo_name,

                           revision=diff1, f_path=c.f_path)

            else:

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

                           revision=diff1, f_path=c.f_path)

            raise HTTPFound(location=_url)

        try:

            if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:

                c.changeset_1 = c.db_repo_scm_instance.get_changeset(diff1)

                try:

                    node1 = c.changeset_1.get_node(f_path)

                    if node1.is_dir():

                        raise NodeError('%s path is a %s not a file'

                                        % (node1, type(node1)))

                except NodeDoesNotExistError:

                    c.changeset_1 = EmptyChangeset(cs=diff1,

                                                   revision=c.changeset_1.revision,

                                                   repo=c.db_repo_scm_instance)

                    node1 = FileNode(f_path, '', changeset=c.changeset_1)

            else:

                c.changeset_1 = EmptyChangeset(repo=c.db_repo_scm_instance)

                node1 = FileNode(f_path, '', changeset=c.changeset_1)

            if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:

                c.changeset_2 = c.db_repo_scm_instance.get_changeset(diff2)

                try:

                    node2 = c.changeset_2.get_node(f_path)

                    if node2.is_dir():

                        raise NodeError('%s path is a %s not a file'

                                        % (node2, type(node2)))

                    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

                            hgrepo = unionrepo.unionrepository(org_scm_instance.baseui,

                                                               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))

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

                    else:

                        show = set()

                        avail_revs = set() # drop revs[0]

                        c.update_msg = _('No 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: revision %s was not found. Please create a new pull request!') % 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 = json.dumps(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.statuses = c.cs_repo.statuses(raw_ids)

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

        c.ignorews_url = _ignorews_url

        c.context_url = _context_url

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

        diff_limit = self.cut_off_limit if not c.fulldiff else None

        # 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:

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

                                                ignore_whitespace=ignore_whitespace,

                                                context=line_context)

        except ChangesetDoesNotExistError:

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

        diff_processor = diffs.DiffProcessor(txtdiff or '', format='gitdiff',

                                             diff_limit=diff_limit)

        _parsed = diff_processor.prepare()

        c.limited_diff = False

        if isinstance(_parsed, LimitedDiffContainer):

            c.limited_diff = True

        c.files = []

        c.changes = {}

        c.lines_added = 0

        c.lines_deleted = 0

        for f in _parsed:

            st = f['stats']

            c.lines_added += st['added']

            c.lines_deleted += st['deleted']

            fid = h.FID('', f['filename'])

            c.files.append([fid, f['operation'], f['filename'], f['stats']])

            htmldiff = diff_processor.as_html(enable_comments=True,

                                              parsed_lines=[f])

            c.changes[fid] = [f['operation'], f['filename'], htmldiff]

        # 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,