import logging

import traceback

from collections import defaultdict

from tg import tmpl_context as c, request, response

from tg.i18n import ugettext as _

from webob.exc import HTTPFound, HTTPForbidden, HTTPBadRequest, HTTPNotFound

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

    ChangesetDoesNotExistError, EmptyRepositoryError

import kallithea.lib.helpers as h

from kallithea.lib.auth import LoginRequired, HasRepoPermissionLevelDecorator, \

    NotAnonymous

from kallithea.lib.base import BaseRepoController, render, jsonify

from kallithea.lib.utils import action_logger

from kallithea.lib.compat import OrderedDict

from kallithea.lib import diffs

from kallithea.model.db import ChangesetComment, ChangesetStatus

from kallithea.model.comment import ChangesetCommentsModel

from kallithea.model.changeset_status import ChangesetStatusModel

from kallithea.model.meta import Session

from kallithea.model.repo import RepoModel

from kallithea.lib.exceptions import StatusChangeOnClosedPullRequestError

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

from kallithea.lib.utils2 import safe_unicode

from kallithea.lib.graphmod import graph_data

log = logging.getLogger(__name__)

def _update_with_GET(params, GET):

    for k in ['diff1', 'diff2', 'diff']:

        params[k] += GET.getall(k)

def anchor_url(revision, path, GET):

    fid = h.FID(revision, path)

    return h.url.current(anchor=fid, **dict(GET))

def get_ignore_ws(fid, GET):

    ig_ws_global = GET.get('ignorews')

    ig_ws = filter(lambda k: k.startswith('WS'), GET.getall(fid))

    if ig_ws:

        try:

            return int(ig_ws[0].split(':')[-1])

                comments.update((st.comment_id, st.comment)

                                for st in ChangesetStatusModel()

                                .get_statuses(c.db_repo.repo_id,

                                              changeset.raw_id, with_revisions=True)

                                if st.comment_id is not None)

                inlines = ChangesetCommentsModel() \

                            .get_inline_comments(c.db_repo.repo_id,

                                                 revision=changeset.raw_id)

                c.inline_comments.extend(inlines)

            cs2 = changeset.raw_id

            cs1 = changeset.parents[0].raw_id if changeset.parents else EmptyChangeset().raw_id

            context_lcl = get_line_ctx('', request.GET)

            ign_whitespace_lcl = get_ignore_ws('', request.GET)

            raw_diff = c.db_repo_scm_instance.get_diff(cs1, cs2,

                ignore_whitespace=ign_whitespace_lcl, context=context_lcl)

            diff_limit = None if c.fulldiff else self.cut_off_limit

            file_diff_data = []

            if method == 'show':

                diff_processor = diffs.DiffProcessor(raw_diff,

                                                     vcs=c.db_repo_scm_instance.alias,

                                                     diff_limit=diff_limit)

                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(changeset.raw_id, filename)

                    url_fid = h.FID('', filename)

                    diff = diff_processor.as_html(enable_comments=enable_comments,

                                                  parsed_lines=[f])

                    file_diff_data.append((fid, url_fid, f['operation'], f['old_filename'], filename, diff, st))

            else:

                # downloads/raw we only need RAW diff nothing else

                file_diff_data.append(('', None, None, None, raw_diff, None))

            c.changes[changeset.raw_id] = (cs1, cs2, file_diff_data)

        # sort comments in creation order

        c.comments = [com for com_id, com in sorted(comments.items())]

        # count inline comments

        for __, lines in c.inline_comments:

            for comments in lines.values():

                c.inline_cnt += len(comments)

        if len(c.cs_ranges) == 1:

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, tmpl_context as c

from tg.i18n import ugettext as _

from webob.exc import HTTPFound, HTTPBadRequest, HTTPNotFound

from kallithea.config.routing import url

from kallithea.lib.utils2 import safe_str, safe_int

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, HasRepoPermissionLevelDecorator

from kallithea.lib import diffs

from kallithea.model.db import Repository

from kallithea.controllers.changeset import _ignorews_url, _context_url

from kallithea.lib.graphmod import graph_data

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))

            # case we want a simple diff without incoming changesets,

            # previewing what will be merged.

            # Make the diff on the other repo (which is known to have other_rev)

            log.debug('Using ancestor %s as rev1 instead of %s',

                      rev1, c.a_rev)

            org_repo = other_repo

        else: # comparing tips, not necessarily linearly related

            if org_repo != other_repo:

                # TODO: we could do this by using hg unionrepo

                log.error('cannot compare across repos %s and %s', org_repo, other_repo)

                h.flash(_('Cannot compare repositories without using common ancestor'), category='error')

                raise HTTPBadRequest

            rev1 = c.a_rev

        diff_limit = None if fulldiff else self.cut_off_limit

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

                  rev1, c.cs_rev, org_repo.scm_instance.path)

        raw_diff = org_repo.scm_instance.get_diff(rev1=rev1, rev2=c.cs_rev,

                                      ignore_whitespace=ignore_whitespace,

                                      context=line_context)

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

        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)

            diff = diff_processor.as_html(enable_comments=False,

                                          parsed_lines=[f])

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

        return render('compare/compare_diff.html')

Feed 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 23, 2010

:author: marcink

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

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

"""

import logging

from tg import response, tmpl_context as c

from tg.i18n import ugettext as _

from beaker.cache import cache_region, region_invalidate

from webhelpers.feedgenerator import Atom1Feed, Rss201rev2Feed

from kallithea import CONFIG

from kallithea.lib import helpers as h

from kallithea.lib.auth import LoginRequired, HasRepoPermissionLevelDecorator

from kallithea.lib.base import BaseRepoController

from kallithea.model.db import CacheInvalidation

from kallithea.lib.utils2 import safe_int, str2bool, safe_unicode

log = logging.getLogger(__name__)

language = 'en-us'

ttl = "5"

class FeedController(BaseRepoController):

    @LoginRequired(api_access=True)

    @HasRepoPermissionLevelDecorator('read')

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

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

    def _get_title(self, cs):

        return h.shorter(cs.message, 160)

    def __get_desc(self, cs):

        desc_msg = [(_('%s committed on %s')

                     % (h.person(cs.author), h.fmt_date(cs.date))) + '<br/>']

        # branches, tags, bookmarks

        if cs.branch:

            desc_msg.append('branch: %s<br/>' % cs.branch)

        for book in cs.bookmarks:

            desc_msg.append('bookmark: %s<br/>' % book)

        for tag in cs.tags:

            desc_msg.append('tag: %s<br/>' % tag)

        changes = []

        diff_limit = safe_int(CONFIG.get('rss_cut_off_limit', 32 * 1024))

        raw_diff = cs.diff()

        diff_processor = DiffProcessor(raw_diff,

                                       diff_limit=diff_limit,

                                       inline_diff=False)

        for st in diff_processor.parsed:

            st.update({'added': st['stats']['added'],

                       'removed': st['stats']['deleted']})

            changes.append('\n %(operation)s %(filename)s '

                           '(%(added)s lines added, %(removed)s lines removed)'

                            % st)

            changes = changes + ['\n ' +

                                 _('Changeset was too big and was cut off...')]

        # rev link

        _url = h.canonical_url('changeset_home', repo_name=c.db_repo.repo_name,

                   revision=cs.raw_id)

        desc_msg.append('changeset: <a href="%s">%s</a>' % (_url, cs.raw_id[:8]))

        desc_msg.append('<pre>')

        desc_msg.append(h.urlify_text(cs.message))

        desc_msg.append('\n')

        desc_msg.extend(changes)

        if str2bool(CONFIG.get('rss_include_diff', False)):

            desc_msg.append('\n\n')

            desc_msg.append(raw_diff)

        desc_msg.append('</pre>')

        return map(safe_unicode, desc_msg)

    def atom(self, repo_name):

        """Produce an atom-1.0 feed via feedgenerator module"""

        @cache_region('long_term', '_get_feed_from_cache')

        def _get_feed_from_cache(key, kind):

            feed = Atom1Feed(

pull requests controller for Kallithea for initializing pull requests

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 7, 2012

:author: marcink

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

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

"""

import logging

import traceback

import formencode

from tg import request, tmpl_context as c

from tg.i18n import ugettext as _

from webob.exc import HTTPFound, HTTPNotFound, HTTPForbidden, HTTPBadRequest

from kallithea.config.routing import url

from kallithea.lib import helpers as h

from kallithea.lib import diffs

from kallithea.lib.auth import LoginRequired, HasRepoPermissionLevelDecorator, \

    NotAnonymous

from kallithea.lib.base import BaseRepoController, render, jsonify

from kallithea.lib.page import Page

from kallithea.lib.utils import action_logger

from kallithea.lib.vcs.exceptions import EmptyRepositoryError, ChangesetDoesNotExistError

from kallithea.lib.vcs.utils import safe_str

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

from kallithea.model.db import PullRequest, ChangesetStatus, ChangesetComment, \

    PullRequestReviewer, Repository, User

from kallithea.model.pull_request import CreatePullRequestAction, CreatePullRequestIterationAction, PullRequestModel

from kallithea.model.meta import Session

from kallithea.model.repo import RepoModel

from kallithea.model.comment import ChangesetCommentsModel

from kallithea.model.changeset_status import ChangesetStatusModel

from kallithea.model.forms import PullRequestForm, PullRequestPostForm

from kallithea.lib.utils2 import safe_int

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

    create_comment

from kallithea.controllers.compare import CompareController

from kallithea.lib.graphmod import graph_data

log = logging.getLogger(__name__)

def _get_reviewer(user_id):

    """Look up user by ID and validate it as a potential reviewer."""

        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 = 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:

            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.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)

            diff = diff_processor.as_html(enable_comments=True,

                                          parsed_lines=[f])

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

                % filenode.__class__)

    repo = filenode_new.changeset.repository

    old_raw_id = getattr(filenode_old.changeset, 'raw_id', repo.EMPTY_CHANGESET)

    new_raw_id = getattr(filenode_new.changeset, 'raw_id', repo.EMPTY_CHANGESET)

    vcs_gitdiff = repo.get_diff(old_raw_id, new_raw_id, filenode_new.path,

                                ignore_whitespace, context)

    return vcs_gitdiff

NEW_FILENODE = 1

DEL_FILENODE = 2

MOD_FILENODE = 3

RENAMED_FILENODE = 4

COPIED_FILENODE = 5

CHMOD_FILENODE = 6

BIN_FILENODE = 7

class DiffLimitExceeded(Exception):

    pass

class DiffProcessor(object):

    """

    Give it a unified or git diff and it returns a list of the files that were

    mentioned in the diff together with a dict of meta information that

    can be used to render it in a HTML template.

    """

    _diff_git_re = re.compile('^diff --git', re.MULTILINE)

    _chunk_re = re.compile(r'^@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*)')

    _newline_marker = re.compile(r'^\\ No newline at end of file')

    _git_header_re = re.compile(r"""

        ^diff[ ]--git

            [ ]a/(?P<a_path>.+?)[ ]b/(?P<b_path>.+?)\n

        (?:^old[ ]mode[ ](?P<old_mode>\d+)\n

           ^new[ ]mode[ ](?P<new_mode>\d+)(?:\n|$))?

        (?:^similarity[ ]index[ ](?P<similarity_index>\d+)%\n

           ^rename[ ]from[ ](?P<rename_from>.+)\n

           ^rename[ ]to[ ](?P<rename_to>.+)(?:\n|$))?

        (?:^new[ ]file[ ]mode[ ](?P<new_file_mode>.+)(?:\n|$))?

        (?:^deleted[ ]file[ ]mode[ ](?P<deleted_file_mode>.+)(?:\n|$))?

        (?:^index[ ](?P<a_blob_id>[0-9A-Fa-f]+)

            \.\.(?P<b_blob_id>[0-9A-Fa-f]+)[ ]?(?P<b_mode>.+)?(?:\n|$))?

        (?:^(?P<bin_patch>GIT[ ]binary[ ]patch)(?:\n|$))?

        (?:^---[ ](a/(?P<a_file>.+?)|/dev/null)\t?(?:\n|$))?

        (?:^\+\+\+[ ](b/(?P<b_file>.+?)|/dev/null)\t?(?:\n|$))?

    """, re.VERBOSE | re.MULTILINE)

    # Used for inline highlighter word split, must match the substitutions in _escaper

    _token_re = re.compile(r'()(&amp;|&lt;|&gt;|<u>\t</u>|<u class="cr"></u>| <i></i>|\W+?)')

    _escape_re = re.compile(r'(&)|(<)|(>)|(\t)|(\r)|(?<=.)( \n| $)')

    def __init__(self, diff, vcs='hg', diff_limit=None, inline_diff=True):

        """

        :param diff:   a text in diff format

        :param vcs: type of version control hg or git

        :param diff_limit: define the size of diff that is considered "big"

            based on that parameter cut off will be triggered, set to None

            to show full diff

        """

        if not isinstance(diff, basestring):

            raise Exception('Diff must be a basestring got %s instead' % type(diff))

        self._diff = diff

        self.adds = 0

        self.removes = 0

        # calculate diff size

        self.diff_limit = diff_limit

        self.cur_diff_size = 0

        self.vcs = vcs

        self.parsed = self._parse_gitdiff(inline_diff=inline_diff)

    def _escaper(self, string):

        """

        Do HTML escaping/markup and check the diff limit

        """

        self.cur_diff_size += len(string)

        # escaper gets iterated on each .next() call and it checks if each

        # parsed line doesn't exceed the diff limit

        if self.diff_limit is not None and self.cur_diff_size > self.diff_limit:

            raise DiffLimitExceeded('Diff Limit Exceeded')

        def substitute(m):

            groups = m.groups()

            if groups[0]:

                return '&'

            if groups[1]:

                return '<'

            if groups[2]:

                return '>'

            if groups[3]:

                return '<u>\t</u>'

        1. A dict with meta info:

            a_path, b_path, similarity_index, rename_from, rename_to,

            old_mode, new_mode, new_file_mode, deleted_file_mode,

            a_blob_id, b_blob_id, b_mode, a_file, b_file

        2. An iterator yielding lines with simple HTML markup.

        """

        match = None

        if self.vcs == 'git':

            match = self._git_header_re.match(diff_chunk)

        elif self.vcs == 'hg':

            match = self._hg_header_re.match(diff_chunk)

        if match is None:

            raise Exception('diff not recognized as valid %s diff' % self.vcs)

        meta_info = match.groupdict()

        rest = diff_chunk[match.end():]

        if rest and not rest.startswith('@') and not rest.startswith('literal ') and not rest.startswith('delta '):

            raise Exception('cannot parse %s diff header: %r followed by %r' % (self.vcs, diff_chunk[:match.end()], rest[:1000]))

        diff_lines = (self._escaper(m.group(0)) for m in re.finditer(r'.*\n|.+$', rest)) # don't split on \r as str.splitlines do

        return meta_info, diff_lines

    def _parse_gitdiff(self, inline_diff):

        """Parse self._diff and return a list of dicts with meta info and chunks for each file.

        Optionally, do an extra pass and to extra markup of one-liner changes.

        """

        _files = [] # list of dicts with meta info and chunks

        starts = [m.start() for m in self._diff_git_re.finditer(self._diff)]

        starts.append(len(self._diff))

        for start, end in zip(starts, starts[1:]):

            head, diff_lines = self._get_header(buffer(self._diff, start, end - start))

            op = None

            stats = {

                'added': 0,

                'deleted': 0,

                'binary': False,

                'ops': {},

            }

            if head['deleted_file_mode']:

                op = 'D'

                stats['binary'] = True

                stats['ops'][DEL_FILENODE] = 'deleted file'

            elif head['new_file_mode']:

                op = 'A'

                stats['binary'] = True

                stats['ops'][NEW_FILENODE] = 'new file %s' % head['new_file_mode']

                    elif head['a_file'] and not head['b_file']:

                        op = 'D'

                        stats['binary'] = True

                        stats['ops'][DEL_FILENODE] = 'deleted file'

                # it's not ADD not DELETE

                if op is None:

                    op = 'M'

                    stats['binary'] = True

                    stats['ops'][MOD_FILENODE] = 'modified file'

            # a real non-binary diff

            if head['a_file'] or head['b_file']:

                try:

                    chunks, added, deleted = self._parse_lines(diff_lines)

                    stats['binary'] = False

                    stats['added'] = added

                    stats['deleted'] = deleted

                    # explicit mark that it's a modified file

                    if op == 'M':

                        stats['ops'][MOD_FILENODE] = 'modified file'

                except DiffLimitExceeded:

                    break

            else:  # Git binary patch (or empty diff)

                # Git binary patch

                if head['bin_patch']:

                    stats['ops'][BIN_FILENODE] = 'binary diff not shown'

                chunks = []

            if op == 'D' and chunks:

                # a way of seeing deleted content could perhaps be nice - but

                # not with the current UI

                chunks = []

            chunks.insert(0, [{

                'old_lineno': '',

                'new_lineno': '',

                'action':     'context',

                'line':       msg,

                } for _op, msg in stats['ops'].iteritems()

                  if _op not in [MOD_FILENODE]])

            _files.append({

                'old_filename':     head['a_path'],

                'filename':         head['b_path'],

                'old_revision':     head['a_blob_id'],

                'new_revision':     head['b_blob_id'],

                'chunks':           chunks,

                'operation':        op,

                'stats':            stats,

            })

        if not inline_diff:

        # highlight inline changes when one del is followed by one add

        for diff_data in _files:

            for chunk in diff_data['chunks']:

                lineiter = iter(chunk)

                try:

                    peekline = lineiter.next()

                    while True:

                        # find a first del line

                        while peekline['action'] != 'del':

                            peekline = lineiter.next()

                        delline = peekline

                        peekline = lineiter.next()

                        # if not followed by add, eat all following del lines

                        if peekline['action'] != 'add':

                            while peekline['action'] == 'del':

                                peekline = lineiter.next()

                            continue

                        # found an add - make sure it is the only one

                        addline = peekline

                        try:

                            peekline = lineiter.next()

                        except StopIteration:

                            # add was last line - ok

                            self._highlight_inline_diff(delline, addline)

                            raise

                        if peekline['action'] != 'add':

                            # there was only one add line - ok

                            self._highlight_inline_diff(delline, addline)

                except StopIteration:

                    pass

    def _parse_lines(self, diff_lines):

        """

        Given an iterator of diff body lines, parse them and return a dict per

        line and added/removed totals.

        """

        added = deleted = 0

        old_line = old_end = new_line = new_end = None

        try:

            chunks = []

            line = diff_lines.next()

            while True:

                lines = []

                chunks.append(lines)

                match = self._chunk_re.match(line)

                if not match:

                    raise Exception('error parsing diff @@ line %r' % line)

                gr = match.groups()

                (old_line, old_end,