diffs: inline prepare() into __init__ and make the result available as .parsed

Make it more clear what the DiffProcessor is: Something that works on a raw
diff as input, mainly compute when initialized, and returns an object where the
result is available in different ways.

kallithea/controllers/changeset.py

@@ -1,460 +1,459 b''
 # -*- 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.changeset
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 changeset controller showing changes between revisions
 
 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 25, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 
 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.diffs import LimitedDiffContainer
 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])
         except ValueError:
             raise HTTPBadRequest()
     return ig_ws_global
 
 
 def _ignorews_url(GET, fileid=None):
     fileid = str(fileid) if fileid else None
     params = defaultdict(list)
     _update_with_GET(params, GET)
     lbl = _('Show whitespace')
     ig_ws = get_ignore_ws(fileid, GET)
     ln_ctx = get_line_ctx(fileid, GET)
     # global option
     if fileid is None:
         if ig_ws is None:
             params['ignorews'] += [1]
             lbl = _('Ignore whitespace')
         ctx_key = 'context'
         ctx_val = ln_ctx
     # per file options
     else:
         if ig_ws is None:
             params[fileid] += ['WS:1']
             lbl = _('Ignore whitespace')
 
         ctx_key = fileid
         ctx_val = 'C:%s' % ln_ctx
     # if we have passed in ln_ctx pass it along to our params
     if ln_ctx:
         params[ctx_key] += [ctx_val]
 
     params['anchor'] = fileid
     icon = h.literal('<i class="icon-strike"></i>')
     return h.link_to(icon, h.url.current(**params), title=lbl, **{'data-toggle': 'tooltip'})
 
 
 def get_line_ctx(fid, GET):
     ln_ctx_global = GET.get('context')
     if fid:
         ln_ctx = filter(lambda k: k.startswith('C'), GET.getall(fid))
     else:
         _ln_ctx = filter(lambda k: k.startswith('C'), GET)
         ln_ctx = GET.get(_ln_ctx[0]) if _ln_ctx else ln_ctx_global
         if ln_ctx:
             ln_ctx = [ln_ctx]
 
     if ln_ctx:
         retval = ln_ctx[0].split(':')[-1]
     else:
         retval = ln_ctx_global
 
     try:
         return int(retval)
     except Exception:
         return 3
 
 
 def _context_url(GET, fileid=None):
     """
     Generates url for context lines
 
     :param fileid:
     """
 
     fileid = str(fileid) if fileid else None
     ig_ws = get_ignore_ws(fileid, GET)
     ln_ctx = (get_line_ctx(fileid, GET) or 3) * 2
 
     params = defaultdict(list)
     _update_with_GET(params, GET)
 
     # global option
     if fileid is None:
         if ln_ctx > 0:
             params['context'] += [ln_ctx]
 
         if ig_ws:
             ig_ws_key = 'ignorews'
             ig_ws_val = 1
 
     # per file option
     else:
         params[fileid] += ['C:%s' % ln_ctx]
         ig_ws_key = fileid
         ig_ws_val = 'WS:%s' % 1
 
     if ig_ws:
         params[ig_ws_key] += [ig_ws_val]
 
     lbl = _('Increase diff context to %(num)s lines') % {'num': ln_ctx}
 
     params['anchor'] = fileid
     icon = h.literal('<i class="icon-sort"></i>')
     return h.link_to(icon, h.url.current(**params), title=lbl, **{'data-toggle': 'tooltip'})
 
 
 # Could perhaps be nice to have in the model but is too high level ...
 def create_comment(text, status, f_path, line_no, revision=None, pull_request_id=None, closing_pr=None):
     """Comment functionality shared between changesets and pullrequests"""
     f_path = f_path or None
     line_no = line_no or None
 
     comment = ChangesetCommentsModel().create(
         text=text,
         repo=c.db_repo.repo_id,
         author=request.authuser.user_id,
         revision=revision,
         pull_request=pull_request_id,
         f_path=f_path,
         line_no=line_no,
         status_change=ChangesetStatus.get_status_lbl(status) if status else None,
         closing_pr=closing_pr,
     )
 
     return comment
 
 
 class ChangesetController(BaseRepoController):
 
     def _before(self, *args, **kwargs):
         super(ChangesetController, self)._before(*args, **kwargs)
         c.affected_files_cut_off = 60
 
     def __load_data(self):
         repo_model = RepoModel()
         c.users_array = repo_model.get_users_js()
         c.user_groups_array = repo_model.get_user_groups_js()
 
     def _index(self, revision, method):
         c.pull_request = None
         c.anchor_url = anchor_url
         c.ignorews_url = _ignorews_url
         c.context_url = _context_url
         c.fulldiff = request.GET.get('fulldiff') # for reporting number of changed files
         # get ranges of revisions if preset
         rev_range = revision.split('...')[:2]
         enable_comments = True
         c.cs_repo = c.db_repo
         try:
             if len(rev_range) == 2:
                 enable_comments = False
                 rev_start = rev_range[0]
                 rev_end = rev_range[1]
                 rev_ranges = c.db_repo_scm_instance.get_changesets(start=rev_start,
                                                              end=rev_end)
             else:
                 rev_ranges = [c.db_repo_scm_instance.get_changeset(revision)]
 
             c.cs_ranges = list(rev_ranges)
             if not c.cs_ranges:
                 raise RepositoryError('Changeset range returned empty result')
 
         except (ChangesetDoesNotExistError, EmptyRepositoryError):
             log.debug(traceback.format_exc())
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
 
         c.changes = OrderedDict()
 
         c.lines_added = 0  # count of lines added
         c.lines_deleted = 0  # count of lines removes
 
         c.changeset_statuses = ChangesetStatus.STATUSES
         comments = dict()
         c.statuses = []
         c.inline_comments = []
         c.inline_cnt = 0
 
         # Iterate over ranges (default changeset view is always one changeset)
         for changeset in c.cs_ranges:
             if method == 'show':
                 c.statuses.extend([ChangesetStatusModel().get_status(
                             c.db_repo.repo_id, changeset.raw_id)])
 
                 # Changeset comments
                 comments.update((com.comment_id, com)
                                 for com in ChangesetCommentsModel()
                                 .get_comments(c.db_repo.repo_id,
                                               revision=changeset.raw_id))
 
                 # Status change comments - mostly from pull requests
                 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)
-                _parsed = diff_processor.prepare()
                 c.limited_diff = False
-                if isinstance(_parsed, LimitedDiffContainer):
+                if isinstance(diff_processor.parsed, LimitedDiffContainer):
                     c.limited_diff = True
-                for f in _parsed:
+                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:
             c.changeset = c.cs_ranges[0]
             c.parent_tmpl = ''.join(['# Parent  %s\n' % x.raw_id
                                      for x in c.changeset.parents])
         if method == 'download':
             response.content_type = 'text/plain'
             response.content_disposition = 'attachment; filename=%s.diff' \
                                             % revision[:12]
             return raw_diff
         elif method == 'patch':
             response.content_type = 'text/plain'
             c.diff = safe_unicode(raw_diff)
             return render('changeset/patch_changeset.html')
         elif method == 'raw':
             response.content_type = 'text/plain'
             return raw_diff
         elif method == 'show':
             self.__load_data()
             if len(c.cs_ranges) == 1:
                 return render('changeset/changeset.html')
             else:
                 c.cs_ranges_org = None
                 c.cs_comments = {}
                 revs = [ctx.revision for ctx in reversed(c.cs_ranges)]
                 c.jsdata = graph_data(c.db_repo_scm_instance, revs)
                 return render('changeset/changeset_range.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def index(self, revision, method='show'):
         return self._index(revision, method=method)
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def changeset_raw(self, revision):
         return self._index(revision, method='raw')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def changeset_patch(self, revision):
         return self._index(revision, method='patch')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def changeset_download(self, revision):
         return self._index(revision, method='download')
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def comment(self, repo_name, revision):
         assert request.environ.get('HTTP_X_PARTIAL_XHR')
 
         status = request.POST.get('changeset_status')
         text = request.POST.get('text', '').strip()
 
         c.comment = create_comment(
             text,
             status,
             revision=revision,
             f_path=request.POST.get('f_path'),
             line_no=request.POST.get('line'),
         )
 
         # get status if set !
         if status:
             # if latest status was from pull request and it's closed
             # disallow changing status ! RLY?
             try:
                 ChangesetStatusModel().set_status(
                     c.db_repo.repo_id,
                     status,
                     request.authuser.user_id,
                     c.comment,
                     revision=revision,
                     dont_allow_on_closed_pull_request=True,
                 )
             except StatusChangeOnClosedPullRequestError:
                 log.debug('cannot change status on %s with closed pull request', revision)
                 raise HTTPBadRequest()
 
         action_logger(request.authuser,
                       'user_commented_revision:%s' % revision,
                       c.db_repo, request.ip_addr)
 
         Session().commit()
 
         data = {
            'target_id': h.safeid(h.safe_unicode(request.POST.get('f_path'))),
         }
         if c.comment is not None:
             data.update(c.comment.get_dict())
             data.update({'rendered_text':
                          render('changeset/changeset_comment_block.html')})
 
         return data
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def delete_comment(self, repo_name, comment_id):
         co = ChangesetComment.get_or_404(comment_id)
         if co.repo.repo_name != repo_name:
             raise HTTPNotFound()
         owner = co.author_id == request.authuser.user_id
         repo_admin = h.HasRepoPermissionLevel('admin')(repo_name)
         if h.HasPermissionAny('hg.admin')() or repo_admin or owner:
             ChangesetCommentsModel().delete(comment=co)
             Session().commit()
             return True
         else:
             raise HTTPForbidden()
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def changeset_info(self, repo_name, revision):
         if request.is_xhr:
             try:
                 return c.db_repo_scm_instance.get_changeset(revision)
             except ChangesetDoesNotExistError as e:
                 return EmptyChangeset(message=str(e))
         else:
             raise HTTPBadRequest()
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def changeset_children(self, repo_name, revision):
         if request.is_xhr:
             changeset = c.db_repo_scm_instance.get_changeset(revision)
             result = {"results": []}
             if changeset.children:
                 result = {"results": changeset.children}
             return result
         else:
             raise HTTPBadRequest()
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def changeset_parents(self, repo_name, revision):
         if request.is_xhr:
             changeset = c.db_repo_scm_instance.get_changeset(revision)
             result = {"results": []}
             if changeset.parents:
                 result = {"results": changeset.parents}
             return result
         else:
             raise HTTPBadRequest()

kallithea/controllers/compare.py

@@ -1,294 +1,293 b''
 # -*- 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 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, 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.lib.diffs import LimitedDiffContainer
 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))
 
         # 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:
                 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
 
             # 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()
     @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()
     @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
         c.is_ajax_preview = partial and request.GET.get('is_ajax_preview')
         # swap url for compare_diff page - never partial and never is_ajax_preview
         c.swap_url = h.url('compare_url',
             repo_name=c.cs_repo.repo_name,
             org_ref_type=other_ref_type, org_ref_name=other_ref_name,
             other_repo=c.a_repo.repo_name,
             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'
         line_context = safe_int(request.GET.get('context'), 3)
 
         c.a_rev = self._get_ref_rev(c.a_repo, org_ref_type, org_ref_name,
             returnempty=True)
         c.cs_rev = self._get_ref_rev(c.cs_repo, other_ref_type, other_ref_name)
 
         c.compare_home = False
         c.a_ref_name = org_ref_name
         c.a_ref_type = org_ref_type
         c.cs_ref_name = other_ref_name
         c.cs_ref_type = other_ref_type
 
         c.cs_ranges, c.cs_ranges_org, c.ancestors = self._get_changesets(
             c.a_repo.scm_instance.alias, c.a_repo.scm_instance, c.a_rev,
             c.cs_repo.scm_instance, c.cs_rev)
         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)
 
         revs = [ctx.revision for ctx in reversed(c.cs_ranges)]
         c.jsdata = graph_data(c.cs_repo.scm_instance, revs)
 
         if partial:
             return render('compare/compare_cs.html')
 
         org_repo = c.a_repo
         other_repo = c.cs_repo
 
         if merge:
             rev1 = msg = None
             if not c.cs_ranges:
                 msg = _('Cannot show empty diff')
             elif not c.ancestors:
                 msg = _('No ancestor found for merge diff')
             elif len(c.ancestors) == 1:
                 rev1 = c.ancestors[0]
             else:
                 msg = _('Multiple merge ancestors found for merge compare')
             if rev1 is None:
                 h.flash(msg, category='error')
                 log.error(msg)
                 raise HTTPNotFound
 
             # 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)
-        txtdiff = org_repo.scm_instance.get_diff(rev1=rev1, rev2=c.cs_rev,
+        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(txtdiff or '', diff_limit=diff_limit)
-        _parsed = diff_processor.prepare()
+        diff_processor = diffs.DiffProcessor(raw_diff or '', diff_limit=diff_limit)
 
         c.limited_diff = False
-        if isinstance(_parsed, LimitedDiffContainer):
+        if isinstance(diff_processor.parsed, LimitedDiffContainer):
             c.limited_diff = True
 
         c.file_diff_data = []
         c.lines_added = 0
         c.lines_deleted = 0
-        for f in _parsed:
+        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')

kallithea/controllers/feed.py

@@ -1,170 +1,170 b''
 # -*- 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.feed
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 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.lib.diffs import DiffProcessor, LimitedDiffContainer
 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)
-        _parsed = diff_processor.prepare(inline_diff=False)
+                                       diff_limit=diff_limit,
+                                       inline_diff=False)
         limited_diff = False
-        if isinstance(_parsed, LimitedDiffContainer):
+        if isinstance(diff_processor.parsed, LimitedDiffContainer):
             limited_diff = True
 
-        for st in _parsed:
+        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)
         if limited_diff:
             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(
                 title=_('%s %s feed') % (c.site_name, repo_name),
                 link=h.canonical_url('summary_home', repo_name=repo_name),
                 description=_('Changes on %s repository') % repo_name,
                 language=language,
                 ttl=ttl
             )
 
             rss_items_per_page = safe_int(CONFIG.get('rss_items_per_page', 20))
             for cs in reversed(list(c.db_repo_scm_instance[-rss_items_per_page:])):
                 feed.add_item(title=self._get_title(cs),
                               link=h.canonical_url('changeset_home', repo_name=repo_name,
                                        revision=cs.raw_id),
                               author_name=cs.author,
                               description=''.join(self.__get_desc(cs)),
                               pubdate=cs.date,
                               )
 
             response.content_type = feed.mime_type
             return feed.writeString('utf-8')
 
         kind = 'ATOM'
         valid = CacheInvalidation.test_and_set_valid(repo_name, kind)
         if not valid:
             region_invalidate(_get_feed_from_cache, None, '_get_feed_from_cache', repo_name, kind)
         return _get_feed_from_cache(repo_name, kind)
 
     def rss(self, repo_name):
         """Produce an rss2 feed via feedgenerator module"""
 
         @cache_region('long_term', '_get_feed_from_cache')
         def _get_feed_from_cache(key, kind):
             feed = Rss201rev2Feed(
                 title=_('%s %s feed') % (c.site_name, repo_name),
                 link=h.canonical_url('summary_home', repo_name=repo_name),
                 description=_('Changes on %s repository') % repo_name,
                 language=language,
                 ttl=ttl
             )
 
             rss_items_per_page = safe_int(CONFIG.get('rss_items_per_page', 20))
             for cs in reversed(list(c.db_repo_scm_instance[-rss_items_per_page:])):
                 feed.add_item(title=self._get_title(cs),
                               link=h.canonical_url('changeset_home', repo_name=repo_name,
                                        revision=cs.raw_id),
                               author_name=cs.author,
                               description=''.join(self.__get_desc(cs)),
                               pubdate=cs.date,
                              )
 
             response.content_type = feed.mime_type
             return feed.writeString('utf-8')
 
         kind = 'RSS'
         valid = CacheInvalidation.test_and_set_valid(repo_name, kind)
         if not valid:
             region_invalidate(_get_feed_from_cache, None, '_get_feed_from_cache', repo_name, kind)
         return _get_feed_from_cache(repo_name, kind)

kallithea/controllers/pullrequests.py

@@ -1,746 +1,745 b''
 # -*- 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.pullrequests
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 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.diffs import LimitedDiffContainer
 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."""
     try:
         user = User.get(int(user_id))
     except ValueError:
         user = None
 
     if user is None or user.is_default_user:
         h.flash(_('Invalid reviewer "%s" specified') % user_id, category='error')
         raise HTTPBadRequest()
 
     return user
 
 
 class PullrequestsController(BaseRepoController):
 
     def _get_repo_refs(self, repo, rev=None, branch=None, branch_rev=None):
         """return a structure with repo's interesting changesets, suitable for
         the selectors in pullrequest.html
 
         rev: a revision that must be in the list somehow and selected by default
         branch: a branch that must be in the list and selected by default - even if closed
         branch_rev: a revision of which peers should be preferred and available."""
         # list named branches that has been merged to this named branch - it should probably merge back
         peers = []
 
         if rev:
             rev = safe_str(rev)
 
         if branch:
             branch = safe_str(branch)
 
         if branch_rev:
             branch_rev = safe_str(branch_rev)
             # a revset not restricting to merge() would be better
             # (especially because it would get the branch point)
             # ... but is currently too expensive
             # including branches of children could be nice too
             peerbranches = set()
             for i in repo._repo.revs(
                 "sort(parents(branch(id(%s)) and merge()) - branch(id(%s)), -rev)",
                 branch_rev, branch_rev):
                 abranch = repo.get_changeset(i).branch
                 if abranch not in peerbranches:
                     n = 'branch:%s:%s' % (abranch, repo.get_changeset(abranch).raw_id)
                     peers.append((n, abranch))
                     peerbranches.add(abranch)
 
         selected = None
         tiprev = repo.tags.get('tip')
         tipbranch = None
 
         branches = []
         for abranch, branchrev in repo.branches.iteritems():
             n = 'branch:%s:%s' % (abranch, branchrev)
             desc = abranch
             if branchrev == tiprev:
                 tipbranch = abranch
                 desc = '%s (current tip)' % desc
             branches.append((n, desc))
             if rev == branchrev:
                 selected = n
             if branch == abranch:
                 if not rev:
                     selected = n
                 branch = None
         if branch:  # branch not in list - it is probably closed
             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:
                     k, v = repo.branches.items()[0]
                     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 _get_is_allowed_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()
     @HasRepoPermissionLevelDecorator('read')
     def show_all(self, repo_name):
         c.from_ = request.GET.get('from_') or ''
         c.closed = request.GET.get('closed') or ''
         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)
 
         return render('/pullrequests/pullrequest_show_all.html')
 
     @LoginRequired()
     @NotAnonymous()
     def show_my(self):
         c.closed = request.GET.get('closed') or ''
 
         c.my_pull_requests = PullRequest.query(
             include_closed=c.closed,
             sorted=True,
         ).filter_by(owner_id=request.authuser.user_id).all()
 
         c.participate_in_pull_requests = []
         c.participate_in_pull_requests_todo = []
         done_status = set([ChangesetStatus.STATUS_APPROVED, ChangesetStatus.STATUS_REJECTED])
         for pr in PullRequest.query(
             include_closed=c.closed,
             reviewer_id=request.authuser.user_id,
             sorted=True,
         ):
             status = pr.user_review_status(request.authuser.user_id) # very inefficient!!!
             if status in done_status:
                 c.participate_in_pull_requests.append(pr)
             else:
                 c.participate_in_pull_requests_todo.append(pr)
 
         return render('/pullrequests/pullrequest_show_my.html')
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     def index(self):
         org_repo = c.db_repo
         org_scm_instance = org_repo.scm_instance
         try:
             org_scm_instance.get_changeset()
         except EmptyRepositoryError as e:
             h.flash(h.literal(_('There are no changesets yet')),
                     category='warning')
             raise HTTPFound(location=url('summary_home', repo_name=org_repo.repo_name))
 
         org_rev = request.GET.get('rev_end')
         # rev_start is not directly useful - its parent could however be used
         # as default for other and thus give a simple compare view
         rev_start = request.GET.get('rev_start')
         other_rev = None
         if rev_start:
             starters = org_repo.get_changeset(rev_start).parents
             if starters:
                 other_rev = starters[0].raw_id
             else:
                 other_rev = org_repo.scm_instance.EMPTY_CHANGESET
         branch = request.GET.get('branch')
 
         c.cs_repos = [(org_repo.repo_name, org_repo.repo_name)]
         c.default_cs_repo = org_repo.repo_name
         c.cs_refs, c.default_cs_ref = self._get_repo_refs(org_scm_instance, rev=org_rev, branch=branch)
 
         default_cs_ref_type, default_cs_branch, default_cs_rev = c.default_cs_ref.split(':')
         if default_cs_ref_type != 'branch':
             default_cs_branch = org_repo.get_changeset(default_cs_rev).branch
 
         # add org repo to other so we can open pull request against peer branches on itself
         c.a_repos = [(org_repo.repo_name, '%s (self)' % org_repo.repo_name)]
 
         if org_repo.parent:
             # add parent of this fork also and select it.
             # use the same branch on destination as on source, if available.
             c.a_repos.append((org_repo.parent.repo_name, '%s (parent)' % org_repo.parent.repo_name))
             c.a_repo = org_repo.parent
             c.a_refs, c.default_a_ref = self._get_repo_refs(
                     org_repo.parent.scm_instance, branch=default_cs_branch, rev=other_rev)
 
         else:
             c.a_repo = org_repo
             c.a_refs, c.default_a_ref = self._get_repo_refs(org_scm_instance, rev=other_rev)
 
         # gather forks and add to this list ... even though it is rare to
         # request forks to pull from their parent
         for fork in org_repo.forks:
             c.a_repos.append((fork.repo_name, fork.repo_name))
 
         return render('/pullrequests/pullrequest.html')
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def repo_info(self, repo_name):
         repo = c.db_repo
         refs, selected_ref = self._get_repo_refs(repo.scm_instance)
         return {
             'description': repo.description.split('\n', 1)[0],
             'selected_ref': selected_ref,
             'refs': refs,
             }
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     def create(self, repo_name):
         repo = c.db_repo
         try:
             _form = PullRequestForm(repo.repo_id)().to_python(request.POST)
         except formencode.Invalid as errors:
             log.error(traceback.format_exc())
             log.error(str(errors))
             msg = _('Error creating pull request: %s') % errors.msg
             h.flash(msg, 'error')
             raise HTTPBadRequest
 
         # heads up: org and other might seem backward here ...
         org_ref = _form['org_ref'] # will have merge_rev as rev but symbolic name
         org_repo = Repository.guess_instance(_form['org_repo'])
 
         other_ref = _form['other_ref'] # will have symbolic name and head revision
         other_repo = Repository.guess_instance(_form['other_repo'])
 
         reviewers = []
 
         title = _form['pullrequest_title']
         description = _form['pullrequest_desc'].strip()
         owner = User.get(request.authuser.user_id)
 
         try:
             cmd = CreatePullRequestAction(org_repo, other_repo, org_ref, other_ref, title, description, owner, reviewers)
         except CreatePullRequestAction.ValidationError as e:
             h.flash(str(e), category='error', logf=log.error)
             raise HTTPNotFound
 
         try:
             pull_request = cmd.execute()
             Session().commit()
         except Exception:
             h.flash(_('Error occurred while creating pull request'),
                     category='error')
             log.error(traceback.format_exc())
             raise HTTPFound(location=url('pullrequest_home', repo_name=repo_name))
 
         h.flash(_('Successfully opened new pull request'),
                 category='success')
         raise HTTPFound(location=pull_request.url())
 
     def create_new_iteration(self, old_pull_request, new_rev, title, description, reviewers):
         owner = User.get(request.authuser.user_id)
         new_org_rev = self._get_ref_rev(old_pull_request.org_repo, 'rev', new_rev)
         new_other_rev = self._get_ref_rev(old_pull_request.other_repo, old_pull_request.other_ref_parts[0], old_pull_request.other_ref_parts[1])
         try:
             cmd = CreatePullRequestIterationAction(old_pull_request, new_org_rev, new_other_rev, title, description, owner, reviewers)
         except CreatePullRequestAction.ValidationError as e:
             h.flash(str(e), category='error', logf=log.error)
             raise HTTPNotFound
 
         try:
             pull_request = cmd.execute()
             Session().commit()
         except Exception:
             h.flash(_('Error occurred while creating pull request'),
                     category='error')
             log.error(traceback.format_exc())
             raise HTTPFound(location=old_pull_request.url())
 
         h.flash(_('New pull request iteration created'),
                 category='success')
         raise HTTPFound(location=pull_request.url())
 
     # pullrequest_post for PR editing
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     def post(self, repo_name, pull_request_id):
         pull_request = PullRequest.get_or_404(pull_request_id)
         if pull_request.is_closed():
             raise HTTPForbidden()
         assert pull_request.other_repo.repo_name == repo_name
         # only owner or admin can update it
         owner = pull_request.owner_id == request.authuser.user_id
         repo_admin = h.HasRepoPermissionLevel('admin')(c.repo_name)
         if not (h.HasPermissionAny('hg.admin')() or repo_admin or owner):
             raise HTTPForbidden()
 
         _form = PullRequestPostForm()().to_python(request.POST)
 
         cur_reviewers = set(pull_request.get_reviewer_users())
         new_reviewers = set(_get_reviewer(s) for s in _form['review_members'])
         old_reviewers = set(_get_reviewer(s) for s in _form['org_review_members'])
 
         other_added = cur_reviewers - old_reviewers
         other_removed = old_reviewers - cur_reviewers
 
         if other_added:
             h.flash(_('Meanwhile, the following reviewers have been added: %s') %
                     (', '.join(u.username for u in other_added)),
                     category='warning')
         if other_removed:
             h.flash(_('Meanwhile, the following reviewers have been removed: %s') %
                     (', '.join(u.username for u in other_removed)),
                     category='warning')
 
         if _form['updaterev']:
             return self.create_new_iteration(pull_request,
                                       _form['updaterev'],
                                       _form['pullrequest_title'],
                                       _form['pullrequest_desc'],
                                       new_reviewers)
 
         added_reviewers = new_reviewers - old_reviewers - cur_reviewers
         removed_reviewers = (old_reviewers - new_reviewers) & cur_reviewers
 
         old_description = pull_request.description
         pull_request.title = _form['pullrequest_title']
         pull_request.description = _form['pullrequest_desc'].strip() or _('No description')
         pull_request.owner = User.get_by_username(_form['owner'])
         user = User.get(request.authuser.user_id)
 
         PullRequestModel().mention_from_description(user, pull_request, old_description)
         PullRequestModel().add_reviewers(user, pull_request, added_reviewers)
         PullRequestModel().remove_reviewers(user, pull_request, removed_reviewers)
 
         Session().commit()
         h.flash(_('Pull request updated'), category='success')
 
         raise HTTPFound(location=pull_request.url())
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     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()
     @HasRepoPermissionLevelDecorator('read')
     def show(self, repo_name, pull_request_id, extra=None):
         repo_model = RepoModel()
         c.users_array = repo_model.get_users_js()
         c.user_groups_array = repo_model.get_user_groups_js()
         c.pull_request = PullRequest.get_or_404(pull_request_id)
         c.allowed_to_change_status = self._get_is_allowed_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
                             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 additional changes are available on %s:') % c.cs_branch_name
                     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:
-            txtdiff = org_scm_instance.get_diff(rev1=safe_str(c.a_rev), rev2=safe_str(c.cs_rev),
+            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:
-            txtdiff = _("The diff can't be shown - the PR revisions could not be found.")
-        diff_processor = diffs.DiffProcessor(txtdiff or '', diff_limit=diff_limit)
-        _parsed = diff_processor.prepare()
+            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 = False
-        if isinstance(_parsed, LimitedDiffContainer):
+        if isinstance(diff_processor.parsed, LimitedDiffContainer):
             c.limited_diff = True
 
         c.file_diff_data = []
         c.lines_added = 0
         c.lines_deleted = 0
 
-        for f in _parsed:
+        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
         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()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def comment(self, repo_name, pull_request_id):
         pull_request = PullRequest.get_or_404(pull_request_id)
 
         status = request.POST.get('changeset_status')
         close_pr = request.POST.get('save_close')
         delete = request.POST.get('save_delete')
         f_path = request.POST.get('f_path')
         line_no = request.POST.get('line')
 
         if (status or close_pr or delete) and (f_path or line_no):
             # status votes and closing is only possible in general comments
             raise HTTPBadRequest()
 
         allowed_to_change_status = self._get_is_allowed_change_status(pull_request)
         if not allowed_to_change_status:
             if status or close_pr:
                 h.flash(_('No permission to change pull request status'), 'error')
                 raise HTTPForbidden()
 
         if delete == "delete":
             if (pull_request.owner_id == request.authuser.user_id or
                 h.HasPermissionAny('hg.admin')() or
                 h.HasRepoPermissionLevel('admin')(pull_request.org_repo.repo_name) or
                 h.HasRepoPermissionLevel('admin')(pull_request.other_repo.repo_name)
                 ) and not pull_request.is_closed():
                 PullRequestModel().delete(pull_request)
                 Session().commit()
                 h.flash(_('Successfully deleted pull request %s') % pull_request_id,
                         category='success')
                 return {
                    'location': url('my_pullrequests'), # or repo pr list?
                 }
                 raise HTTPFound(location=url('my_pullrequests')) # or repo pr list?
             raise HTTPForbidden()
 
         text = request.POST.get('text', '').strip()
 
         comment = create_comment(
             text,
             status,
             pull_request_id=pull_request_id,
             f_path=f_path,
             line_no=line_no,
             closing_pr=close_pr,
         )
 
         action_logger(request.authuser,
                       'user_commented_pull_request:%s' % pull_request_id,
                       c.db_repo, request.ip_addr)
 
         if status:
             ChangesetStatusModel().set_status(
                 c.db_repo.repo_id,
                 status,
                 request.authuser.user_id,
                 comment,
                 pull_request=pull_request_id
             )
 
         if close_pr:
             PullRequestModel().close_pull_request(pull_request_id)
             action_logger(request.authuser,
                           'user_closed_pull_request:%s' % pull_request_id,
                           c.db_repo, request.ip_addr)
 
         Session().commit()
 
         if not request.environ.get('HTTP_X_PARTIAL_XHR'):
             raise HTTPFound(location=pull_request.url())
 
         data = {
            'target_id': h.safeid(h.safe_unicode(request.POST.get('f_path'))),
         }
         if comment is not None:
             c.comment = comment
             data.update(comment.get_dict())
             data.update({'rendered_text':
                          render('changeset/changeset_comment_block.html')})
 
         return data
 
     @LoginRequired()
     @NotAnonymous()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def delete_comment(self, repo_name, comment_id):
         co = ChangesetComment.get(comment_id)
         if co.pull_request.is_closed():
             # don't allow deleting comments on closed pull request
             raise HTTPForbidden()
 
         owner = co.author_id == request.authuser.user_id
         repo_admin = h.HasRepoPermissionLevel('admin')(c.repo_name)
         if h.HasPermissionAny('hg.admin')() or repo_admin or owner:
             ChangesetCommentsModel().delete(comment=co)
             Session().commit()
             return True
         else:
             raise HTTPForbidden()

kallithea/lib/diffs.py

@@ -1,700 +1,685 b''
 # -*- 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.lib.diffs
 ~~~~~~~~~~~~~~~~~~~
 
 Set of diffing helpers, previously part of vcs
 
 
 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: Dec 4, 2011
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import re
 import difflib
 import logging
 
 from tg.i18n import ugettext as _
 
 from kallithea.lib.vcs.exceptions import VCSError
 from kallithea.lib.vcs.nodes import FileNode, SubModuleNode
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.helpers import escape
 from kallithea.lib.utils2 import safe_unicode
 
 log = logging.getLogger(__name__)
 
 
 def wrap_to_table(html):
     """Given a string with html, return it wrapped in a table, similar to what
     DiffProcessor returns."""
     return '''\
               <table class="code-difftable">
                 <tr class="line no-comment">
                 <td class="lineno new"></td>
                 <td class="code no-comment"><pre>%s</pre></td>
                 </tr>
               </table>''' % html
 
 
 def wrapped_diff(filenode_old, filenode_new, diff_limit=None,
                 ignore_whitespace=True, line_context=3,
                 enable_comments=False):
     """
     Returns a file diff wrapped into a table.
     Checks for diff_limit and presents a message if the diff is too big.
     """
     if filenode_old is None:
         filenode_old = FileNode(filenode_new.path, '', EmptyChangeset())
 
     op = None
     a_path = filenode_old.path # default, might be overriden by actual rename in diff
     if filenode_old.is_binary or filenode_new.is_binary:
         diff = wrap_to_table(_('Binary file'))
         stats = (0, 0)
 
     elif diff_limit != -1 and (
             diff_limit is None or
             (filenode_old.size < diff_limit and filenode_new.size < diff_limit)):
 
-        f_gitdiff = get_gitdiff(filenode_old, filenode_new,
+        raw_diff = get_gitdiff(filenode_old, filenode_new,
                                 ignore_whitespace=ignore_whitespace,
                                 context=line_context)
-        diff_processor = DiffProcessor(f_gitdiff)
-        _parsed = diff_processor.prepare()
-        if _parsed: # there should be exactly one element, for the specified file
-            f = _parsed[0]
+        diff_processor = DiffProcessor(raw_diff)
+        if diff_processor.parsed: # there should be exactly one element, for the specified file
+            f = diff_processor.parsed[0]
             op = f['operation']
             a_path = f['old_filename']
 
         diff = diff_processor.as_html(enable_comments=enable_comments)
         stats = diff_processor.stat()
 
     else:
         diff = wrap_to_table(_('Changeset was too big and was cut off, use '
                                'diff menu to display this diff'))
         stats = (0, 0)
 
     if not diff:
         submodules = filter(lambda o: isinstance(o, SubModuleNode),
                             [filenode_new, filenode_old])
         if submodules:
             diff = wrap_to_table(escape('Submodule %r' % submodules[0]))
         else:
             diff = wrap_to_table(_('No changes detected'))
 
     cs1 = filenode_old.changeset.raw_id
     cs2 = filenode_new.changeset.raw_id
 
     return cs1, cs2, a_path, diff, stats, op
 
 
 def get_gitdiff(filenode_old, filenode_new, ignore_whitespace=True, context=3):
     """
     Returns git style diff between given ``filenode_old`` and ``filenode_new``.
     """
     # make sure we pass in default context
     context = context or 3
     submodules = filter(lambda o: isinstance(o, SubModuleNode),
                         [filenode_new, filenode_old])
     if submodules:
         return ''
 
     for filenode in (filenode_old, filenode_new):
         if not isinstance(filenode, FileNode):
             raise VCSError("Given object should be FileNode object, not %s"
                 % 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 LimitedDiffContainer(object):
 
     def __init__(self, diff_limit, cur_diff_size, diff):
         self.diff = diff
         self.diff_limit = diff_limit
         self.cur_diff_size = cur_diff_size
 
     def __iter__(self):
         for l in self.diff:
             yield l
 
 
 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)
     _hg_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|$))?
         (?:^copy[ ]from[ ](?P<copy_from>.+)\n
            ^copy[ ]to[ ](?P<copy_to>.+)(?:\n|$))?
         (?:^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):
+    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_size = len(diff)
         self.diff_limit = diff_limit
         self.cur_diff_size = 0
-        self.parsed = False
-        self.parsed_diff = []
         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 '&amp;'
             if groups[1]:
                 return '&lt;'
             if groups[2]:
                 return '&gt;'
             if groups[3]:
                 return '<u>\t</u>'
             if groups[4]:
                 return '<u class="cr"></u>'
             if groups[5]:
                 return ' <i></i>'
             assert False
 
         return self._escape_re.sub(substitute, safe_unicode(string))
 
     def _highlight_inline_diff(self, old, new):
         """
         Highlight simple add/remove in two lines given as info dicts. They are
         modified in place and given markup with <del>/<ins>.
         """
         assert old['action'] == 'del'
         assert new['action'] == 'add'
 
         oldwords = self._token_re.split(old['line'])
         newwords = self._token_re.split(new['line'])
         sequence = difflib.SequenceMatcher(None, oldwords, newwords)
 
         oldfragments, newfragments = [], []
         for tag, i1, i2, j1, j2 in sequence.get_opcodes():
             oldfrag = ''.join(oldwords[i1:i2])
             newfrag = ''.join(newwords[j1:j2])
             if tag != 'equal':
                 if oldfrag:
                     oldfrag = '<del>%s</del>' % oldfrag
                 if newfrag:
                     newfrag = '<ins>%s</ins>' % newfrag
             oldfragments.append(oldfrag)
             newfragments.append(newfrag)
 
         old['line'] = "".join(oldfragments)
         new['line'] = "".join(newfragments)
 
     def _get_header(self, diff_chunk):
         """
         Parses a Git diff for a single file (header and chunks) and returns a tuple with:
 
         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=True):
+    def _parse_gitdiff(self, inline_diff):
         """Parse self._diff and return a list of dicts with meta info and chunks for each file.
         If diff is truncated, wrap it in LimitedDiffContainer.
         Optionally, do an extra pass and to extra markup of one-liner changes.
         """
         _files = [] # list of dicts with meta info and chunks
         diff_container = lambda arg: arg
 
         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']
             else:  # modify operation, can be cp, rename, chmod
                 # CHMOD
                 if head['new_mode'] and head['old_mode']:
                     op = 'M'
                     stats['binary'] = True
                     stats['ops'][CHMOD_FILENODE] = ('modified file chmod %s => %s'
                                         % (head['old_mode'], head['new_mode']))
                 # RENAME
                 if (head['rename_from'] and head['rename_to']
                       and head['rename_from'] != head['rename_to']):
                     op = 'R'
                     stats['binary'] = True
                     stats['ops'][RENAMED_FILENODE] = ('file renamed from %s to %s'
                                     % (head['rename_from'], head['rename_to']))
                 # COPY
                 if head.get('copy_from') and head.get('copy_to'):
                     op = 'M'
                     stats['binary'] = True
                     stats['ops'][COPIED_FILENODE] = ('file copied from %s to %s'
                                         % (head['copy_from'], head['copy_to']))
                 # FALL BACK: detect missed old style add or remove
                 if op is None:
                     if not head['a_file'] and head['b_file']:
                         op = 'A'
                         stats['binary'] = True
                         stats['ops'][NEW_FILENODE] = 'new file'
 
                     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:
                     diff_container = lambda _diff: \
                         LimitedDiffContainer(self.diff_limit,
                                             self.cur_diff_size, _diff)
                     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:
             return diff_container(_files)
 
         # 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
 
         return diff_container(_files)
 
     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,
                  new_line, new_end) = [int(x or 1) for x in gr[:-1]]
                 old_line -= 1
                 new_line -= 1
 
                 context = len(gr) == 5
                 old_end += old_line
                 new_end += new_line
 
                 if context:
                     # skip context only if it's first line
                     if int(gr[0]) > 1:
                         lines.append({
                             'old_lineno': '...',
                             'new_lineno': '...',
                             'action':     'context',
                             'line':       line,
                         })
 
                 line = diff_lines.next()
 
                 while old_line < old_end or new_line < new_end:
                     if not line:
                         raise Exception('error parsing diff - empty line at -%s+%s' % (old_line, new_line))
 
                     affects_old = affects_new = False
 
                     command = line[0]
                     if command == '+':
                         affects_new = True
                         action = 'add'
                         added += 1
                     elif command == '-':
                         affects_old = True
                         action = 'del'
                         deleted += 1
                     elif command == ' ':
                         affects_old = affects_new = True
                         action = 'unmod'
                     else:
                         raise Exception('error parsing diff - unknown command in line %r at -%s+%s' % (line, old_line, new_line))
 
                     if not self._newline_marker.match(line):
                         old_line += affects_old
                         new_line += affects_new
                         lines.append({
                             'old_lineno':   affects_old and old_line or '',
                             'new_lineno':   affects_new and new_line or '',
                             'action':       action,
                             'line':         line[1:],
                         })
 
                     line = diff_lines.next()
 
                     if self._newline_marker.match(line):
                         # we need to append to lines, since this is not
                         # counted in the line specs of diff
                         lines.append({
                             'old_lineno':   '...',
                             'new_lineno':   '...',
                             'action':       'context',
                             'line':         line,
                         })
                         line = diff_lines.next()
                 if old_line > old_end:
                     raise Exception('error parsing diff - more than %s "-" lines at -%s+%s' % (old_end, old_line, new_line))
                 if new_line > new_end:
                     raise Exception('error parsing diff - more than %s "+" lines at -%s+%s' % (new_end, old_line, new_line))
         except StopIteration:
             pass
         if old_line != old_end or new_line != new_end:
             raise Exception('diff processing broken when old %s<>%s or new %s<>%s line %r' % (old_line, old_end, new_line, new_end, line))
 
         return chunks, added, deleted
 
     def _safe_id(self, idstring):
         """Make a string safe for including in an id attribute.
 
         The HTML spec says that id attributes 'must begin with
         a letter ([A-Za-z]) and may be followed by any number
         of letters, digits ([0-9]), hyphens ("-"), underscores
         ("_"), colons (":"), and periods (".")'. These regexps
         are slightly over-zealous, in that they remove colons
         and periods unnecessarily.
 
         Whitespace is transformed into underscores, and then
         anything which is not a hyphen or a character that
         matches \w (alphanumerics and underscore) is removed.
 
         """
         # Transform all whitespace to underscore
         idstring = re.sub(r'\s', "_", idstring)
         # Remove everything that is not a hyphen or a member of \w
         idstring = re.sub(r'(?!-)\W', "", idstring).lower()
         return idstring
 
-    def prepare(self, inline_diff=True):
-        """
-        Prepare the passed udiff for HTML rendering. It'll return a list
-        of dicts with diff information
-        """
-        parsed = self._parse_gitdiff(inline_diff=inline_diff)
-        self.parsed = True
-        self.parsed_diff = parsed
-        return parsed
-
     def as_html(self, table_class='code-difftable', line_class='line',
                 old_lineno_class='lineno old', new_lineno_class='lineno new',
                 no_lineno_class='lineno',
                 code_class='code', enable_comments=False, parsed_lines=None):
         """
         Return given diff as html table with customized css classes
         """
         def _link_to_if(condition, label, url):
             """
             Generates a link if condition is meet or just the label if not.
             """
 
             if condition:
                 return '''<a href="%(url)s">%(label)s</a>''' % {
                     'url': url,
                     'label': label
                 }
             else:
                 return label
-        if not self.parsed:
-            self.prepare()
 
-        diff_lines = self.parsed_diff
+        diff_lines = self.parsed
         if parsed_lines:
             diff_lines = parsed_lines
 
         _html_empty = True
         _html = []
         _html.append('''<table class="%(table_class)s">\n''' % {
             'table_class': table_class
         })
 
         for diff in diff_lines:
             for line in diff['chunks']:
                 _html_empty = False
                 for change in line:
                     _html.append('''<tr class="%(lc)s %(action)s">\n''' % {
                         'lc': line_class,
                         'action': change['action']
                     })
                     anchor_old_id = ''
                     anchor_new_id = ''
                     anchor_old = "%(filename)s_o%(oldline_no)s" % {
                         'filename': self._safe_id(diff['filename']),
                         'oldline_no': change['old_lineno']
                     }
                     anchor_new = "%(filename)s_n%(oldline_no)s" % {
                         'filename': self._safe_id(diff['filename']),
                         'oldline_no': change['new_lineno']
                     }
                     cond_old = (change['old_lineno'] != '...' and
                                 change['old_lineno'])
                     cond_new = (change['new_lineno'] != '...' and
                                 change['new_lineno'])
                     no_lineno = (change['old_lineno'] == '...' and
                                  change['new_lineno'] == '...')
                     if cond_old:
                         anchor_old_id = 'id="%s"' % anchor_old
                     if cond_new:
                         anchor_new_id = 'id="%s"' % anchor_new
                     ###########################################################
                     # OLD LINE NUMBER
                     ###########################################################
                     _html.append('''\t<td %(a_id)s class="%(olc)s" %(colspan)s>''' % {
                         'a_id': anchor_old_id,
                         'olc': no_lineno_class if no_lineno else old_lineno_class,
                         'colspan': 'colspan="2"' if no_lineno else ''
                     })
 
                     _html.append('''%(link)s''' % {
                         'link': _link_to_if(not no_lineno, change['old_lineno'],
                                             '#%s' % anchor_old)
                     })
                     _html.append('''</td>\n''')
                     ###########################################################
                     # NEW LINE NUMBER
                     ###########################################################
 
                     if not no_lineno:
                         _html.append('''\t<td %(a_id)s class="%(nlc)s">''' % {
                             'a_id': anchor_new_id,
                             'nlc': new_lineno_class
                         })
 
                         _html.append('''%(link)s''' % {
                             'link': _link_to_if(True, change['new_lineno'],
                                                 '#%s' % anchor_new)
                         })
                         _html.append('''</td>\n''')
                     ###########################################################
                     # CODE
                     ###########################################################
                     comments = '' if enable_comments else 'no-comment'
                     _html.append('''\t<td class="%(cc)s %(inc)s">''' % {
                         'cc': code_class,
                         'inc': comments
                     })
                     _html.append('''\n\t\t<div class="add-bubble"><div>&nbsp;</div></div><pre>%(code)s</pre>\n''' % {
                         'code': change['line']
                     })
 
                     _html.append('''\t</td>''')
                     _html.append('''\n</tr>\n''')
         _html.append('''</table>''')
         if _html_empty:
             return None
         return ''.join(_html)
 
     def stat(self):
         """
         Returns tuple of added, and removed lines for this instance
         """
         return self.adds, self.removes

kallithea/tests/models/test_diff_parsers.py

@@ -1,317 +1,315 b''
 from kallithea.tests.base import *
 from kallithea.lib.diffs import DiffProcessor, NEW_FILENODE, DEL_FILENODE, \
     MOD_FILENODE, RENAMED_FILENODE, CHMOD_FILENODE, BIN_FILENODE, COPIED_FILENODE
 from kallithea.tests.fixture import Fixture
 
 fixture = Fixture()
 
 
 DIFF_FIXTURES = {
     'hg_diff_add_single_binary_file.diff': [
         ('US Warszawa.jpg', 'A',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {NEW_FILENODE: 'new file 100755',
                   BIN_FILENODE: 'binary diff not shown'}}),
     ],
     'hg_diff_mod_single_binary_file.diff': [
         ('US Warszawa.jpg', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {MOD_FILENODE: 'modified file',
                   BIN_FILENODE: 'binary diff not shown'}}),
     ],
 
     'hg_diff_mod_single_file_and_rename_and_chmod.diff': [
         ('README', 'R',
          {'added': 3,
           'deleted': 0,
           'binary': False,
           'ops': {RENAMED_FILENODE: 'file renamed from README.rst to README',
                   CHMOD_FILENODE: 'modified file chmod 100755 => 100644'}}),
     ],
     'hg_diff_mod_file_and_rename.diff': [
         ('README.rst', 'R',
          {'added': 3,
           'deleted': 0,
           'binary': False,
           'ops': {RENAMED_FILENODE: 'file renamed from README to README.rst'}}),
     ],
     'hg_diff_del_single_binary_file.diff': [
         ('US Warszawa.jpg', 'D',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {DEL_FILENODE: 'deleted file',
                   BIN_FILENODE: 'binary diff not shown'}}),
     ],
     'hg_diff_chmod_and_mod_single_binary_file.diff': [
         ('gravatar.png', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {CHMOD_FILENODE: 'modified file chmod 100644 => 100755',
                   BIN_FILENODE: 'binary diff not shown'}}),
     ],
     'hg_diff_chmod.diff': [
         ('file', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {CHMOD_FILENODE: 'modified file chmod 100755 => 100644'}}),
     ],
     'hg_diff_rename_file.diff': [
         ('file_renamed', 'R',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {RENAMED_FILENODE: 'file renamed from file to file_renamed'}}),
     ],
     'hg_diff_rename_and_chmod_file.diff': [
         ('README', 'R',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {CHMOD_FILENODE: 'modified file chmod 100644 => 100755',
                   RENAMED_FILENODE: 'file renamed from README.rst to README'}}),
     ],
     'hg_diff_binary_and_normal.diff': [
         ('img/baseline-10px.png', 'A',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {NEW_FILENODE: 'new file 100644',
                   BIN_FILENODE: 'binary diff not shown'}}),
         ('img/baseline-20px.png', 'D',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {DEL_FILENODE: 'deleted file',
                   BIN_FILENODE: 'binary diff not shown'}}),
         ('index.html', 'M',
          {'added': 3,
           'deleted': 2,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('js/global.js', 'D',
          {'added': 0,
           'deleted': 75,
           'binary': False,
           'ops': {DEL_FILENODE: 'deleted file'}}),
         ('js/jquery/hashgrid.js', 'A',
          {'added': 340,
           'deleted': 0,
           'binary': False,
           'ops': {NEW_FILENODE: 'new file 100755'}}),
         ('less/docs.less', 'M',
          {'added': 34,
           'deleted': 0,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('less/scaffolding.less', 'M',
          {'added': 1,
           'deleted': 3,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('readme.markdown', 'M',
          {'added': 1,
           'deleted': 10,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
     ],
     'git_diff_chmod.diff': [
         ('work-horus.xls', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {CHMOD_FILENODE: 'modified file chmod 100644 => 100755'}})
     ],
     'git_diff_rename_file.diff': [
         ('file.xls', 'R',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {RENAMED_FILENODE: 'file renamed from work-horus.xls to file.xls'}}),
         ('files/var/www/favicon.ico/DEFAULT',
          'R',
          {'added': 0,
           'binary': True,
           'deleted': 0,
           'ops': {4: 'file renamed from files/var/www/favicon.ico to files/var/www/favicon.ico/DEFAULT',
                   6: 'modified file chmod 100644 => 100755'}})
     ],
     'git_diff_mod_single_binary_file.diff': [
         ('US Warszawa.jpg', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {MOD_FILENODE: 'modified file',
                   BIN_FILENODE: 'binary diff not shown'}})
     ],
     'git_diff_binary_and_normal.diff': [
         ('img/baseline-10px.png', 'A',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {NEW_FILENODE: 'new file 100644',
                   BIN_FILENODE: 'binary diff not shown'}}),
         ('img/baseline-20px.png', 'D',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {DEL_FILENODE: 'deleted file',
                   BIN_FILENODE: 'binary diff not shown'}}),
         ('index.html', 'M',
          {'added': 3,
           'deleted': 2,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('js/global.js', 'D',
          {'added': 0,
           'deleted': 75,
           'binary': False,
           'ops': {DEL_FILENODE: 'deleted file'}}),
         ('js/jquery/hashgrid.js', 'A',
          {'added': 340,
           'deleted': 0,
           'binary': False,
           'ops': {NEW_FILENODE: 'new file 100755'}}),
         ('less/docs.less', 'M',
          {'added': 34,
           'deleted': 0,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('less/scaffolding.less', 'M',
          {'added': 1,
           'deleted': 3,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('readme.markdown', 'M',
          {'added': 1,
           'deleted': 10,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
     ],
     'diff_with_diff_data.diff': [
         ('vcs/backends/base.py', 'M',
          {'added': 18,
           'deleted': 2,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('vcs/backends/git/repository.py', 'M',
          {'added': 46,
           'deleted': 15,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('vcs/backends/hg.py', 'M',
          {'added': 22,
           'deleted': 3,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('vcs/tests/test_git.py', 'M',
          {'added': 5,
           'deleted': 5,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
         ('vcs/tests/test_repository.py', 'M',
          {'added': 174,
           'deleted': 2,
           'binary': False,
           'ops': {MOD_FILENODE: 'modified file'}}),
     ],
     'git_diff_modify_binary_file.diff': [
         ('file.name', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {MOD_FILENODE: 'modified file',
                   BIN_FILENODE: 'binary diff not shown'}})
     ],
     'hg_diff_copy_file.diff': [
         ('file2', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {COPIED_FILENODE: 'file copied from file1 to file2'}}),
     ],
     'hg_diff_copy_and_modify_file.diff': [
         ('file3', 'M',
          {'added': 1,
           'deleted': 0,
           'binary': False,
           'ops': {COPIED_FILENODE: 'file copied from file2 to file3',
                   MOD_FILENODE: 'modified file'}}),
     ],
     'hg_diff_copy_and_chmod_file.diff': [
         ('file4', 'M',
          {'added': 0,
           'deleted': 0,
           'binary': True,
           'ops': {COPIED_FILENODE: 'file copied from file3 to file4',
                   CHMOD_FILENODE: 'modified file chmod 100644 => 100755'}}),
     ],
     'hg_diff_copy_chmod_and_edit_file.diff': [
         ('file5', 'M',
          {'added': 2,
           'deleted': 1,
           'binary': False,
           'ops': {COPIED_FILENODE: 'file copied from file4 to file5',
                   CHMOD_FILENODE: 'modified file chmod 100755 => 100644',
                   MOD_FILENODE: 'modified file'}}),
     ],
     'hg_diff_rename_space_cr.diff': [
         ('oh yes', 'R',
          {'added': 3,
           'deleted': 2,
           'binary': False,
           'ops': {RENAMED_FILENODE: 'file renamed from oh no to oh yes'}}),
     ],
 }
 
 
 class TestDiffLib(TestController):
 
     @parametrize('diff_fixture', DIFF_FIXTURES)
     def test_diff(self, diff_fixture):
-        diff = fixture.load_resource(diff_fixture, strip=False)
+        raw_diff = fixture.load_resource(diff_fixture, strip=False)
         vcs = 'hg'
         if diff_fixture.startswith('git_'):
             vcs = 'git'
-        diff_proc = DiffProcessor(diff, vcs=vcs)
-        diff_proc_d = diff_proc.prepare()
-        data = [(x['filename'], x['operation'], x['stats']) for x in diff_proc_d]
+        diff_processor = DiffProcessor(raw_diff, vcs=vcs)
+        data = [(x['filename'], x['operation'], x['stats']) for x in diff_processor.parsed]
         expected_data = DIFF_FIXTURES[diff_fixture]
         assert expected_data == data
 
     def test_diff_markup(self):
-        diff = fixture.load_resource('markuptest.diff', strip=False)
-        diff_proc = DiffProcessor(diff)
-        diff_proc_d = diff_proc.prepare()
-        chunks = diff_proc_d[0]['chunks']
+        raw_diff = fixture.load_resource('markuptest.diff', strip=False)
+        diff_processor = DiffProcessor(raw_diff)
+        chunks = diff_processor.parsed[0]['chunks']
         assert not chunks[0]
         #from pprint import pprint; pprint(chunks[1])
         l = ['\n']
         for d in chunks[1]:
             l.append('%(action)-7s %(new_lineno)3s %(old_lineno)3s %(line)r\n' % d)
         s = ''.join(l)
         print s
         assert s == r'''
 context ... ... u'@@ -51,6 +51,13 @@\n'
 unmod    51  51 u'<u>\t</u>begin();\n'
 unmod    52  52 u'<u>\t</u>\n'
 add      53     u'<u>\t</u>int foo;<u class="cr"></u>\n'
 add      54     u'<u>\t</u>int bar; <u class="cr"></u>\n'
 add      55     u'<u>\t</u>int baz;<u>\t</u><u class="cr"></u>\n'
 add      56     u'<u>\t</u>int space; <i></i>'
 add      57     u'<u>\t</u>int tab;<u>\t</u>\n'
 add      58     u'<u>\t</u>\n'
 unmod    59  53 u' <i></i>'
 del          54 u'<u>\t</u>#define MAX_STEPS (48)\n'
 add      60     u'<u>\t</u><u class="cr"></u>\n'
 add      61     u'<u>\t</u>#define MAX_STEPS (64)<u class="cr"></u>\n'
 unmod    62  55 u'\n'
 del          56 u'<u>\t</u>#define MIN_STEPS (<del>48</del>)\n'
 add      63     u'<u>\t</u>#define MIN_STEPS (<ins>42</ins>)\n'
 '''