diffs: drop the noop as_raw method - just use the raw diff directly and with proper variable naming

kallithea/controllers/changeset.py

@@ -1,461 +1,460 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)
 
-            _diff = c.db_repo_scm_instance.get_diff(cs1, cs2,
+            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
-            diff_processor = diffs.DiffProcessor(_diff,
-                                                 vcs=c.db_repo_scm_instance.alias,
-                                                 diff_limit=diff_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):
                     c.limited_diff = True
                 for f in _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
-                diff = diff_processor.as_raw()
-                file_diff_data.append(('', None, None, None, diff, None))
+                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 diff
+            return raw_diff
         elif method == 'patch':
             response.content_type = 'text/plain'
-            c.diff = safe_unicode(diff)
+            c.diff = safe_unicode(raw_diff)
             return render('changeset/patch_changeset.html')
         elif method == 'raw':
             response.content_type = 'text/plain'
-            return diff
+            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/feed.py

@@ -1,171 +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 __changes(self, cs):
+    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))
-        diff_processor = DiffProcessor(cs.diff(),
+        raw_diff = cs.diff()
+        diff_processor = DiffProcessor(raw_diff,
                                        diff_limit=diff_limit)
         _parsed = diff_processor.prepare(inline_diff=False)
         limited_diff = False
         if isinstance(_parsed, LimitedDiffContainer):
             limited_diff = True
 
         for st in _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...')]
-        return diff_processor, changes
 
-    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)
-        diff_processor, changes = self.__changes(cs)
         # 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(diff_processor.as_raw())
+            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/files.py

@@ -1,785 +1,782 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.files
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 Files controller for Kallithea
 
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Apr 21, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 
 import os
 import posixpath
 import logging
 import traceback
 import tempfile
 import shutil
 
 from tg import request, response, tmpl_context as c
 from tg.i18n import ugettext as _
 from webob.exc import HTTPFound
 
 from kallithea.config.routing import url
 from kallithea.lib.utils import action_logger
 from kallithea.lib import diffs
 from kallithea.lib import helpers as h
 
 from kallithea.lib.compat import OrderedDict
 from kallithea.lib.utils2 import convert_line_endings, detect_mode, safe_str, \
     str2bool, safe_int
 from kallithea.lib.auth import LoginRequired, HasRepoPermissionLevelDecorator
 from kallithea.lib.base import BaseRepoController, render, jsonify
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import RepositoryError, \
     ChangesetDoesNotExistError, EmptyRepositoryError, \
     ImproperArchiveTypeError, VCSError, NodeAlreadyExistsError, \
     NodeDoesNotExistError, ChangesetError, NodeError
 from kallithea.lib.vcs.nodes import FileNode
 
 from kallithea.model.repo import RepoModel
 from kallithea.model.scm import ScmModel
 from kallithea.model.db import Repository
 
 from kallithea.controllers.changeset import anchor_url, _ignorews_url, \
     _context_url, get_line_ctx, get_ignore_ws
 from webob.exc import HTTPNotFound
 from kallithea.lib.exceptions import NonRelativePathError
 
 
 log = logging.getLogger(__name__)
 
 
 class FilesController(BaseRepoController):
 
     def _before(self, *args, **kwargs):
         super(FilesController, self)._before(*args, **kwargs)
 
     def __get_cs(self, rev, silent_empty=False):
         """
         Safe way to get changeset if error occur it redirects to tip with
         proper message
 
         :param rev: revision to fetch
         :silent_empty: return None if repository is empty
         """
 
         try:
             return c.db_repo_scm_instance.get_changeset(rev)
         except EmptyRepositoryError as e:
             if silent_empty:
                 return None
             url_ = url('files_add_home',
                        repo_name=c.repo_name,
                        revision=0, f_path='', anchor='edit')
             add_new = h.link_to(_('Click here to add new file'), url_, class_="alert-link")
             h.flash(h.literal(_('There are no files yet. %s') % add_new),
                     category='warning')
             raise HTTPNotFound()
         except (ChangesetDoesNotExistError, LookupError):
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         except RepositoryError as e:
             h.flash(safe_str(e), category='error')
             raise HTTPNotFound()
 
     def __get_filenode(self, cs, path):
         """
         Returns file_node or raise HTTP error.
 
         :param cs: given changeset
         :param path: path to lookup
         """
 
         try:
             file_node = cs.get_node(path)
             if file_node.is_dir():
                 raise RepositoryError('given path is a directory')
         except ChangesetDoesNotExistError:
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         except RepositoryError as e:
             h.flash(safe_str(e), category='error')
             raise HTTPNotFound()
 
         return file_node
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def index(self, repo_name, revision, f_path, annotate=False):
         # redirect to given revision from form if given
         post_revision = request.POST.get('at_rev', None)
         if post_revision:
             cs = self.__get_cs(post_revision) # FIXME - unused!
 
         c.revision = revision
         c.changeset = self.__get_cs(revision)
         c.branch = request.GET.get('branch', None)
         c.f_path = f_path
         c.annotate = annotate
         cur_rev = c.changeset.revision
         # used in files_source.html:
         c.cut_off_limit = self.cut_off_limit
         c.fulldiff = request.GET.get('fulldiff')
 
         # prev link
         try:
             prev_rev = c.db_repo_scm_instance.get_changeset(cur_rev).prev(c.branch)
             c.url_prev = url('files_home', repo_name=c.repo_name,
                          revision=prev_rev.raw_id, f_path=f_path)
             if c.branch:
                 c.url_prev += '?branch=%s' % c.branch
         except (ChangesetDoesNotExistError, VCSError):
             c.url_prev = '#'
 
         # next link
         try:
             next_rev = c.db_repo_scm_instance.get_changeset(cur_rev).next(c.branch)
             c.url_next = url('files_home', repo_name=c.repo_name,
                      revision=next_rev.raw_id, f_path=f_path)
             if c.branch:
                 c.url_next += '?branch=%s' % c.branch
         except (ChangesetDoesNotExistError, VCSError):
             c.url_next = '#'
 
         # files or dirs
         try:
             c.file = c.changeset.get_node(f_path)
 
             if c.file.is_file():
                 c.load_full_history = False
                 # determine if we're on branch head
                 _branches = c.db_repo_scm_instance.branches
                 c.on_branch_head = revision in _branches.keys() + _branches.values()
                 _hist = []
                 c.file_history = []
                 if c.load_full_history:
                     c.file_history, _hist = self._get_node_history(c.changeset, f_path)
 
                 c.authors = []
                 for a in set([x.author for x in _hist]):
                     c.authors.append((h.email(a), h.person(a)))
             else:
                 c.authors = c.file_history = []
         except RepositoryError as e:
             h.flash(safe_str(e), category='error')
             raise HTTPNotFound()
 
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             return render('files/files_ypjax.html')
 
         # TODO: tags and bookmarks?
         c.revision_options = [(c.changeset.raw_id,
                               _('%s at %s') % (c.changeset.branch, h.short_id(c.changeset.raw_id)))] + \
             [(n, b) for b, n in c.db_repo_scm_instance.branches.items()]
         if c.db_repo_scm_instance.closed_branches:
             prefix = _('(closed)') + ' '
             c.revision_options += [('-', '-')] + \
                 [(n, prefix + b) for b, n in c.db_repo_scm_instance.closed_branches.items()]
 
         return render('files/files.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def history(self, repo_name, revision, f_path):
         changeset = self.__get_cs(revision)
         _file = changeset.get_node(f_path)
         if _file.is_file():
             file_history, _hist = self._get_node_history(changeset, f_path)
 
             res = []
             for obj in file_history:
                 res.append({
                     'text': obj[1],
                     'children': [{'id': o[0], 'text': o[1]} for o in obj[0]]
                 })
 
             data = {
                 'more': False,
                 'results': res
             }
             return data
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def authors(self, repo_name, revision, f_path):
         changeset = self.__get_cs(revision)
         _file = changeset.get_node(f_path)
         if _file.is_file():
             file_history, _hist = self._get_node_history(changeset, f_path)
             c.authors = []
             for a in set([x.author for x in _hist]):
                 c.authors.append((h.email(a), h.person(a)))
             return render('files/files_history_box.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def rawfile(self, repo_name, revision, f_path):
         cs = self.__get_cs(revision)
         file_node = self.__get_filenode(cs, f_path)
 
         response.content_disposition = 'attachment; filename=%s' % \
             safe_str(f_path.split(Repository.url_sep())[-1])
 
         response.content_type = file_node.mimetype
         return file_node.content
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def raw(self, repo_name, revision, f_path):
         cs = self.__get_cs(revision)
         file_node = self.__get_filenode(cs, f_path)
 
         raw_mimetype_mapping = {
             # map original mimetype to a mimetype used for "show as raw"
             # you can also provide a content-disposition to override the
             # default "attachment" disposition.
             # orig_type: (new_type, new_dispo)
 
             # show images inline:
             'image/x-icon': ('image/x-icon', 'inline'),
             'image/png': ('image/png', 'inline'),
             'image/gif': ('image/gif', 'inline'),
             'image/jpeg': ('image/jpeg', 'inline'),
             'image/svg+xml': ('image/svg+xml', 'inline'),
         }
 
         mimetype = file_node.mimetype
         try:
             mimetype, dispo = raw_mimetype_mapping[mimetype]
         except KeyError:
             # we don't know anything special about this, handle it safely
             if file_node.is_binary:
                 # do same as download raw for binary files
                 mimetype, dispo = 'application/octet-stream', 'attachment'
             else:
                 # do not just use the original mimetype, but force text/plain,
                 # otherwise it would serve text/html and that might be unsafe.
                 # Note: underlying vcs library fakes text/plain mimetype if the
                 # mimetype can not be determined and it thinks it is not
                 # binary.This might lead to erroneous text display in some
                 # cases, but helps in other cases, like with text files
                 # without extension.
                 mimetype, dispo = 'text/plain', 'inline'
 
         if dispo == 'attachment':
             dispo = 'attachment; filename=%s' % \
                         safe_str(f_path.split(os.sep)[-1])
 
         response.content_disposition = dispo
         response.content_type = mimetype
         return file_node.content
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def delete(self, repo_name, revision, f_path):
         repo = c.db_repo
         if repo.enable_locking and repo.locked[0]:
             h.flash(_('This repository has been locked by %s on %s')
                 % (h.person_by_id(repo.locked[0]),
                    h.fmt_date(h.time_to_datetime(repo.locked[1]))),
                 'warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip'))
 
         # check if revision is a branch identifier- basically we cannot
         # create multiple heads via file editing
         _branches = repo.scm_instance.branches
         # check if revision is a branch name or branch hash
         if revision not in _branches.keys() + _branches.values():
             h.flash(_('You can only delete files with revision '
                       'being a valid branch'), category='warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip',
                                   f_path=f_path))
 
         r_post = request.POST
 
         c.cs = self.__get_cs(revision)
         c.file = self.__get_filenode(c.cs, f_path)
 
         c.default_message = _('Deleted file %s via Kallithea') % (f_path)
         c.f_path = f_path
         node_path = f_path
         author = request.authuser.full_contact
 
         if r_post:
             message = r_post.get('message') or c.default_message
 
             try:
                 nodes = {
                     node_path: {
                         'content': ''
                     }
                 }
                 self.scm_model.delete_nodes(
                     user=request.authuser.user_id, repo=c.db_repo,
                     message=message,
                     nodes=nodes,
                     parent_cs=c.cs,
                     author=author,
                 )
 
                 h.flash(_('Successfully deleted file %s') % f_path,
                         category='success')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
 
         return render('files/files_delete.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def edit(self, repo_name, revision, f_path):
         repo = c.db_repo
         if repo.enable_locking and repo.locked[0]:
             h.flash(_('This repository has been locked by %s on %s')
                 % (h.person_by_id(repo.locked[0]),
                    h.fmt_date(h.time_to_datetime(repo.locked[1]))),
                 'warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip'))
 
         # check if revision is a branch identifier- basically we cannot
         # create multiple heads via file editing
         _branches = repo.scm_instance.branches
         # check if revision is a branch name or branch hash
         if revision not in _branches.keys() + _branches.values():
             h.flash(_('You can only edit files with revision '
                       'being a valid branch'), category='warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip',
                                   f_path=f_path))
 
         r_post = request.POST
 
         c.cs = self.__get_cs(revision)
         c.file = self.__get_filenode(c.cs, f_path)
 
         if c.file.is_binary:
             raise HTTPFound(location=url('files_home', repo_name=c.repo_name,
                             revision=c.cs.raw_id, f_path=f_path))
         c.default_message = _('Edited file %s via Kallithea') % (f_path)
         c.f_path = f_path
 
         if r_post:
 
             old_content = c.file.content
             sl = old_content.splitlines(1)
             first_line = sl[0] if sl else ''
             # modes:  0 - Unix, 1 - Mac, 2 - DOS
             mode = detect_mode(first_line, 0)
             content = convert_line_endings(r_post.get('content', ''), mode)
 
             message = r_post.get('message') or c.default_message
             author = request.authuser.full_contact
 
             if content == old_content:
                 h.flash(_('No changes'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             try:
                 self.scm_model.commit_change(repo=c.db_repo_scm_instance,
                                              repo_name=repo_name, cs=c.cs,
                                              user=request.authuser.user_id,
                                              author=author, message=message,
                                              content=content, f_path=f_path)
                 h.flash(_('Successfully committed to %s') % f_path,
                         category='success')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
 
         return render('files/files_edit.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def add(self, repo_name, revision, f_path):
 
         repo = c.db_repo
         if repo.enable_locking and repo.locked[0]:
             h.flash(_('This repository has been locked by %s on %s')
                 % (h.person_by_id(repo.locked[0]),
                    h.fmt_date(h.time_to_datetime(repo.locked[1]))),
                   'warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip'))
 
         r_post = request.POST
         c.cs = self.__get_cs(revision, silent_empty=True)
         if c.cs is None:
             c.cs = EmptyChangeset(alias=c.db_repo_scm_instance.alias)
         c.default_message = (_('Added file via Kallithea'))
         c.f_path = f_path
 
         if r_post:
             unix_mode = 0
             content = convert_line_endings(r_post.get('content', ''), unix_mode)
 
             message = r_post.get('message') or c.default_message
             filename = r_post.get('filename')
             location = r_post.get('location', '')
             file_obj = r_post.get('upload_file', None)
 
             if file_obj is not None and hasattr(file_obj, 'filename'):
                 filename = file_obj.filename
                 content = file_obj.file
 
                 if hasattr(content, 'file'):
                     # non posix systems store real file under file attr
                     content = content.file
 
             if not content:
                 h.flash(_('No content'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             if not filename:
                 h.flash(_('No filename'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             # strip all crap out of file, just leave the basename
             filename = os.path.basename(filename)
             node_path = posixpath.join(location, filename)
             author = request.authuser.full_contact
 
             try:
                 nodes = {
                     node_path: {
                         'content': content
                     }
                 }
                 self.scm_model.create_nodes(
                     user=request.authuser.user_id, repo=c.db_repo,
                     message=message,
                     nodes=nodes,
                     parent_cs=c.cs,
                     author=author,
                 )
 
                 h.flash(_('Successfully committed to %s') % node_path,
                         category='success')
             except NonRelativePathError as e:
                 h.flash(_('Location must be relative path and must not '
                           'contain .. in path'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             except (NodeError, NodeAlreadyExistsError) as e:
                 h.flash(_(e), category='error')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
 
         return render('files/files_add.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def archivefile(self, repo_name, fname):
         fileformat = None
         revision = None
         ext = None
         subrepos = request.GET.get('subrepos') == 'true'
 
         for a_type, ext_data in settings.ARCHIVE_SPECS.items():
             archive_spec = fname.split(ext_data[1])
             if len(archive_spec) == 2 and archive_spec[1] == '':
                 fileformat = a_type or ext_data[1]
                 revision = archive_spec[0]
                 ext = ext_data[1]
 
         try:
             dbrepo = RepoModel().get_by_repo_name(repo_name)
             if not dbrepo.enable_downloads:
                 return _('Downloads disabled') # TODO: do something else?
 
             if c.db_repo_scm_instance.alias == 'hg':
                 # patch and reset hooks section of UI config to not run any
                 # hooks on fetching archives with subrepos
                 for k, v in c.db_repo_scm_instance._repo.ui.configitems('hooks'):
                     c.db_repo_scm_instance._repo.ui.setconfig('hooks', k, None)
 
             cs = c.db_repo_scm_instance.get_changeset(revision)
             content_type = settings.ARCHIVE_SPECS[fileformat][0]
         except ChangesetDoesNotExistError:
             return _('Unknown revision %s') % revision
         except EmptyRepositoryError:
             return _('Empty repository')
         except (ImproperArchiveTypeError, KeyError):
             return _('Unknown archive type')
 
         from kallithea import CONFIG
         rev_name = cs.raw_id[:12]
         archive_name = '%s-%s%s' % (safe_str(repo_name.replace('/', '_')),
                                     safe_str(rev_name), ext)
 
         archive_path = None
         cached_archive_path = None
         archive_cache_dir = CONFIG.get('archive_cache_dir')
         if archive_cache_dir and not subrepos: # TODO: subrepo caching?
             if not os.path.isdir(archive_cache_dir):
                 os.makedirs(archive_cache_dir)
             cached_archive_path = os.path.join(archive_cache_dir, archive_name)
             if os.path.isfile(cached_archive_path):
                 log.debug('Found cached archive in %s', cached_archive_path)
                 archive_path = cached_archive_path
             else:
                 log.debug('Archive %s is not yet cached', archive_name)
 
         if archive_path is None:
             # generate new archive
             fd, archive_path = tempfile.mkstemp()
             log.debug('Creating new temp archive in %s', archive_path)
             with os.fdopen(fd, 'wb') as stream:
                 cs.fill_archive(stream=stream, kind=fileformat, subrepos=subrepos)
                 # stream (and thus fd) has been closed by cs.fill_archive
             if cached_archive_path is not None:
                 # we generated the archive - move it to cache
                 log.debug('Storing new archive in %s', cached_archive_path)
                 shutil.move(archive_path, cached_archive_path)
                 archive_path = cached_archive_path
 
         def get_chunked_archive(archive_path):
             stream = open(archive_path, 'rb')
             while True:
                 data = stream.read(16 * 1024)
                 if not data:
                     break
                 yield data
             stream.close()
             if archive_path != cached_archive_path:
                 log.debug('Destroying temp archive %s', archive_path)
                 os.remove(archive_path)
 
         action_logger(user=request.authuser,
                       action='user_downloaded_archive:%s' % (archive_name),
                       repo=repo_name, ipaddr=request.ip_addr, commit=True)
 
         response.content_disposition = str('attachment; filename=%s' % (archive_name))
         response.content_type = str(content_type)
         return get_chunked_archive(archive_path)
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def diff(self, repo_name, f_path):
         ignore_whitespace = request.GET.get('ignorews') == '1'
         line_context = safe_int(request.GET.get('context'), 3)
         diff2 = request.GET.get('diff2', '')
         diff1 = request.GET.get('diff1', '') or diff2
         c.action = request.GET.get('diff')
         c.no_changes = diff1 == diff2
         c.f_path = f_path
         c.big_diff = False
         fulldiff = request.GET.get('fulldiff')
         c.anchor_url = anchor_url
         c.ignorews_url = _ignorews_url
         c.context_url = _context_url
         c.changes = OrderedDict()
         c.changes[diff2] = []
 
         # special case if we want a show rev only, it's impl here
         # to reduce JS and callbacks
 
         if request.GET.get('show_rev'):
             if str2bool(request.GET.get('annotate', 'False')):
                 _url = url('files_annotate_home', repo_name=c.repo_name,
                            revision=diff1, f_path=c.f_path)
             else:
                 _url = url('files_home', repo_name=c.repo_name,
                            revision=diff1, f_path=c.f_path)
 
             raise HTTPFound(location=_url)
         try:
             if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_1 = c.db_repo_scm_instance.get_changeset(diff1)
                 try:
                     node1 = c.changeset_1.get_node(f_path)
                     if node1.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node1, type(node1)))
                 except NodeDoesNotExistError:
                     c.changeset_1 = EmptyChangeset(cs=diff1,
                                                    revision=c.changeset_1.revision,
                                                    repo=c.db_repo_scm_instance)
                     node1 = FileNode(f_path, '', changeset=c.changeset_1)
             else:
                 c.changeset_1 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node1 = FileNode(f_path, '', changeset=c.changeset_1)
 
             if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_2 = c.db_repo_scm_instance.get_changeset(diff2)
                 try:
                     node2 = c.changeset_2.get_node(f_path)
                     if node2.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node2, type(node2)))
                 except NodeDoesNotExistError:
                     c.changeset_2 = EmptyChangeset(cs=diff2,
                                                    revision=c.changeset_2.revision,
                                                    repo=c.db_repo_scm_instance)
                     node2 = FileNode(f_path, '', changeset=c.changeset_2)
             else:
                 c.changeset_2 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node2 = FileNode(f_path, '', changeset=c.changeset_2)
         except (RepositoryError, NodeError):
             log.error(traceback.format_exc())
             raise HTTPFound(location=url('files_home', repo_name=c.repo_name,
                                 f_path=f_path))
 
         if c.action == 'download':
-            _diff = diffs.get_gitdiff(node1, node2,
+            raw_diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
-            diff = diffs.DiffProcessor(_diff)
-
             diff_name = '%s_vs_%s.diff' % (diff1, diff2)
             response.content_type = 'text/plain'
             response.content_disposition = (
                 'attachment; filename=%s' % diff_name
             )
-            return diff.as_raw()
+            return raw_diff
 
         elif c.action == 'raw':
-            _diff = diffs.get_gitdiff(node1, node2,
+            raw_diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
-            diff = diffs.DiffProcessor(_diff)
             response.content_type = 'text/plain'
-            return diff.as_raw()
+            return raw_diff
 
         else:
             fid = h.FID(diff2, node2.path)
             line_context_lcl = get_line_ctx(fid, request.GET)
             ign_whitespace_lcl = get_ignore_ws(fid, request.GET)
 
             diff_limit = None if fulldiff else self.cut_off_limit
             c.a_rev, c.cs_rev, a_path, diff, st, op = diffs.wrapped_diff(filenode_old=node1,
                                          filenode_new=node2,
                                          diff_limit=diff_limit,
                                          ignore_whitespace=ign_whitespace_lcl,
                                          line_context=line_context_lcl,
                                          enable_comments=False)
             c.file_diff_data = [(fid, fid, op, a_path, node2.path, diff, st)]
 
             return render('files/file_diff.html')
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     def diff_2way(self, repo_name, f_path):
         diff1 = request.GET.get('diff1', '')
         diff2 = request.GET.get('diff2', '')
         try:
             if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_1 = c.db_repo_scm_instance.get_changeset(diff1)
                 try:
                     node1 = c.changeset_1.get_node(f_path)
                     if node1.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node1, type(node1)))
                 except NodeDoesNotExistError:
                     c.changeset_1 = EmptyChangeset(cs=diff1,
                                                    revision=c.changeset_1.revision,
                                                    repo=c.db_repo_scm_instance)
                     node1 = FileNode(f_path, '', changeset=c.changeset_1)
             else:
                 c.changeset_1 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node1 = FileNode(f_path, '', changeset=c.changeset_1)
 
             if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_2 = c.db_repo_scm_instance.get_changeset(diff2)
                 try:
                     node2 = c.changeset_2.get_node(f_path)
                     if node2.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node2, type(node2)))
                 except NodeDoesNotExistError:
                     c.changeset_2 = EmptyChangeset(cs=diff2,
                                                    revision=c.changeset_2.revision,
                                                    repo=c.db_repo_scm_instance)
                     node2 = FileNode(f_path, '', changeset=c.changeset_2)
             else:
                 c.changeset_2 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node2 = FileNode(f_path, '', changeset=c.changeset_2)
         except ChangesetDoesNotExistError as e:
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         c.node1 = node1
         c.node2 = node2
         c.cs1 = c.changeset_1
         c.cs2 = c.changeset_2
 
         return render('files/diff_2way.html')
 
     def _get_node_history(self, cs, f_path, changesets=None):
         """
         get changesets history for given node
 
         :param cs: changeset to calculate history
         :param f_path: path for node to calculate history for
         :param changesets: if passed don't calculate history and take
             changesets defined in this list
         """
         # calculate history based on tip
         tip_cs = c.db_repo_scm_instance.get_changeset()
         if changesets is None:
             try:
                 changesets = tip_cs.get_file_history(f_path)
             except (NodeDoesNotExistError, ChangesetError):
                 # this node is not present at tip !
                 changesets = cs.get_file_history(f_path)
         hist_l = []
 
         changesets_group = ([], _("Changesets"))
         branches_group = ([], _("Branches"))
         tags_group = ([], _("Tags"))
         for chs in changesets:
             #_branch = '(%s)' % chs.branch if (cs.repository.alias == 'hg') else ''
             _branch = chs.branch
             n_desc = '%s (%s)' % (h.show_id(chs), _branch)
             changesets_group[0].append((chs.raw_id, n_desc,))
         hist_l.append(changesets_group)
 
         for name, chs in c.db_repo_scm_instance.branches.items():
             branches_group[0].append((chs, name),)
         hist_l.append(branches_group)
 
         for name, chs in c.db_repo_scm_instance.tags.items():
             tags_group[0].append((chs, name),)
         hist_l.append(tags_group)
 
         return hist_l, changesets
 
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def nodelist(self, repo_name, revision, f_path):
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             cs = self.__get_cs(revision)
             _d, _f = ScmModel().get_nodes(repo_name, cs.raw_id, f_path,
                                           flat=False)
             return {'nodes': _d + _f}

kallithea/lib/diffs.py

@@ -1,713 +1,707 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 itertools import imap
 
 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,
                                 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]
             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.
     """
     _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"""
         # has already been split on this:
         # ^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"""
         # has already been split on this:
         # ^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):
         """
         :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
 
     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]))
         difflines = imap(self._escaper, re.findall(r'.*\n|.+$', rest)) # don't split on \r as str.splitlines do
         return meta_info, difflines
 
     def _clean_line(self, line, command):
         """Given a diff line, strip the leading character if it is a plus/minus/context line."""
         if command in ['+', '-', ' ']:
             line = line[1:]
         return line
 
     def _parse_gitdiff(self, inline_diff=True):
         """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
 
         # split the diff in chunks of separate --git a/file b/file chunks
         for raw_diff in ('\n' + self._diff).split('\ndiff --git')[1:]:
             head, diff = self._get_header(raw_diff)
 
             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)
                     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):
         """
         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.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.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':         self._clean_line(line, command)
                         })
 
                     line = diff.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':         self._clean_line(line, command)
                         })
                         line = diff.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_raw(self):
-        """
-        Returns raw string diff, exactly as it was passed in the first place.
-        """
-        return self._diff
-
     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
         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