various fixes for git and mercurial with InMemoryCommit backend and non-ascii files

- hg use tolocal for unicode objects
- git use core.quotepath=false for commands

rhodecode/controllers/files.py

@@ -1,486 +1,486 b''
 # -*- coding: utf-8 -*-
 """
     rhodecode.controllers.files
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
     Files controller for RhodeCode
 
     :created_on: Apr 21, 2010
     :author: marcink
     :copyright: (C) 2010-2012 Marcin Kuzminski <[e-mail]>
     :license: GPLv3, see COPYING for more details.
 """
 # 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/>.
 
 import os
 import logging
 import traceback
 
 from pylons import request, response, tmpl_context as c, url
 from pylons.i18n.translation import _
 from pylons.controllers.util import redirect
 from pylons.decorators import jsonify
 
 from rhodecode.lib import diffs
 from rhodecode.lib import helpers as h
 
 from rhodecode.lib.compat import OrderedDict
 from rhodecode.lib.utils2 import convert_line_endings, detect_mode, safe_str
 from rhodecode.lib.auth import LoginRequired, HasRepoPermissionAnyDecorator
 from rhodecode.lib.base import BaseRepoController, render
 from rhodecode.lib.utils import EmptyChangeset
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.exceptions import RepositoryError, \
     ChangesetDoesNotExistError, EmptyRepositoryError, \
     ImproperArchiveTypeError, VCSError, NodeAlreadyExistsError
 from rhodecode.lib.vcs.nodes import FileNode
 
 from rhodecode.model.repo import RepoModel
 from rhodecode.model.scm import ScmModel
 
 from rhodecode.controllers.changeset import anchor_url, _ignorews_url,\
     _context_url, get_line_ctx, get_ignore_ws
 
 
 log = logging.getLogger(__name__)
 
 
 class FilesController(BaseRepoController):
 
     @LoginRequired()
     def __before__(self):
         super(FilesController, self).__before__()
         c.cut_off_limit = self.cut_off_limit
 
     def __get_cs_or_redirect(self, rev, repo_name, redirect_after=True):
         """
         Safe way to get changeset if error occur it redirects to tip with
         proper message
 
         :param rev: revision to fetch
         :param repo_name: repo name to redirect after
         """
 
         try:
             return c.rhodecode_repo.get_changeset(rev)
         except EmptyRepositoryError, e:
             if not redirect_after:
                 return None
             url_ = url('files_add_home',
                        repo_name=c.repo_name,
                        revision=0, f_path='')
             add_new = '<a href="%s">[%s]</a>' % (url_, _('add new'))
             h.flash(h.literal(_('There are no files yet %s' % add_new)),
                     category='warning')
             redirect(h.url('summary_home', repo_name=repo_name))
 
         except RepositoryError, e:
             h.flash(str(e), category='warning')
             redirect(h.url('files_home', repo_name=repo_name, revision='tip'))
 
     def __get_filenode_or_redirect(self, repo_name, cs, path):
         """
         Returns file_node, if error occurs or given path is directory,
         it'll redirect to top level path
 
         :param repo_name: repo_name
         :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 RepositoryError, e:
             h.flash(str(e), category='warning')
             redirect(h.url('files_home', repo_name=repo_name,
                            revision=cs.raw_id))
 
         return file_node
 
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     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_or_redirect(post_revision, repo_name)
             redirect(url('files_home', repo_name=c.repo_name,
                          revision=cs.raw_id, f_path=f_path))
 
         c.changeset = self.__get_cs_or_redirect(revision, repo_name)
         c.branch = request.GET.get('branch', None)
         c.f_path = f_path
         c.annotate = annotate
         cur_rev = c.changeset.revision
 
         # prev link
         try:
             prev_rev = c.rhodecode_repo.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.rhodecode_repo.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.file_history = self._get_node_history(c.changeset, f_path)
             else:
                 c.file_history = []
         except RepositoryError, e:
             h.flash(str(e), category='warning')
             redirect(h.url('files_home', repo_name=repo_name,
                            revision=revision))
 
         return render('files/files.html')
 
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     def rawfile(self, repo_name, revision, f_path):
         cs = self.__get_cs_or_redirect(revision, repo_name)
         file_node = self.__get_filenode_or_redirect(repo_name, cs, f_path)
 
         response.content_disposition = 'attachment; filename=%s' % \
             safe_str(f_path.split(os.sep)[-1])
 
         response.content_type = file_node.mimetype
         return file_node.content
 
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     def raw(self, repo_name, revision, f_path):
         cs = self.__get_cs_or_redirect(revision, repo_name)
         file_node = self.__get_filenode_or_redirect(repo_name, 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
 
     @HasRepoPermissionAnyDecorator('repository.write', 'repository.admin')
     def edit(self, repo_name, revision, f_path):
         r_post = request.POST
 
         c.cs = self.__get_cs_or_redirect(revision, repo_name)
         c.file = self.__get_filenode_or_redirect(repo_name, c.cs, f_path)
 
         if c.file.is_binary:
             return redirect(url('files_home', repo_name=c.repo_name,
                          revision=c.cs.raw_id, f_path=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 (_('Edited %s via RhodeCode')
                                                 % (f_path))
             author = self.rhodecode_user.full_contact
 
             if content == old_content:
                 h.flash(_('No changes'),
                     category='warning')
                 return redirect(url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
 
             try:
                 self.scm_model.commit_change(repo=c.rhodecode_repo,
                                              repo_name=repo_name, cs=c.cs,
                                              user=self.rhodecode_user,
                                              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')
             return redirect(url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
 
         return render('files/files_edit.html')
 
     @HasRepoPermissionAnyDecorator('repository.write', 'repository.admin')
     def add(self, repo_name, revision, f_path):
         r_post = request.POST
         c.cs = self.__get_cs_or_redirect(revision, repo_name,
                                          redirect_after=False)
         if c.cs is None:
             c.cs = EmptyChangeset(alias=c.rhodecode_repo.alias)
 
         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 (_('Added %s via RhodeCode')
                                                 % (f_path))
             location = r_post.get('location')
             filename = r_post.get('filename')
             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
 
             node_path = os.path.join(location, filename)
             author = self.rhodecode_user.full_contact
 
             if not content:
                 h.flash(_('No content'), category='warning')
                 return redirect(url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             if not filename:
                 h.flash(_('No filename'), category='warning')
                 return redirect(url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
 
             try:
                 self.scm_model.create_node(repo=c.rhodecode_repo,
-                                             repo_name=repo_name, cs=c.cs,
-                                             user=self.rhodecode_user,
-                                             author=author, message=message,
-                                             content=content, f_path=node_path)
+                                           repo_name=repo_name, cs=c.cs,
+                                           user=self.rhodecode_user,
+                                           author=author, message=message,
+                                           content=content, f_path=node_path)
                 h.flash(_('Successfully committed to %s' % node_path),
                         category='success')
             except NodeAlreadyExistsError, e:
                 h.flash(_(e), category='error')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             return redirect(url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
 
         return render('files/files_add.html')
 
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     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 dbrepo.enable_downloads is False:
                 return _('downloads disabled')
 
             if c.rhodecode_repo.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.rhodecode_repo._repo.ui.configitems('hooks'):
                     c.rhodecode_repo._repo.ui.setconfig('hooks', k, None)
 
             cs = c.rhodecode_repo.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')
 
         response.content_type = content_type
         response.content_disposition = 'attachment; filename=%s-%s%s' \
             % (repo_name, revision, ext)
 
         import tempfile
         archive = tempfile.mkstemp()[1]
         t = open(archive, 'wb')
         cs.fill_archive(stream=t, kind=fileformat, subrepos=subrepos)
 
         def get_chunked_archive(archive):
             stream = open(archive, 'rb')
             while True:
                 data = stream.read(4096)
                 if not data:
                     os.remove(archive)
                     break
                 yield data
 
         return get_chunked_archive(archive)
 
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     def diff(self, repo_name, f_path):
         ignore_whitespace = request.GET.get('ignorews') == '1'
         line_context = request.GET.get('context', 3)
         diff1 = request.GET.get('diff1', '')
         diff2 = request.GET.get('diff2', '')
         c.action = request.GET.get('diff')
         c.no_changes = diff1 == diff2
         c.f_path = f_path
         c.big_diff = False
         c.anchor_url = anchor_url
         c.ignorews_url = _ignorews_url
         c.context_url = _context_url
         c.changes = OrderedDict()
         c.changes[diff2] = []
         try:
             if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_1 = c.rhodecode_repo.get_changeset(diff1)
                 node1 = c.changeset_1.get_node(f_path)
             else:
                 c.changeset_1 = EmptyChangeset(repo=c.rhodecode_repo)
                 node1 = FileNode('.', '', changeset=c.changeset_1)
 
             if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_2 = c.rhodecode_repo.get_changeset(diff2)
                 node2 = c.changeset_2.get_node(f_path)
             else:
                 c.changeset_2 = EmptyChangeset(repo=c.rhodecode_repo)
                 node2 = FileNode('.', '', changeset=c.changeset_2)
         except RepositoryError:
             return redirect(url('files_home', repo_name=c.repo_name,
                                 f_path=f_path))
 
         if c.action == 'download':
             _diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
             diff = diffs.DiffProcessor(_diff, format='gitdiff')
 
             diff_name = '%s_vs_%s.diff' % (diff1, diff2)
             response.content_type = 'text/plain'
             response.content_disposition = (
                 'attachment; filename=%s' % diff_name
             )
             return diff.raw_diff()
 
         elif c.action == 'raw':
             _diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
             diff = diffs.DiffProcessor(_diff, format='gitdiff')
             response.content_type = 'text/plain'
             return diff.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)
 
             lim = request.GET.get('fulldiff') or self.cut_off_limit
             _, cs1, cs2, diff, st = diffs.wrapped_diff(filenode_old=node1,
                                          filenode_new=node2,
                                          cut_off_limit=lim,
                                          ignore_whitespace=ign_whitespace_lcl,
                                          line_context=line_context_lcl,
                                          enable_comments=False)
 
             c.changes = [('', node2, diff, cs1, cs2, st,)]
 
         return render('files/file_diff.html')
 
     def _get_node_history(self, cs, f_path):
         changesets = cs.get_file_history(f_path)
         hist_l = []
 
         changesets_group = ([], _("Changesets"))
         branches_group = ([], _("Branches"))
         tags_group = ([], _("Tags"))
         _hg = cs.repository.alias == 'hg'
         for chs in changesets:
             _branch = '(%s)' % chs.branch if _hg else ''
             n_desc = 'r%s:%s %s' % (chs.revision, chs.short_id, _branch)
             changesets_group[0].append((chs.raw_id, n_desc,))
 
         hist_l.append(changesets_group)
 
         for name, chs in c.rhodecode_repo.branches.items():
             branches_group[0].append((chs, name),)
         hist_l.append(branches_group)
 
         for name, chs in c.rhodecode_repo.tags.items():
             tags_group[0].append((chs, name),)
         hist_l.append(tags_group)
 
         return hist_l
 
     @jsonify
     @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',
                                    'repository.admin')
     def nodelist(self, repo_name, revision, f_path):
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             cs = self.__get_cs_or_redirect(revision, repo_name)
             _d, _f = ScmModel().get_nodes(repo_name, cs.raw_id, f_path,
                                           flat=False)
             return _d + _f

rhodecode/lib/vcs/backends/git/changeset.py

@@ -1,446 +1,447 b''
 import re
 from itertools import chain
 from dulwich import objects
 from subprocess import Popen, PIPE
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.exceptions import ChangesetError
 from rhodecode.lib.vcs.exceptions import NodeDoesNotExistError
 from rhodecode.lib.vcs.exceptions import VCSError
 from rhodecode.lib.vcs.exceptions import ChangesetDoesNotExistError
 from rhodecode.lib.vcs.exceptions import ImproperArchiveTypeError
 from rhodecode.lib.vcs.backends.base import BaseChangeset
 from rhodecode.lib.vcs.nodes import FileNode, DirNode, NodeKind, RootNode, RemovedFileNode
 from rhodecode.lib.vcs.utils import safe_unicode
 from rhodecode.lib.vcs.utils import date_fromtimestamp
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 
 
 class GitChangeset(BaseChangeset):
     """
     Represents state of the repository at single revision.
     """
 
     def __init__(self, repository, revision):
         self._stat_modes = {}
         self.repository = repository
         self.raw_id = revision
         self.revision = repository.revisions.index(revision)
 
         self.short_id = self.raw_id[:12]
         self.id = self.raw_id
         try:
             commit = self.repository._repo.get_object(self.raw_id)
         except KeyError:
             raise RepositoryError("Cannot get object with id %s" % self.raw_id)
         self._commit = commit
         self._tree_id = commit.tree
 
         try:
             self.message = safe_unicode(commit.message[:-1])
             # Always strip last eol
         except UnicodeDecodeError:
             self.message = commit.message[:-1].decode(commit.encoding
                 or 'utf-8')
         #self.branch = None
         self.tags = []
         #tree = self.repository.get_object(self._tree_id)
         self.nodes = {}
         self._paths = {}
 
     @LazyProperty
     def author(self):
         return safe_unicode(self._commit.committer)
 
     @LazyProperty
     def date(self):
         return date_fromtimestamp(self._commit.commit_time,
                                   self._commit.commit_timezone)
 
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.changed, self.added, self.removed
 
     @LazyProperty
     def branch(self):
 
         heads = self.repository._heads(reverse=False)
 
         ref = heads.get(self.raw_id)
         if ref:
             return safe_unicode(ref)
 
-
     def _fix_path(self, path):
         """
         Paths are stored without trailing slash so we need to get rid off it if
         needed.
         """
         if path.endswith('/'):
             path = path.rstrip('/')
         return path
 
     def _get_id_for_path(self, path):
 
         # FIXME: Please, spare a couple of minutes and make those codes cleaner;
         if not path in self._paths:
             path = path.strip('/')
             # set root tree
             tree = self.repository._repo[self._commit.tree]
             if path == '':
                 self._paths[''] = tree.id
                 return tree.id
             splitted = path.split('/')
             dirs, name = splitted[:-1], splitted[-1]
             curdir = ''
 
             # initially extract things from root dir
             for item, stat, id in tree.iteritems():
                 if curdir:
                     name = '/'.join((curdir, item))
                 else:
                     name = item
                 self._paths[name] = id
                 self._stat_modes[name] = stat
 
             for dir in dirs:
                 if curdir:
                     curdir = '/'.join((curdir, dir))
                 else:
                     curdir = dir
                 dir_id = None
                 for item, stat, id in tree.iteritems():
                     if dir == item:
                         dir_id = id
                 if dir_id:
                     # Update tree
                     tree = self.repository._repo[dir_id]
                     if not isinstance(tree, objects.Tree):
                         raise ChangesetError('%s is not a directory' % curdir)
                 else:
                     raise ChangesetError('%s have not been found' % curdir)
 
                 # cache all items from the given traversed tree
                 for item, stat, id in tree.iteritems():
                     if curdir:
                         name = '/'.join((curdir, item))
                     else:
                         name = item
                     self._paths[name] = id
                     self._stat_modes[name] = stat
-
             if not path in self._paths:
                 raise NodeDoesNotExistError("There is no file nor directory "
                     "at the given path %r at revision %r"
                     % (path, self.short_id))
         return self._paths[path]
 
     def _get_kind(self, path):
         id = self._get_id_for_path(path)
         obj = self.repository._repo[id]
         if isinstance(obj, objects.Blob):
             return NodeKind.FILE
         elif isinstance(obj, objects.Tree):
             return NodeKind.DIR
 
     def _get_file_nodes(self):
         return chain(*(t[2] for t in self.walk()))
 
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         return [self.repository.get_changeset(parent)
             for parent in self._commit.parents]
 
     def next(self, branch=None):
 
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
 
         def _next(changeset, branch):
             try:
                 next_ = changeset.revision + 1
                 next_rev = changeset.repository.revisions[next_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = changeset.repository.get_changeset(next_rev)
 
             if branch and branch != cs.branch:
                 return _next(cs, branch)
 
             return cs
 
         return _next(self, branch)
 
     def prev(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
 
         def _prev(changeset, branch):
             try:
                 prev_ = changeset.revision - 1
                 if prev_ < 0:
                     raise IndexError
                 prev_rev = changeset.repository.revisions[prev_]
             except IndexError:
                 raise ChangesetDoesNotExistError
 
             cs = changeset.repository.get_changeset(prev_rev)
 
             if branch and branch != cs.branch:
                 return _prev(cs, branch)
 
             return cs
 
         return _prev(self, branch)
 
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         # ensure path is traversed
         self._get_id_for_path(path)
         return self._stat_modes[path]
 
     def get_file_content(self, path):
         """
         Returns content of the file at given ``path``.
         """
         id = self._get_id_for_path(path)
         blob = self.repository._repo[id]
         return blob.as_pretty_string()
 
     def get_file_size(self, path):
         """
         Returns size of the file at given ``path``.
         """
         id = self._get_id_for_path(path)
         blob = self.repository._repo[id]
         return blob.raw_length()
 
     def get_file_changeset(self, path):
         """
         Returns last commit of the file at the given ``path``.
         """
         node = self.get_node(path)
         return node.history[0]
 
     def get_file_history(self, path):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
 
         TODO: This function now uses os underlying 'git' and 'grep' commands
         which is generally not good. Should be replaced with algorithm
         iterating commits.
         """
         cmd = 'log --pretty="format: %%H" --name-status -p %s -- "%s"' % (
                   self.id, path
                )
         so, se = self.repository.run_git_command(cmd)
         ids = re.findall(r'\w{40}', so)
         return [self.repository.get_changeset(id) for id in ids]
 
     def get_file_annotate(self, path):
         """
         Returns a list of three element tuples with lineno,changeset and line
 
         TODO: This function now uses os underlying 'git' command which is
         generally not good. Should be replaced with algorithm iterating
         commits.
         """
         cmd = 'blame -l --root -r %s -- "%s"' % (self.id, path)
         # -l     ==> outputs long shas (and we need all 40 characters)
         # --root ==> doesn't put '^' character for bounderies
         # -r sha ==> blames for the given revision
         so, se = self.repository.run_git_command(cmd)
         annotate = []
         for i, blame_line in enumerate(so.split('\n')[:-1]):
             ln_no = i + 1
             id, line = re.split(r' \(.+?\) ', blame_line, 1)
             annotate.append((ln_no, self.repository.get_changeset(id), line))
         return annotate
 
     def fill_archive(self, stream=None, kind='tgz', prefix=None,
                      subrepos=False):
         """
         Fills up given stream.
 
         :param stream: file like object.
         :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``.
             Default: ``tgz``.
         :param prefix: name of root directory in archive.
             Default is repository name and changeset's raw_id joined with dash
             (``repo-tip.<KIND>``).
         :param subrepos: include subrepos in this archive.
 
         :raise ImproperArchiveTypeError: If given kind is wrong.
         :raise VcsError: If given stream is None
 
         """
         allowed_kinds = settings.ARCHIVE_SPECS.keys()
         if kind not in allowed_kinds:
             raise ImproperArchiveTypeError('Archive kind not supported use one'
                 'of %s', allowed_kinds)
 
         if prefix is None:
             prefix = '%s-%s' % (self.repository.name, self.short_id)
         elif prefix.startswith('/'):
             raise VCSError("Prefix cannot start with leading slash")
         elif prefix.strip() == '':
             raise VCSError("Prefix cannot be empty")
 
         if kind == 'zip':
             frmt = 'zip'
         else:
             frmt = 'tar'
         cmd = 'git archive --format=%s --prefix=%s/ %s' % (frmt, prefix,
             self.raw_id)
         if kind == 'tgz':
             cmd += ' | gzip -9'
         elif kind == 'tbz2':
             cmd += ' | bzip2 -9'
 
         if stream is None:
             raise VCSError('You need to pass in a valid stream for filling'
                            ' with archival data')
         popen = Popen(cmd, stdout=PIPE, stderr=PIPE, shell=True,
             cwd=self.repository.path)
 
         buffer_size = 1024 * 8
         chunk = popen.stdout.read(buffer_size)
         while chunk:
             stream.write(chunk)
             chunk = popen.stdout.read(buffer_size)
         # Make sure all descriptors would be read
         popen.communicate()
 
     def get_nodes(self, path):
         if self._get_kind(path) != NodeKind.DIR:
             raise ChangesetError("Directory does not exist for revision %r at "
                 " %r" % (self.revision, path))
         path = self._fix_path(path)
         id = self._get_id_for_path(path)
         tree = self.repository._repo[id]
         dirnodes = []
         filenodes = []
         for name, stat, id in tree.iteritems():
             obj = self.repository._repo.get_object(id)
             if path != '':
                 obj_path = '/'.join((path, name))
             else:
                 obj_path = name
             if obj_path not in self._stat_modes:
                 self._stat_modes[obj_path] = stat
             if isinstance(obj, objects.Tree):
                 dirnodes.append(DirNode(obj_path, changeset=self))
             elif isinstance(obj, objects.Blob):
                 filenodes.append(FileNode(obj_path, changeset=self, mode=stat))
             else:
                 raise ChangesetError("Requested object should be Tree "
                                      "or Blob, is %r" % type(obj))
         nodes = dirnodes + filenodes
         for node in nodes:
             if not node.path in self.nodes:
                 self.nodes[node.path] = node
         nodes.sort()
         return nodes
 
     def get_node(self, path):
         if isinstance(path, unicode):
             path = path.encode('utf-8')
         path = self._fix_path(path)
         if not path in self.nodes:
             try:
                 id = self._get_id_for_path(path)
             except ChangesetError:
                 raise NodeDoesNotExistError("Cannot find one of parents' "
                     "directories for a given path: %s" % path)
             obj = self.repository._repo.get_object(id)
             if isinstance(obj, objects.Tree):
                 if path == '':
                     node = RootNode(changeset=self)
                 else:
                     node = DirNode(path, changeset=self)
                 node._tree = obj
             elif isinstance(obj, objects.Blob):
                 node = FileNode(path, changeset=self)
                 node._blob = obj
             else:
                 raise NodeDoesNotExistError("There is no file nor directory "
                     "at the given path %r at revision %r"
                     % (path, self.short_id))
             # cache node
             self.nodes[path] = node
         return self.nodes[path]
 
     @LazyProperty
     def affected_files(self):
         """
         Get's a fast accessible file changes for given changeset
         """
 
         return self.added + self.changed
 
     @LazyProperty
     def _diff_name_status(self):
         output = []
         for parent in self.parents:
-            cmd = 'diff --name-status %s %s' % (parent.raw_id, self.raw_id)
+            cmd = 'diff --name-status %s %s --encoding=utf8' % (parent.raw_id, self.raw_id)
             so, se = self.repository.run_git_command(cmd)
             output.append(so.strip())
         return '\n'.join(output)
 
     def _get_paths_for_status(self, status):
         """
         Returns sorted list of paths for given ``status``.
 
         :param status: one of: *added*, *modified* or *deleted*
         """
         paths = set()
         char = status[0].upper()
         for line in self._diff_name_status.splitlines():
             if not line:
                 continue
+
             if line.startswith(char):
-                splitted = line.split(char,1)
+                splitted = line.split(char, 1)
                 if not len(splitted) == 2:
                     raise VCSError("Couldn't parse diff result:\n%s\n\n and "
                         "particularly that line: %s" % (self._diff_name_status,
                         line))
-                paths.add(splitted[1].strip())
+                _path = splitted[1].strip()
+                paths.add(_path)
+
         return sorted(paths)
 
     @LazyProperty
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         if not self.parents:
             return list(self._get_file_nodes())
         return [self.get_node(path) for path in self._get_paths_for_status('added')]
 
     @LazyProperty
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return [self.get_node(path) for path in self._get_paths_for_status('modified')]
 
     @LazyProperty
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return [RemovedFileNode(path) for path in self._get_paths_for_status('deleted')]

rhodecode/lib/vcs/backends/git/inmemory.py

@@ -1,192 +1,193 b''
 import time
 import datetime
 import posixpath
 from dulwich import objects
 from dulwich.repo import Repo
 from rhodecode.lib.vcs.backends.base import BaseInMemoryChangeset
 from rhodecode.lib.vcs.exceptions import RepositoryError
+from rhodecode.lib.vcs.utils import safe_str
 
 
 class GitInMemoryChangeset(BaseInMemoryChangeset):
 
     def commit(self, message, author, parents=None, branch=None, date=None,
-            **kwargs):
+               **kwargs):
         """
         Performs in-memory commit (doesn't check workdir in any way) and
         returns newly created ``Changeset``. Updates repository's
         ``revisions``.
 
         :param message: message of the commit
         :param author: full username, i.e. "Joe Doe <[e-mail]>"
         :param parents: single parent or sequence of parents from which commit
           would be derieved
         :param date: ``datetime.datetime`` instance. Defaults to
           ``datetime.datetime.now()``.
         :param branch: branch name, as string. If none given, default backend's
           branch would be used.
 
         :raises ``CommitError``: if any error occurs while committing
         """
         self.check_integrity(parents)
 
         from .repository import GitRepository
         if branch is None:
             branch = GitRepository.DEFAULT_BRANCH_NAME
 
         repo = self.repository._repo
         object_store = repo.object_store
 
         ENCODING = "UTF-8"
         DIRMOD = 040000
 
         # Create tree and populates it with blobs
         commit_tree = self.parents[0] and repo[self.parents[0]._commit.tree] or\
             objects.Tree()
         for node in self.added + self.changed:
             # Compute subdirs if needed
             dirpath, nodename = posixpath.split(node.path)
             dirnames = dirpath and dirpath.split('/') or []
             parent = commit_tree
             ancestors = [('', parent)]
 
             # Tries to dig for the deepest existing tree
             while dirnames:
                 curdir = dirnames.pop(0)
                 try:
                     dir_id = parent[curdir][1]
                 except KeyError:
                     # put curdir back into dirnames and stops
                     dirnames.insert(0, curdir)
                     break
                 else:
                     # If found, updates parent
                     parent = self.repository._repo[dir_id]
                     ancestors.append((curdir, parent))
             # Now parent is deepest exising tree and we need to create subtrees
             # for dirnames (in reverse order) [this only applies for nodes from added]
             new_trees = []
             blob = objects.Blob.from_string(node.content.encode(ENCODING))
             node_path = node.name.encode(ENCODING)
             if dirnames:
                 # If there are trees which should be created we need to build
                 # them now (in reverse order)
                 reversed_dirnames = list(reversed(dirnames))
                 curtree = objects.Tree()
                 curtree[node_path] = node.mode, blob.id
                 new_trees.append(curtree)
                 for dirname in reversed_dirnames[:-1]:
                     newtree = objects.Tree()
                     #newtree.add(DIRMOD, dirname, curtree.id)
                     newtree[dirname] = DIRMOD, curtree.id
                     new_trees.append(newtree)
                     curtree = newtree
                 parent[reversed_dirnames[-1]] = DIRMOD, curtree.id
             else:
                 parent.add(node.mode, node_path, blob.id)
             new_trees.append(parent)
             # Update ancestors
             for parent, tree, path in reversed([(a[1], b[1], b[0]) for a, b in
                 zip(ancestors, ancestors[1:])]):
                 parent[path] = DIRMOD, tree.id
                 object_store.add_object(tree)
 
             object_store.add_object(blob)
             for tree in new_trees:
                 object_store.add_object(tree)
         for node in self.removed:
             paths = node.path.split('/')
             tree = commit_tree
             trees = [tree]
             # Traverse deep into the forest...
             for path in paths:
                 try:
                     obj = self.repository._repo[tree[path][1]]
                     if isinstance(obj, objects.Tree):
                         trees.append(obj)
                         tree = obj
                 except KeyError:
                     break
             # Cut down the blob and all rotten trees on the way back...
             for path, tree in reversed(zip(paths, trees)):
                 del tree[path]
                 if tree:
                     # This tree still has elements - don't remove it or any
                     # of it's parents
                     break
 
         object_store.add_object(commit_tree)
 
         # Create commit
         commit = objects.Commit()
         commit.tree = commit_tree.id
         commit.parents = [p._commit.id for p in self.parents if p]
-        commit.author = commit.committer = author
+        commit.author = commit.committer = safe_str(author)
         commit.encoding = ENCODING
-        commit.message = message + ' '
+        commit.message = safe_str(message) + ' '
 
         # Compute date
         if date is None:
             date = time.time()
         elif isinstance(date, datetime.datetime):
             date = time.mktime(date.timetuple())
 
         author_time = kwargs.pop('author_time', date)
         commit.commit_time = int(date)
         commit.author_time = int(author_time)
         tz = time.timezone
         author_tz = kwargs.pop('author_timezone', tz)
         commit.commit_timezone = tz
         commit.author_timezone = author_tz
 
         object_store.add_object(commit)
 
         ref = 'refs/heads/%s' % branch
         repo.refs[ref] = commit.id
         repo.refs.set_symbolic_ref('HEAD', ref)
 
         # Update vcs repository object & recreate dulwich repo
         self.repository.revisions.append(commit.id)
         self.repository._repo = Repo(self.repository.path)
         tip = self.repository.get_changeset()
         self.reset()
         return tip
 
     def _get_missing_trees(self, path, root_tree):
         """
         Creates missing ``Tree`` objects for the given path.
 
         :param path: path given as a string. It may be a path to a file node
           (i.e. ``foo/bar/baz.txt``) or directory path - in that case it must
           end with slash (i.e. ``foo/bar/``).
         :param root_tree: ``dulwich.objects.Tree`` object from which we start
           traversing (should be commit's root tree)
         """
         dirpath = posixpath.split(path)[0]
         dirs = dirpath.split('/')
         if not dirs or dirs == ['']:
             return []
 
         def get_tree_for_dir(tree, dirname):
             for name, mode, id in tree.iteritems():
                 if name == dirname:
                     obj = self.repository._repo[id]
                     if isinstance(obj, objects.Tree):
                         return obj
                     else:
                         raise RepositoryError("Cannot create directory %s "
                         "at tree %s as path is occupied and is not a "
                         "Tree" % (dirname, tree))
             return None
 
         trees = []
         parent = root_tree
         for dirname in dirs:
             tree = get_tree_for_dir(parent, dirname)
             if tree is None:
                 tree = objects.Tree()
                 dirmode = 040000
                 parent.add(dirmode, dirname, tree.id)
                 parent = tree
             # Always append tree
             trees.append(tree)
         return trees

rhodecode/lib/vcs/backends/git/repository.py

@@ -1,521 +1,523 b''
 # -*- coding: utf-8 -*-
 """
     vcs.backends.git
     ~~~~~~~~~~~~~~~~
 
     Git backend implementation.
 
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 
 import os
 import re
 import time
 import posixpath
 from dulwich.repo import Repo, NotGitRepository
 #from dulwich.config import ConfigFile
 from string import Template
 from subprocess import Popen, PIPE
 from rhodecode.lib.vcs.backends.base import BaseRepository
 from rhodecode.lib.vcs.exceptions import BranchDoesNotExistError
 from rhodecode.lib.vcs.exceptions import ChangesetDoesNotExistError
 from rhodecode.lib.vcs.exceptions import EmptyRepositoryError
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.exceptions import TagAlreadyExistError
 from rhodecode.lib.vcs.exceptions import TagDoesNotExistError
 from rhodecode.lib.vcs.utils import safe_unicode, makedate, date_fromtimestamp
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.ordered_dict import OrderedDict
 from rhodecode.lib.vcs.utils.paths import abspath
 from rhodecode.lib.vcs.utils.paths import get_user_home
 from .workdir import GitWorkdir
 from .changeset import GitChangeset
 from .inmemory import GitInMemoryChangeset
 from .config import ConfigFile
 
 
 class GitRepository(BaseRepository):
     """
     Git repository backend.
     """
     DEFAULT_BRANCH_NAME = 'master'
     scm = 'git'
 
     def __init__(self, repo_path, create=False, src_url=None,
                  update_after_clone=False, bare=False):
 
         self.path = abspath(repo_path)
         self._repo = self._get_repo(create, src_url, update_after_clone, bare)
         try:
             self.head = self._repo.head()
         except KeyError:
             self.head = None
 
         self._config_files = [
             bare and abspath(self.path, 'config') or abspath(self.path, '.git',
                 'config'),
             abspath(get_user_home(), '.gitconfig'),
         ]
 
     @LazyProperty
     def revisions(self):
         """
         Returns list of revisions' ids, in ascending order.  Being lazy
         attribute allows external tools to inject shas from cache.
         """
         return self._get_all_revisions()
 
     def run_git_command(self, cmd):
         """
         Runs given ``cmd`` as git command and returns tuple
         (returncode, stdout, stderr).
 
         .. note::
            This method exists only until log/blame functionality is implemented
            at Dulwich (see https://bugs.launchpad.net/bugs/645142). Parsing
            os command's output is road to hell...
 
         :param cmd: git command to be executed
         """
 
-        #cmd = '(cd %s && git %s)' % (self.path, cmd)
+        _copts = ['-c', 'core.quotepath=false', ]
+
         if isinstance(cmd, basestring):
-            cmd = 'GIT_CONFIG_NOGLOBAL=1 git %s' % cmd
-        else:
-            cmd = ['GIT_CONFIG_NOGLOBAL=1', 'git'] + cmd
+            cmd = [cmd]
+
+        cmd = ['GIT_CONFIG_NOGLOBAL=1', 'git'] + _copts + cmd
+
         try:
             opts = dict(
                 shell=isinstance(cmd, basestring),
                 stdout=PIPE,
                 stderr=PIPE)
             if os.path.isdir(self.path):
                 opts['cwd'] = self.path
             p = Popen(cmd, **opts)
         except OSError, err:
             raise RepositoryError("Couldn't run git command (%s).\n"
                 "Original error was:%s" % (cmd, err))
         so, se = p.communicate()
         if not se.startswith("fatal: bad default revision 'HEAD'") and \
             p.returncode != 0:
             raise RepositoryError("Couldn't run git command (%s).\n"
                 "stderr:\n%s" % (cmd, se))
         return so, se
 
     def _check_url(self, url):
         """
         Functon will check given url and try to verify if it's a valid
         link. Sometimes it may happened that mercurial will issue basic
         auth request that can cause whole API to hang when used from python
         or other external calls.
 
         On failures it'll raise urllib2.HTTPError
         """
 
         #TODO: implement this
         pass
 
     def _get_repo(self, create, src_url=None, update_after_clone=False,
             bare=False):
         if create and os.path.exists(self.path):
             raise RepositoryError("Location already exist")
         if src_url and not create:
             raise RepositoryError("Create should be set to True if src_url is "
                                   "given (clone operation creates repository)")
         try:
             if create and src_url:
                 self._check_url(src_url)
                 self.clone(src_url, update_after_clone, bare)
                 return Repo(self.path)
             elif create:
                 os.mkdir(self.path)
                 if bare:
                     return Repo.init_bare(self.path)
                 else:
                     return Repo.init(self.path)
             else:
                 return Repo(self.path)
         except (NotGitRepository, OSError), err:
             raise RepositoryError(err)
 
     def _get_all_revisions(self):
         cmd = 'rev-list --all --date-order'
         try:
             so, se = self.run_git_command(cmd)
         except RepositoryError:
             # Can be raised for empty repositories
             return []
         revisions = so.splitlines()
         revisions.reverse()
         return revisions
 
     def _get_revision(self, revision):
         """
         For git backend we always return integer here. This way we ensure
         that changset's revision attribute would become integer.
         """
         pattern = re.compile(r'^[[0-9a-fA-F]{12}|[0-9a-fA-F]{40}]$')
         is_bstr = lambda o: isinstance(o, (str, unicode))
         is_null = lambda o: len(o) == revision.count('0')
 
         if len(self.revisions) == 0:
             raise EmptyRepositoryError("There are no changesets yet")
 
         if revision in (None, '', 'tip', 'HEAD', 'head', -1):
             revision = self.revisions[-1]
 
         if ((is_bstr(revision) and revision.isdigit() and len(revision) < 12)
             or isinstance(revision, int) or is_null(revision)):
             try:
                 revision = self.revisions[int(revision)]
             except:
                 raise ChangesetDoesNotExistError("Revision %r does not exist "
                     "for this repository %s" % (revision, self))
 
         elif is_bstr(revision):
             if not pattern.match(revision) or revision not in self.revisions:
                 raise ChangesetDoesNotExistError("Revision %r does not exist "
                     "for this repository %s" % (revision, self))
 
         # Ensure we return full id
         if not pattern.match(str(revision)):
             raise ChangesetDoesNotExistError("Given revision %r not recognized"
                 % revision)
         return revision
 
     def _get_archives(self, archive_name='tip'):
 
         for i in [('zip', '.zip'), ('gz', '.tar.gz'), ('bz2', '.tar.bz2')]:
                 yield {"type": i[0], "extension": i[1], "node": archive_name}
 
     def _get_url(self, url):
         """
         Returns normalized url. If schema is not given, would fall to
         filesystem (``file:///``) schema.
         """
         url = str(url)
         if url != 'default' and not '://' in url:
             url = ':///'.join(('file', url))
         return url
 
     @LazyProperty
     def name(self):
         return os.path.basename(self.path)
 
     @LazyProperty
     def last_change(self):
         """
         Returns last change made on this repository as datetime object
         """
         return date_fromtimestamp(self._get_mtime(), makedate()[1])
 
     def _get_mtime(self):
         try:
             return time.mktime(self.get_changeset().date.timetuple())
         except RepositoryError:
             # fallback to filesystem
             in_path = os.path.join(self.path, '.git', "index")
             he_path = os.path.join(self.path, '.git', "HEAD")
             if os.path.exists(in_path):
                 return os.stat(in_path).st_mtime
             else:
                 return os.stat(he_path).st_mtime
 
     @LazyProperty
     def description(self):
         undefined_description = u'unknown'
         description_path = os.path.join(self.path, '.git', 'description')
         if os.path.isfile(description_path):
             return safe_unicode(open(description_path).read())
         else:
             return undefined_description
 
     @LazyProperty
     def contact(self):
         undefined_contact = u'Unknown'
         return undefined_contact
 
     @property
     def branches(self):
         if not self.revisions:
             return {}
         refs = self._repo.refs.as_dict()
         sortkey = lambda ctx: ctx[0]
         _branches = [('/'.join(ref.split('/')[2:]), head)
             for ref, head in refs.items()
             if ref.startswith('refs/heads/') and not ref.endswith('/HEAD')]
         return OrderedDict(sorted(_branches, key=sortkey, reverse=False))
 
     def _heads(self, reverse=False):
         refs = self._repo.get_refs()
         heads = {}
 
         for key, val in refs.items():
             for ref_key in ['refs/heads/', 'refs/remotes/origin/']:
                 if key.startswith(ref_key):
                     n = key[len(ref_key):]
                     if n not in ['HEAD']:
                         heads[n] = val
 
         return heads if reverse else dict((y,x) for x,y in heads.iteritems())
 
     def _get_tags(self):
         if not self.revisions:
             return {}
         sortkey = lambda ctx: ctx[0]
         _tags = [('/'.join(ref.split('/')[2:]), head) for ref, head in
             self._repo.get_refs().items() if ref.startswith('refs/tags/')]
         return OrderedDict(sorted(_tags, key=sortkey, reverse=True))
 
     @LazyProperty
     def tags(self):
         return self._get_tags()
 
     def tag(self, name, user, revision=None, message=None, date=None,
             **kwargs):
         """
         Creates and returns a tag for the given ``revision``.
 
         :param name: name for new tag
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param revision: changeset id for which new tag would be created
         :param message: message of the tag's commit
         :param date: date of tag's commit
 
         :raises TagAlreadyExistError: if tag with same name already exists
         """
         if name in self.tags:
             raise TagAlreadyExistError("Tag %s already exists" % name)
         changeset = self.get_changeset(revision)
         message = message or "Added tag %s for commit %s" % (name,
             changeset.raw_id)
         self._repo.refs["refs/tags/%s" % name] = changeset._commit.id
 
         self.tags = self._get_tags()
         return changeset
 
     def remove_tag(self, name, user, message=None, date=None):
         """
         Removes tag with the given ``name``.
 
         :param name: name of the tag to be removed
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param message: message of the tag's removal commit
         :param date: date of tag's removal commit
 
         :raises TagDoesNotExistError: if tag with given name does not exists
         """
         if name not in self.tags:
             raise TagDoesNotExistError("Tag %s does not exist" % name)
         tagpath = posixpath.join(self._repo.refs.path, 'refs', 'tags', name)
         try:
             os.remove(tagpath)
             self.tags = self._get_tags()
         except OSError, e:
             raise RepositoryError(e.strerror)
 
     def get_changeset(self, revision=None):
         """
         Returns ``GitChangeset`` object representing commit from git repository
         at the given revision or head (most recent commit) if None given.
         """
         if isinstance(revision, GitChangeset):
             return revision
         revision = self._get_revision(revision)
         changeset = GitChangeset(repository=self, revision=revision)
         return changeset
 
     def get_changesets(self, start=None, end=None, start_date=None,
            end_date=None, branch_name=None, reverse=False):
         """
         Returns iterator of ``GitChangeset`` objects from start to end (both
         are inclusive), in ascending date order (unless ``reverse`` is set).
 
         :param start: changeset ID, as str; first returned changeset
         :param end: changeset ID, as str; last returned changeset
         :param start_date: if specified, changesets with commit date less than
           ``start_date`` would be filtered out from returned set
         :param end_date: if specified, changesets with commit date greater than
           ``end_date`` would be filtered out from returned set
         :param branch_name: if specified, changesets not reachable from given
           branch would be filtered out from returned set
         :param reverse: if ``True``, returned generator would be reversed
           (meaning that returned changesets would have descending date order)
 
         :raise BranchDoesNotExistError: If given ``branch_name`` does not
             exist.
         :raise ChangesetDoesNotExistError: If changeset for given ``start`` or
           ``end`` could not be found.
 
         """
         if branch_name and branch_name not in self.branches:
             raise BranchDoesNotExistError("Branch '%s' not found" \
                                           % branch_name)
         # %H at format means (full) commit hash, initial hashes are retrieved
         # in ascending date order
         cmd_template = 'log --date-order --reverse --pretty=format:"%H"'
         cmd_params = {}
         if start_date:
             cmd_template += ' --since "$since"'
             cmd_params['since'] = start_date.strftime('%m/%d/%y %H:%M:%S')
         if end_date:
             cmd_template += ' --until "$until"'
             cmd_params['until'] = end_date.strftime('%m/%d/%y %H:%M:%S')
         if branch_name:
             cmd_template += ' $branch_name'
             cmd_params['branch_name'] = branch_name
         else:
             cmd_template += ' --all'
 
         cmd = Template(cmd_template).safe_substitute(**cmd_params)
         revs = self.run_git_command(cmd)[0].splitlines()
         start_pos = 0
         end_pos = len(revs)
         if start:
             _start = self._get_revision(start)
             try:
                 start_pos = revs.index(_start)
             except ValueError:
                 pass
 
         if end is not None:
             _end = self._get_revision(end)
             try:
                 end_pos = revs.index(_end)
             except ValueError:
                 pass
 
         if None not in [start, end] and start_pos > end_pos:
             raise RepositoryError('start cannot be after end')
 
         if end_pos is not None:
             end_pos += 1
 
         revs = revs[start_pos:end_pos]
         if reverse:
             revs = reversed(revs)
         for rev in revs:
             yield self.get_changeset(rev)
 
     def get_diff(self, rev1, rev2, path=None, ignore_whitespace=False,
             context=3):
         """
         Returns (git like) *diff*, as plain text. Shows changes introduced by
         ``rev2`` since ``rev1``.
 
         :param rev1: Entry point from which diff is shown. Can be
           ``self.EMPTY_CHANGESET`` - in this case, patch showing all
           the changes since empty state of the repository until ``rev2``
         :param rev2: Until which revision changes should be shown.
         :param ignore_whitespace: If set to ``True``, would not show whitespace
           changes. Defaults to ``False``.
         :param context: How many lines before/after changed lines should be
           shown. Defaults to ``3``.
         """
         flags = ['-U%s' % context]
         if ignore_whitespace:
             flags.append('-w')
 
         if rev1 == self.EMPTY_CHANGESET:
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ' '.join(['show'] + flags + [rev2])
         else:
             rev1 = self.get_changeset(rev1).raw_id
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ' '.join(['diff'] + flags + [rev1, rev2])
 
         if path:
             cmd += ' -- "%s"' % path
         stdout, stderr = self.run_git_command(cmd)
         # If we used 'show' command, strip first few lines (until actual diff
         # starts)
         if rev1 == self.EMPTY_CHANGESET:
             lines = stdout.splitlines()
             x = 0
             for line in lines:
                 if line.startswith('diff'):
                     break
                 x += 1
             # Append new line just like 'diff' command do
             stdout = '\n'.join(lines[x:]) + '\n'
         return stdout
 
     @LazyProperty
     def in_memory_changeset(self):
         """
         Returns ``GitInMemoryChangeset`` object for this repository.
         """
         return GitInMemoryChangeset(self)
 
     def clone(self, url, update_after_clone=True, bare=False):
         """
         Tries to clone changes from external location.
 
         :param update_after_clone: If set to ``False``, git won't checkout
           working directory
         :param bare: If set to ``True``, repository would be cloned into
           *bare* git repository (no working directory at all).
         """
         url = self._get_url(url)
         cmd = ['clone']
         if bare:
             cmd.append('--bare')
         elif not update_after_clone:
             cmd.append('--no-checkout')
         cmd += ['--', '"%s"' % url, '"%s"' % self.path]
         cmd = ' '.join(cmd)
         # If error occurs run_git_command raises RepositoryError already
         self.run_git_command(cmd)
 
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         return GitWorkdir(self)
 
     def get_config_value(self, section, name, config_file=None):
         """
         Returns configuration value for a given [``section``] and ``name``.
 
         :param section: Section we want to retrieve value from
         :param name: Name of configuration we want to retrieve
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         if config_file is None:
             config_file = []
         elif isinstance(config_file, basestring):
             config_file = [config_file]
 
         def gen_configs():
             for path in config_file + self._config_files:
                 try:
                     yield ConfigFile.from_path(path)
                 except (IOError, OSError, ValueError):
                     continue
 
         for config in gen_configs():
             try:
                 return config.get(section, name)
             except KeyError:
                 continue
         return None
 
     def get_user_name(self, config_file=None):
         """
         Returns user's name from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'name', config_file)
 
     def get_user_email(self, config_file=None):
         """
         Returns user's email from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'email', config_file)

rhodecode/lib/vcs/backends/hg/inmemory.py

@@ -1,110 +1,112 b''
 import datetime
 import errno
 
 from rhodecode.lib.vcs.backends.base import BaseInMemoryChangeset
 from rhodecode.lib.vcs.exceptions import RepositoryError
 
-from ...utils.hgcompat import memfilectx, memctx, hex
+from ...utils.hgcompat import memfilectx, memctx, hex, tolocal
 
 
 class MercurialInMemoryChangeset(BaseInMemoryChangeset):
 
     def commit(self, message, author, parents=None, branch=None, date=None,
             **kwargs):
         """
         Performs in-memory commit (doesn't check workdir in any way) and
         returns newly created ``Changeset``. Updates repository's
         ``revisions``.
 
         :param message: message of the commit
         :param author: full username, i.e. "Joe Doe <[e-mail]>"
         :param parents: single parent or sequence of parents from which commit
           would be derieved
         :param date: ``datetime.datetime`` instance. Defaults to
           ``datetime.datetime.now()``.
         :param branch: branch name, as string. If none given, default backend's
           branch would be used.
 
         :raises ``CommitError``: if any error occurs while committing
         """
         self.check_integrity(parents)
 
         from .repository import MercurialRepository
-        if not isinstance(message, str) or not isinstance(author, str):
+        if not isinstance(message, unicode) or not isinstance(author, unicode):
             raise RepositoryError('Given message and author needs to be '
-                                  'an <str> instance')
+                                  'an <unicode> instance')
 
         if branch is None:
             branch = MercurialRepository.DEFAULT_BRANCH_NAME
         kwargs['branch'] = branch
 
         def filectxfn(_repo, memctx, path):
             """
             Marks given path as added/changed/removed in a given _repo. This is
             for internal mercurial commit function.
             """
 
             # check if this path is removed
             if path in (node.path for node in self.removed):
                 # Raising exception is a way to mark node for removal
                 raise IOError(errno.ENOENT, '%s is deleted' % path)
 
             # check if this path is added
             for node in self.added:
                 if node.path == path:
                     return memfilectx(path=node.path,
                         data=(node.content.encode('utf8')
                               if not node.is_binary else node.content),
                         islink=False,
                         isexec=node.is_executable,
                         copied=False)
 
             # or changed
             for node in self.changed:
                 if node.path == path:
                     return memfilectx(path=node.path,
                         data=(node.content.encode('utf8')
                               if not node.is_binary else node.content),
                         islink=False,
                         isexec=node.is_executable,
                         copied=False)
 
             raise RepositoryError("Given path haven't been marked as added,"
-                "changed or removed (%s)" % path)
+                                  "changed or removed (%s)" % path)
 
         parents = [None, None]
         for i, parent in enumerate(self.parents):
             if parent is not None:
                 parents[i] = parent._ctx.node()
 
         if date and isinstance(date, datetime.datetime):
             date = date.ctime()
 
         commit_ctx = memctx(repo=self.repository._repo,
             parents=parents,
             text='',
             files=self.get_paths(),
             filectxfn=filectxfn,
             user=author,
             date=date,
             extra=kwargs)
 
+        loc = lambda u: tolocal(u.encode('utf-8'))
+
         # injecting given _repo params
-        commit_ctx._text = message
-        commit_ctx._user = author
+        commit_ctx._text = loc(message)
+        commit_ctx._user = loc(author)
         commit_ctx._date = date
 
         # TODO: Catch exceptions!
         n = self.repository._repo.commitctx(commit_ctx)
         # Returns mercurial node
         self._commit_ctx = commit_ctx  # For reference
         # Update vcs repository object & recreate mercurial _repo
         # new_ctx = self.repository._repo[node]
         # new_tip = self.repository.get_changeset(new_ctx.hex())
         new_id = hex(n)
         self.repository.revisions.append(new_id)
         self._repo = self.repository._get_repo(create=False)
         self.repository.branches = self.repository._get_branches()
         tip = self.repository.get_changeset()
         self.reset()
         return tip

rhodecode/lib/vcs/nodes.py

@@ -1,559 +1,563 b''
 # -*- coding: utf-8 -*-
 """
     vcs.nodes
     ~~~~~~~~~
 
     Module holding everything related to vcs nodes.
 
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import stat
 import posixpath
 import mimetypes
 
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils import safe_unicode
 from rhodecode.lib.vcs.exceptions import NodeError
 from rhodecode.lib.vcs.exceptions import RemovedFileNodeError
 
 from pygments import lexers
 
 
 class NodeKind:
     DIR = 1
     FILE = 2
 
 
 class NodeState:
     ADDED = u'added'
     CHANGED = u'changed'
     NOT_CHANGED = u'not changed'
     REMOVED = u'removed'
 
 
 class NodeGeneratorBase(object):
     """
     Base class for removed added and changed filenodes, it's a lazy generator
     class that will create filenodes only on iteration or call
 
     The len method doesn't need to create filenodes at all
     """
 
     def __init__(self, current_paths, cs):
         self.cs = cs
         self.current_paths = current_paths
 
     def __call__(self):
         return [n for n in self]
 
     def __getslice__(self, i, j):
         for p in self.current_paths[i:j]:
             yield self.cs.get_node(p)
 
     def __len__(self):
         return len(self.current_paths)
 
     def __iter__(self):
         for p in self.current_paths:
             yield self.cs.get_node(p)
 
 
 class AddedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding Added files for current changeset
     """
     pass
 
 
 class ChangedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding Changed files for current changeset
     """
     pass
 
 
 class RemovedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding removed files for current changeset
     """
     def __iter__(self):
         for p in self.current_paths:
             yield RemovedFileNode(path=p)
 
     def __getslice__(self, i, j):
         for p in self.current_paths[i:j]:
             yield RemovedFileNode(path=p)
 
 
 class Node(object):
     """
     Simplest class representing file or directory on repository.  SCM backends
     should use ``FileNode`` and ``DirNode`` subclasses rather than ``Node``
     directly.
 
     Node's ``path`` cannot start with slash as we operate on *relative* paths
     only. Moreover, every single node is identified by the ``path`` attribute,
     so it cannot end with slash, too. Otherwise, path could lead to mistakes.
     """
 
     def __init__(self, path, kind):
         if path.startswith('/'):
             raise NodeError("Cannot initialize Node objects with slash at "
                 "the beginning as only relative paths are supported")
         self.path = path.rstrip('/')
         if path == '' and kind != NodeKind.DIR:
             raise NodeError("Only DirNode and its subclasses may be "
                             "initialized with empty path")
         self.kind = kind
         #self.dirs, self.files = [], []
         if self.is_root() and not self.is_dir():
             raise NodeError("Root node cannot be FILE kind")
 
     @LazyProperty
     def parent(self):
         parent_path = self.get_parent_path()
         if parent_path:
             if self.changeset:
                 return self.changeset.get_node(parent_path)
             return DirNode(parent_path)
         return None
 
     @LazyProperty
+    def unicode_path(self):
+        return safe_unicode(self.path)
+
+    @LazyProperty
     def name(self):
         """
         Returns name of the node so if its path
         then only last part is returned.
         """
         return safe_unicode(self.path.rstrip('/').split('/')[-1])
 
     def _get_kind(self):
         return self._kind
 
     def _set_kind(self, kind):
         if hasattr(self, '_kind'):
             raise NodeError("Cannot change node's kind")
         else:
             self._kind = kind
             # Post setter check (path's trailing slash)
             if self.path.endswith('/'):
                 raise NodeError("Node's path cannot end with slash")
 
     kind = property(_get_kind, _set_kind)
 
     def __cmp__(self, other):
         """
         Comparator using name of the node, needed for quick list sorting.
         """
         kind_cmp = cmp(self.kind, other.kind)
         if kind_cmp:
             return kind_cmp
         return cmp(self.name, other.name)
 
     def __eq__(self, other):
         for attr in ['name', 'path', 'kind']:
             if getattr(self, attr) != getattr(other, attr):
                 return False
         if self.is_file():
             if self.content != other.content:
                 return False
         else:
             # For DirNode's check without entering each dir
             self_nodes_paths = list(sorted(n.path for n in self.nodes))
             other_nodes_paths = list(sorted(n.path for n in self.nodes))
             if self_nodes_paths != other_nodes_paths:
                 return False
         return True
 
     def __nq__(self, other):
         return not self.__eq__(other)
 
     def __repr__(self):
         return '<%s %r>' % (self.__class__.__name__, self.path)
 
     def __str__(self):
         return self.__repr__()
 
     def __unicode__(self):
         return self.name
 
     def get_parent_path(self):
         """
         Returns node's parent path or empty string if node is root.
         """
         if self.is_root():
             return ''
         return posixpath.dirname(self.path.rstrip('/')) + '/'
 
     def is_file(self):
         """
         Returns ``True`` if node's kind is ``NodeKind.FILE``, ``False``
         otherwise.
         """
         return self.kind == NodeKind.FILE
 
     def is_dir(self):
         """
         Returns ``True`` if node's kind is ``NodeKind.DIR``, ``False``
         otherwise.
         """
         return self.kind == NodeKind.DIR
 
     def is_root(self):
         """
         Returns ``True`` if node is a root node and ``False`` otherwise.
         """
         return self.kind == NodeKind.DIR and self.path == ''
 
     @LazyProperty
     def added(self):
         return self.state is NodeState.ADDED
 
     @LazyProperty
     def changed(self):
         return self.state is NodeState.CHANGED
 
     @LazyProperty
     def not_changed(self):
         return self.state is NodeState.NOT_CHANGED
 
     @LazyProperty
     def removed(self):
         return self.state is NodeState.REMOVED
 
 
 class FileNode(Node):
     """
     Class representing file nodes.
 
     :attribute: path: path to the node, relative to repostiory's root
     :attribute: content: if given arbitrary sets content of the file
     :attribute: changeset: if given, first time content is accessed, callback
     :attribute: mode: octal stat mode for a node. Default is 0100644.
     """
 
     def __init__(self, path, content=None, changeset=None, mode=None):
         """
         Only one of ``content`` and ``changeset`` may be given. Passing both
         would raise ``NodeError`` exception.
 
         :param path: relative path to the node
         :param content: content may be passed to constructor
         :param changeset: if given, will use it to lazily fetch content
         :param mode: octal representation of ST_MODE (i.e. 0100644)
         """
 
         if content and changeset:
             raise NodeError("Cannot use both content and changeset")
         super(FileNode, self).__init__(path, kind=NodeKind.FILE)
         self.changeset = changeset
         self._content = content
         self._mode = mode or 0100644
 
     @LazyProperty
     def mode(self):
         """
         Returns lazily mode of the FileNode. If ``changeset`` is not set, would
         use value given at initialization or 0100644 (default).
         """
         if self.changeset:
             mode = self.changeset.get_file_mode(self.path)
         else:
             mode = self._mode
         return mode
 
     @property
     def content(self):
         """
         Returns lazily content of the FileNode. If possible, would try to
         decode content from UTF-8.
         """
         if self.changeset:
             content = self.changeset.get_file_content(self.path)
         else:
             content = self._content
 
         if bool(content and '\0' in content):
             return content
         return safe_unicode(content)
 
     @LazyProperty
     def size(self):
         if self.changeset:
             return self.changeset.get_file_size(self.path)
         raise NodeError("Cannot retrieve size of the file without related "
             "changeset attribute")
 
     @LazyProperty
     def message(self):
         if self.changeset:
             return self.last_changeset.message
         raise NodeError("Cannot retrieve message of the file without related "
             "changeset attribute")
 
     @LazyProperty
     def last_changeset(self):
         if self.changeset:
             return self.changeset.get_file_changeset(self.path)
         raise NodeError("Cannot retrieve last changeset of the file without "
             "related changeset attribute")
 
     def get_mimetype(self):
         """
         Mimetype is calculated based on the file's content. If ``_mimetype``
         attribute is available, it will be returned (backends which store
         mimetypes or can easily recognize them, should set this private
         attribute to indicate that type should *NOT* be calculated).
         """
         if hasattr(self, '_mimetype'):
             if (isinstance(self._mimetype, (tuple, list,)) and
                 len(self._mimetype) == 2):
                 return self._mimetype
             else:
                 raise NodeError('given _mimetype attribute must be an 2 '
                                'element list or tuple')
 
         mtype, encoding = mimetypes.guess_type(self.name)
 
         if mtype is None:
             if self.is_binary:
                 mtype = 'application/octet-stream'
                 encoding = None
             else:
                 mtype = 'text/plain'
                 encoding = None
         return mtype, encoding
 
     @LazyProperty
     def mimetype(self):
         """
         Wrapper around full mimetype info. It returns only type of fetched
         mimetype without the encoding part. use get_mimetype function to fetch
         full set of (type,encoding)
         """
         return self.get_mimetype()[0]
 
     @LazyProperty
     def mimetype_main(self):
         return self.mimetype.split('/')[0]
 
     @LazyProperty
     def lexer(self):
         """
         Returns pygment's lexer class. Would try to guess lexer taking file's
         content, name and mimetype.
         """
         try:
             lexer = lexers.guess_lexer_for_filename(self.name, self.content)
         except lexers.ClassNotFound:
             lexer = lexers.TextLexer()
         # returns first alias
         return lexer
 
     @LazyProperty
     def lexer_alias(self):
         """
         Returns first alias of the lexer guessed for this file.
         """
         return self.lexer.aliases[0]
 
     @LazyProperty
     def history(self):
         """
         Returns a list of changeset for this file in which the file was changed
         """
         if self.changeset is None:
             raise NodeError('Unable to get changeset for this FileNode')
         return self.changeset.get_file_history(self.path)
 
     @LazyProperty
     def annotate(self):
         """
         Returns a list of three element tuples with lineno,changeset and line
         """
         if self.changeset is None:
             raise NodeError('Unable to get changeset for this FileNode')
         return self.changeset.get_file_annotate(self.path)
 
     @LazyProperty
     def state(self):
         if not self.changeset:
             raise NodeError("Cannot check state of the node if it's not "
                 "linked with changeset")
         elif self.path in (node.path for node in self.changeset.added):
             return NodeState.ADDED
         elif self.path in (node.path for node in self.changeset.changed):
             return NodeState.CHANGED
         else:
             return NodeState.NOT_CHANGED
 
     @property
     def is_binary(self):
         """
         Returns True if file has binary content.
         """
         _bin = '\0' in self.content
         return _bin
 
     @LazyProperty
     def extension(self):
         """Returns filenode extension"""
         return self.name.split('.')[-1]
 
     def is_executable(self):
         """
         Returns ``True`` if file has executable flag turned on.
         """
         return bool(self.mode & stat.S_IXUSR)
 
     def __repr__(self):
         return '<%s %r @ %s>' % (self.__class__.__name__, self.path,
                                  self.changeset.short_id)
 
 
 class RemovedFileNode(FileNode):
     """
     Dummy FileNode class - trying to access any public attribute except path,
     name, kind or state (or methods/attributes checking those two) would raise
     RemovedFileNodeError.
     """
     ALLOWED_ATTRIBUTES = ['name', 'path', 'state', 'is_root', 'is_file',
         'is_dir', 'kind', 'added', 'changed', 'not_changed', 'removed']
 
     def __init__(self, path):
         """
         :param path: relative path to the node
         """
         super(RemovedFileNode, self).__init__(path=path)
 
     def __getattribute__(self, attr):
         if attr.startswith('_') or attr in RemovedFileNode.ALLOWED_ATTRIBUTES:
             return super(RemovedFileNode, self).__getattribute__(attr)
         raise RemovedFileNodeError("Cannot access attribute %s on "
             "RemovedFileNode" % attr)
 
     @LazyProperty
     def state(self):
         return NodeState.REMOVED
 
 
 class DirNode(Node):
     """
     DirNode stores list of files and directories within this node.
     Nodes may be used standalone but within repository context they
     lazily fetch data within same repositorty's changeset.
     """
 
     def __init__(self, path, nodes=(), changeset=None):
         """
         Only one of ``nodes`` and ``changeset`` may be given. Passing both
         would raise ``NodeError`` exception.
 
         :param path: relative path to the node
         :param nodes: content may be passed to constructor
         :param changeset: if given, will use it to lazily fetch content
         :param size: always 0 for ``DirNode``
         """
         if nodes and changeset:
             raise NodeError("Cannot use both nodes and changeset")
         super(DirNode, self).__init__(path, NodeKind.DIR)
         self.changeset = changeset
         self._nodes = nodes
 
     @LazyProperty
     def content(self):
         raise NodeError("%s represents a dir and has no ``content`` attribute"
             % self)
 
     @LazyProperty
     def nodes(self):
         if self.changeset:
             nodes = self.changeset.get_nodes(self.path)
         else:
             nodes = self._nodes
         self._nodes_dict = dict((node.path, node) for node in nodes)
         return sorted(nodes)
 
     @LazyProperty
     def files(self):
         return sorted((node for node in self.nodes if node.is_file()))
 
     @LazyProperty
     def dirs(self):
         return sorted((node for node in self.nodes if node.is_dir()))
 
     def __iter__(self):
         for node in self.nodes:
             yield node
 
     def get_node(self, path):
         """
         Returns node from within this particular ``DirNode``, so it is now
         allowed to fetch, i.e. node located at 'docs/api/index.rst' from node
         'docs'. In order to access deeper nodes one must fetch nodes between
         them first - this would work::
 
            docs = root.get_node('docs')
            docs.get_node('api').get_node('index.rst')
 
         :param: path - relative to the current node
 
         .. note::
            To access lazily (as in example above) node have to be initialized
            with related changeset object - without it node is out of
            context and may know nothing about anything else than nearest
            (located at same level) nodes.
         """
         try:
             path = path.rstrip('/')
             if path == '':
                 raise NodeError("Cannot retrieve node without path")
             self.nodes  # access nodes first in order to set _nodes_dict
             paths = path.split('/')
             if len(paths) == 1:
                 if not self.is_root():
                     path = '/'.join((self.path, paths[0]))
                 else:
                     path = paths[0]
                 return self._nodes_dict[path]
             elif len(paths) > 1:
                 if self.changeset is None:
                     raise NodeError("Cannot access deeper "
                                     "nodes without changeset")
                 else:
                     path1, path2 = paths[0], '/'.join(paths[1:])
                     return self.get_node(path1).get_node(path2)
             else:
                 raise KeyError
         except KeyError:
             raise NodeError("Node does not exist at %s" % path)
 
     @LazyProperty
     def state(self):
         raise NodeError("Cannot access state of DirNode")
 
     @LazyProperty
     def size(self):
         size = 0
         for root, dirs, files in self.changeset.walk(self.path):
             for f in files:
                 size += f.size
 
         return size
 
     def __repr__(self):
         return '<%s %r @ %s>' % (self.__class__.__name__, self.path,
                                  self.changeset.short_id)
 
 
 class RootNode(DirNode):
     """
     DirNode being the root node of the repository.
     """
 
     def __init__(self, nodes=(), changeset=None):
         super(RootNode, self).__init__(path='', nodes=nodes,
             changeset=changeset)
 
     def __repr__(self):
         return '<%s>' % self.__class__.__name__

rhodecode/lib/vcs/utils/hgcompat.py

@@ -1,12 +1,14 b''
-"""Mercurial libs compatibility
+"""
+Mercurial libs compatibility
+"""
 
-"""
 from mercurial import archival, merge as hg_merge, patch, ui
 from mercurial.commands import clone, nullid, pull
 from mercurial.context import memctx, memfilectx
 from mercurial.error import RepoError, RepoLookupError, Abort
 from mercurial.hgweb.common import get_contact
 from mercurial.localrepo import localrepository
 from mercurial.match import match
 from mercurial.mdiff import diffopts
 from mercurial.node import hex
+from mercurial.encoding import tolocal
  No newline at end of file

rhodecode/model/scm.py

@@ -1,459 +1,461 b''
 # -*- coding: utf-8 -*-
 """
     rhodecode.model.scm
     ~~~~~~~~~~~~~~~~~~~
 
     Scm model for RhodeCode
 
     :created_on: Apr 9, 2010
     :author: marcink
     :copyright: (C) 2010-2012 Marcin Kuzminski <[e-mail]>
     :license: GPLv3, see COPYING for more details.
 """
 # 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/>.
 import os
 import time
 import traceback
 import logging
 import cStringIO
 
 from rhodecode.lib.vcs import get_backend
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.nodes import FileNode
 
 from rhodecode import BACKENDS
 from rhodecode.lib import helpers as h
-from rhodecode.lib.utils2 import safe_str
+from rhodecode.lib.utils2 import safe_str, safe_unicode
 from rhodecode.lib.auth import HasRepoPermissionAny, HasReposGroupPermissionAny
 from rhodecode.lib.utils import get_repos as get_filesystem_repos, make_ui, \
     action_logger, EmptyChangeset, REMOVED_REPO_PAT
 from rhodecode.model import BaseModel
 from rhodecode.model.db import Repository, RhodeCodeUi, CacheInvalidation, \
     UserFollowing, UserLog, User, RepoGroup
 
 log = logging.getLogger(__name__)
 
 
 class UserTemp(object):
     def __init__(self, user_id):
         self.user_id = user_id
 
     def __repr__(self):
         return "<%s('id:%s')>" % (self.__class__.__name__, self.user_id)
 
 
 class RepoTemp(object):
     def __init__(self, repo_id):
         self.repo_id = repo_id
 
     def __repr__(self):
         return "<%s('id:%s')>" % (self.__class__.__name__, self.repo_id)
 
 
 class CachedRepoList(object):
 
     def __init__(self, db_repo_list, repos_path, order_by=None):
         self.db_repo_list = db_repo_list
         self.repos_path = repos_path
         self.order_by = order_by
         self.reversed = (order_by or '').startswith('-')
 
     def __len__(self):
         return len(self.db_repo_list)
 
     def __repr__(self):
         return '<%s (%s)>' % (self.__class__.__name__, self.__len__())
 
     def __iter__(self):
         for dbr in self.db_repo_list:
             scmr = dbr.scm_instance_cached
             # check permission at this level
             if not HasRepoPermissionAny(
                 'repository.read', 'repository.write', 'repository.admin'
             )(dbr.repo_name, 'get repo check'):
                 continue
 
             if scmr is None:
                 log.error(
                     '%s this repository is present in database but it '
                     'cannot be created as an scm instance' % dbr.repo_name
                 )
                 continue
 
             last_change = scmr.last_change
             tip = h.get_changeset_safe(scmr, 'tip')
 
             tmp_d = {}
             tmp_d['name'] = dbr.repo_name
             tmp_d['name_sort'] = tmp_d['name'].lower()
             tmp_d['description'] = dbr.description
             tmp_d['description_sort'] = tmp_d['description']
             tmp_d['last_change'] = last_change
             tmp_d['last_change_sort'] = time.mktime(last_change.timetuple())
             tmp_d['tip'] = tip.raw_id
             tmp_d['tip_sort'] = tip.revision
             tmp_d['rev'] = tip.revision
             tmp_d['contact'] = dbr.user.full_contact
             tmp_d['contact_sort'] = tmp_d['contact']
             tmp_d['owner_sort'] = tmp_d['contact']
             tmp_d['repo_archives'] = list(scmr._get_archives())
             tmp_d['last_msg'] = tip.message
             tmp_d['author'] = tip.author
             tmp_d['dbrepo'] = dbr.get_dict()
             tmp_d['dbrepo_fork'] = dbr.fork.get_dict() if dbr.fork else {}
             yield tmp_d
 
 
 class GroupList(object):
 
     def __init__(self, db_repo_group_list):
         self.db_repo_group_list = db_repo_group_list
 
     def __len__(self):
         return len(self.db_repo_group_list)
 
     def __repr__(self):
         return '<%s (%s)>' % (self.__class__.__name__, self.__len__())
 
     def __iter__(self):
         for dbgr in self.db_repo_group_list:
             # check permission at this level
             if not HasReposGroupPermissionAny(
                 'group.read', 'group.write', 'group.admin'
             )(dbgr.group_name, 'get group repo check'):
                 continue
 
             yield dbgr
 
 
 class ScmModel(BaseModel):
     """
     Generic Scm Model
     """
 
     def __get_repo(self, instance):
         cls = Repository
         if isinstance(instance, cls):
             return instance
         elif isinstance(instance, int) or str(instance).isdigit():
             return cls.get(instance)
         elif isinstance(instance, basestring):
             return cls.get_by_repo_name(instance)
         elif instance:
             raise Exception('given object must be int, basestr or Instance'
                             ' of %s got %s' % (type(cls), type(instance)))
 
     @LazyProperty
     def repos_path(self):
         """
         Get's the repositories root path from database
         """
 
         q = self.sa.query(RhodeCodeUi).filter(RhodeCodeUi.ui_key == '/').one()
 
         return q.ui_value
 
     def repo_scan(self, repos_path=None):
         """
         Listing of repositories in given path. This path should not be a
         repository itself. Return a dictionary of repository objects
 
         :param repos_path: path to directory containing repositories
         """
 
         if repos_path is None:
             repos_path = self.repos_path
 
         log.info('scanning for repositories in %s' % repos_path)
 
         baseui = make_ui('db')
         repos = {}
 
         for name, path in get_filesystem_repos(repos_path, recursive=True):
             # skip removed repos
             if REMOVED_REPO_PAT.match(name):
                 continue
 
             # name need to be decomposed and put back together using the /
             # since this is internal storage separator for rhodecode
             name = Repository.url_sep().join(name.split(os.sep))
 
             try:
                 if name in repos:
                     raise RepositoryError('Duplicate repository name %s '
                                           'found in %s' % (name, path))
                 else:
 
                     klass = get_backend(path[0])
 
                     if path[0] == 'hg' and path[0] in BACKENDS.keys():
                         repos[name] = klass(safe_str(path[1]), baseui=baseui)
 
                     if path[0] == 'git' and path[0] in BACKENDS.keys():
                         repos[name] = klass(path[1])
             except OSError:
                 continue
 
         return repos
 
     def get_repos(self, all_repos=None, sort_key=None):
         """
         Get all repos from db and for each repo create it's
         backend instance and fill that backed with information from database
 
         :param all_repos: list of repository names as strings
             give specific repositories list, good for filtering
         """
         if all_repos is None:
             all_repos = self.sa.query(Repository)\
                         .filter(Repository.group_id == None)\
                         .order_by(Repository.repo_name).all()
 
         repo_iter = CachedRepoList(all_repos, repos_path=self.repos_path,
                                    order_by=sort_key)
 
         return repo_iter
 
     def get_repos_groups(self, all_groups=None):
         if all_groups is None:
             all_groups = RepoGroup.query()\
                 .filter(RepoGroup.group_parent_id == None).all()
         group_iter = GroupList(all_groups)
 
         return group_iter
 
     def mark_for_invalidation(self, repo_name):
         """
         Puts cache invalidation task into db for
         further global cache invalidation
 
         :param repo_name: this repo that should invalidation take place
         """
         CacheInvalidation.set_invalidate(repo_name)
 
     def toggle_following_repo(self, follow_repo_id, user_id):
 
         f = self.sa.query(UserFollowing)\
             .filter(UserFollowing.follows_repo_id == follow_repo_id)\
             .filter(UserFollowing.user_id == user_id).scalar()
 
         if f is not None:
             try:
                 self.sa.delete(f)
                 action_logger(UserTemp(user_id),
                               'stopped_following_repo',
                               RepoTemp(follow_repo_id))
                 return
             except:
                 log.error(traceback.format_exc())
                 raise
 
         try:
             f = UserFollowing()
             f.user_id = user_id
             f.follows_repo_id = follow_repo_id
             self.sa.add(f)
 
             action_logger(UserTemp(user_id),
                           'started_following_repo',
                           RepoTemp(follow_repo_id))
         except:
             log.error(traceback.format_exc())
             raise
 
     def toggle_following_user(self, follow_user_id, user_id):
         f = self.sa.query(UserFollowing)\
             .filter(UserFollowing.follows_user_id == follow_user_id)\
             .filter(UserFollowing.user_id == user_id).scalar()
 
         if f is not None:
             try:
                 self.sa.delete(f)
                 return
             except:
                 log.error(traceback.format_exc())
                 raise
 
         try:
             f = UserFollowing()
             f.user_id = user_id
             f.follows_user_id = follow_user_id
             self.sa.add(f)
         except:
             log.error(traceback.format_exc())
             raise
 
     def is_following_repo(self, repo_name, user_id, cache=False):
         r = self.sa.query(Repository)\
             .filter(Repository.repo_name == repo_name).scalar()
 
         f = self.sa.query(UserFollowing)\
             .filter(UserFollowing.follows_repository == r)\
             .filter(UserFollowing.user_id == user_id).scalar()
 
         return f is not None
 
     def is_following_user(self, username, user_id, cache=False):
         u = User.get_by_username(username)
 
         f = self.sa.query(UserFollowing)\
             .filter(UserFollowing.follows_user == u)\
             .filter(UserFollowing.user_id == user_id).scalar()
 
         return f is not None
 
     def get_followers(self, repo_id):
         if not isinstance(repo_id, int):
             repo_id = getattr(Repository.get_by_repo_name(repo_id), 'repo_id')
 
         return self.sa.query(UserFollowing)\
                 .filter(UserFollowing.follows_repo_id == repo_id).count()
 
     def get_forks(self, repo_id):
         if not isinstance(repo_id, int):
             repo_id = getattr(Repository.get_by_repo_name(repo_id), 'repo_id')
 
         return self.sa.query(Repository)\
                 .filter(Repository.fork_id == repo_id).count()
 
     def mark_as_fork(self, repo, fork, user):
         repo = self.__get_repo(repo)
         fork = self.__get_repo(fork)
         repo.fork = fork
         self.sa.add(repo)
         return repo
 
     def pull_changes(self, repo_name, username):
         dbrepo = Repository.get_by_repo_name(repo_name)
         clone_uri = dbrepo.clone_uri
         if not clone_uri:
             raise Exception("This repository doesn't have a clone uri")
 
         repo = dbrepo.scm_instance
         try:
             extras = {'ip': '',
                       'username': username,
                       'action': 'push_remote',
                       'repository': repo_name}
 
             #inject ui extra param to log this action via push logger
             for k, v in extras.items():
                 repo._repo.ui.setconfig('rhodecode_extras', k, v)
 
             repo.pull(clone_uri)
             self.mark_for_invalidation(repo_name)
         except:
             log.error(traceback.format_exc())
             raise
 
     def commit_change(self, repo, repo_name, cs, user, author, message,
                       content, f_path):
 
         if repo.alias == 'hg':
             from rhodecode.lib.vcs.backends.hg import MercurialInMemoryChangeset as IMC
         elif repo.alias == 'git':
             from rhodecode.lib.vcs.backends.git import GitInMemoryChangeset as IMC
 
         # decoding here will force that we have proper encoded values
         # in any other case this will throw exceptions and deny commit
         content = safe_str(content)
-        message = safe_str(message)
         path = safe_str(f_path)
-        author = safe_str(author)
+        # message and author needs to be unicode
+        # proper backend should then translate that into required type
+        message = safe_unicode(message)
+        author = safe_unicode(author)
         m = IMC(repo)
         m.change(FileNode(path, content))
         tip = m.commit(message=message,
-                 author=author,
-                 parents=[cs], branch=cs.branch)
+                       author=author,
+                       parents=[cs], branch=cs.branch)
 
         new_cs = tip.short_id
         action = 'push_local:%s' % new_cs
 
         action_logger(user, action, repo_name)
 
         self.mark_for_invalidation(repo_name)
 
     def create_node(self, repo, repo_name, cs, user, author, message, content,
                       f_path):
         if repo.alias == 'hg':
             from rhodecode.lib.vcs.backends.hg import MercurialInMemoryChangeset as IMC
         elif repo.alias == 'git':
             from rhodecode.lib.vcs.backends.git import GitInMemoryChangeset as IMC
         # decoding here will force that we have proper encoded values
         # in any other case this will throw exceptions and deny commit
 
         if isinstance(content, (basestring,)):
             content = safe_str(content)
         elif isinstance(content, (file, cStringIO.OutputType,)):
             content = content.read()
         else:
             raise Exception('Content is of unrecognized type %s' % (
                 type(content)
             ))
 
-        message = safe_str(message)
+        message = safe_unicode(message)
+        author = safe_unicode(author)
         path = safe_str(f_path)
-        author = safe_str(author)
         m = IMC(repo)
 
         if isinstance(cs, EmptyChangeset):
-            # Emptychangeset means we we're editing empty repository
+            # EmptyChangeset means we we're editing empty repository
             parents = None
         else:
             parents = [cs]
 
         m.add(FileNode(path, content=content))
         tip = m.commit(message=message,
-                 author=author,
-                 parents=parents, branch=cs.branch)
+                       author=author,
+                       parents=parents, branch=cs.branch)
         new_cs = tip.short_id
         action = 'push_local:%s' % new_cs
 
         action_logger(user, action, repo_name)
 
         self.mark_for_invalidation(repo_name)
 
     def get_nodes(self, repo_name, revision, root_path='/', flat=True):
         """
         recursive walk in root dir and return a set of all path in that dir
         based on repository walk function
 
         :param repo_name: name of repository
         :param revision: revision for which to list nodes
         :param root_path: root path to list
         :param flat: return as a list, if False returns a dict with decription
 
         """
         _files = list()
         _dirs = list()
         try:
             _repo = self.__get_repo(repo_name)
             changeset = _repo.scm_instance.get_changeset(revision)
             root_path = root_path.lstrip('/')
             for topnode, dirs, files in changeset.walk(root_path):
                 for f in files:
                     _files.append(f.path if flat else {"name": f.path,
                                                        "type": "file"})
                 for d in dirs:
                     _dirs.append(d.path if flat else {"name": d.path,
                                                       "type": "dir"})
         except RepositoryError:
             log.debug(traceback.format_exc())
             raise
 
         return _dirs, _files
 
     def get_unread_journal(self):
         return self.sa.query(UserLog).count()

rhodecode/templates/files/files_edit.html

@@ -1,78 +1,78 b''
 <%inherit file="/base/base.html"/>
 
 <%def name="title()">
     ${c.repo_name} ${_('Edit file')} - ${c.rhodecode_name}
 </%def>
 
 <%def name="js_extra()">
 <script type="text/javascript" src="${h.url('/js/codemirror.js')}"></script>
 </%def>
 <%def name="css_extra()">
 <link rel="stylesheet" type="text/css" href="${h.url('/css/codemirror.css')}"/>
 </%def>
 
 <%def name="breadcrumbs_links()">
     ${h.link_to(u'Home',h.url('/'))}
     &raquo;
     ${h.link_to(c.repo_name,h.url('summary_home',repo_name=c.repo_name))}
     &raquo;
     ${_('edit file')} @ R${c.cs.revision}:${h.short_id(c.cs.raw_id)}
 </%def>
 
 <%def name="page_nav()">
 		${self.menu('files')}
 </%def>
 <%def name="main()">
 <div class="box">
     <!-- box / title -->
     <div class="title">
         ${self.breadcrumbs()}
         <ul class="links">
             <li>
               <span style="text-transform: uppercase;">
               <a href="#">${_('branch')}: ${c.cs.branch}</a></span>
             </li>
         </ul>
     </div>
     <div class="table">
 		<div id="files_data">
 			<h3 class="files_location">${_('Location')}: ${h.files_breadcrumbs(c.repo_name,c.cs.revision,c.file.path)}</h3>
 			${h.form(h.url.current(),method='post',id='eform')}
 			<div id="body" class="codeblock">
             <div class="code-header">
                 <div class="stats">
                     <div class="left"><img src="${h.url('/images/icons/file.png')}"/></div>
                     <div class="left item">${h.link_to("r%s:%s" % (c.file.changeset.revision,h.short_id(c.file.changeset.raw_id)),h.url('changeset_home',repo_name=c.repo_name,revision=c.file.changeset.raw_id))}</div>
                     <div class="left item">${h.format_byte_size(c.file.size,binary=True)}</div>
                     <div class="left item last">${c.file.mimetype}</div>
                     <div class="buttons">
                       ${h.link_to(_('show annotation'),h.url('files_annotate_home',repo_name=c.repo_name,revision=c.cs.raw_id,f_path=c.f_path),class_="ui-btn")}
                       ${h.link_to(_('show as raw'),h.url('files_raw_home',repo_name=c.repo_name,revision=c.cs.raw_id,f_path=c.f_path),class_="ui-btn")}
                       ${h.link_to(_('download as raw'),h.url('files_rawfile_home',repo_name=c.repo_name,revision=c.cs.raw_id,f_path=c.f_path),class_="ui-btn")}
                       % if h.HasRepoPermissionAny('repository.write','repository.admin')(c.repo_name):
                        % if not c.file.is_binary:
                         ${h.link_to(_('source'),h.url('files_home',repo_name=c.repo_name,revision=c.cs.raw_id,f_path=c.f_path),class_="ui-btn")}
                        % endif
                       % endif
                     </div>
                 </div>
-                <div class="commit">${_('Editing file')}: ${c.file.path}</div>
+                <div class="commit">${_('Editing file')}: ${c.file.unicode_path}</div>
             </div>
 			    <pre id="editor_pre"></pre>
 				<textarea id="editor" name="content" style="display:none">${h.escape(c.file.content)|n}</textarea>
 				<div style="padding: 10px;color:#666666">${_('commit message')}</div>
 				<textarea id="commit" name="message" style="height: 60px;width: 99%;margin-left:4px"></textarea>
 			</div>
 			<div style="text-align: left;padding-top: 5px">
             ${h.submit('commit',_('Commit changes'),class_="ui-btn")}
             ${h.reset('reset',_('Reset'),class_="ui-btn")}
 			</div>
 			${h.end_form()}
 			<script type="text/javascript">
 			var reset_url = "${h.url('files_home',repo_name=c.repo_name,revision=c.cs.raw_id,f_path=c.file.path)}";
 			initCodeMirror('editor',reset_url);
 			</script>
 		</div>
     </div>
 </div>
 </%def>