kallithea Changeset - 5b1f43027662

Changeset - 5b1f43027662

Parent rev.

Child rev.

[Not reviewed]

default

0 15 0

Mads Kiilerich - 6 years ago 2020-03-10 15:08:46
mads@kiilerich.com

scripts: drop isort --wrap-length 160 - it is broken with py3 and not really necessary

Under Python 3, isort 4.3.21 failed with
https://github.com/timothycrosley/isort/issues/741 on
kallithea/controllers/api/api.py :

Traceback (most recent call last):
File "data/env/bin/isort", line 10, in <module>
sys.exit(main())
File ".../env/lib64/python3.7/site-packages/isort/main.py", line 379, in main
for sort_attempt in attempt_iterator:
File ".../env/lib64/python3.7/site-packages/isort/main.py", line 377, in <genexpr>
attempt_iterator = (sort_imports(file_name, **arguments) for file_name in file_names)
File ".../env/lib64/python3.7/site-packages/isort/main.py", line 88, in sort_imports
result = SortImports(file_name, **arguments)
File ".../env/lib64/python3.7/site-packages/isort/isort.py", line 207, in __init__
self._add_formatted_imports()
File ".../env/lib64/python3.7/site-packages/isort/isort.py", line 606, in _add_formatted_imports
self._add_from_imports(from_modules, section, section_output, sort_ignore_case)
File ".../env/lib64/python3.7/site-packages/isort/isort.py", line 526, in _add_from_imports
import_statement = self._multi_line_reformat(import_start, from_import_section, comments)
File ".../env/lib64/python3.7/site-packages/isort/isort.py", line 552, in _multi_line_reformat
dynamic_indent, indent, line_length, comments)
File ".../env/lib64/python3.7/site-packages/isort/isort.py", line 705, in _output_grid
if len(next_statement.split(self.line_separator)[-1]) + 1 > line_length:
TypeError: '>' not supported between instances of 'int' and 'str'

9 files changed:

kallithea/controllers/api/api.py

kallithea/controllers/files.py

kallithea/lib/auth.py

kallithea/lib/vcs/backends/base.py

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

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

kallithea/lib/vcs/backends/hg/changeset.py

kallithea/lib/vcs/backends/hg/repository.py

kallithea/model/db.py

Changeset was too big and was cut off... Show full diff anyway

0 comments (0 inline, 0 general)

kallithea/controllers/api/api.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.controllers.api.api
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 API controller for Kallithea
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Aug 20, 2011
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 import traceback
 from datetime import datetime
 from tg import request
 from kallithea.controllers.api import JSONRPCController, JSONRPCError
 from kallithea.lib.auth import (
     AuthUser, HasPermissionAny, HasPermissionAnyDecorator, HasRepoGroupPermissionLevel, HasRepoPermissionLevel, HasUserGroupPermissionLevel)
 from kallithea.lib.auth import (AuthUser, HasPermissionAny, HasPermissionAnyDecorator, HasRepoGroupPermissionLevel, HasRepoPermissionLevel,
                                 HasUserGroupPermissionLevel)
 from kallithea.lib.exceptions import DefaultUserException, UserGroupsAssignedException
 from kallithea.lib.utils import action_logger, repo2db_mapper
 from kallithea.lib.utils2 import OAttr, Optional
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.vcs.exceptions import EmptyRepositoryError
 from kallithea.model.changeset_status import ChangesetStatusModel
 from kallithea.model.comment import ChangesetCommentsModel
 from kallithea.model.db import ChangesetStatus, Gist, Permission, PullRequest, RepoGroup, Repository, Setting, User, UserGroup, UserIpMap
 from kallithea.model.gist import GistModel
 from kallithea.model.meta import Session
 from kallithea.model.pull_request import PullRequestModel
 from kallithea.model.repo import RepoModel
 from kallithea.model.repo_group import RepoGroupModel
 from kallithea.model.scm import ScmModel, UserGroupList
 from kallithea.model.user import UserModel
 from kallithea.model.user_group import UserGroupModel
 log = logging.getLogger(__name__)
 def store_update(updates, attr, name):
     """
     Stores param in updates dict if it's not instance of Optional
     allows easy updates of passed in params
     """
     if not isinstance(attr, Optional):
         updates[name] = attr
 def get_user_or_error(userid):
     """
     Get user by id or name or return JsonRPCError if not found
     :param userid:
     """
     user = UserModel().get_user(userid)
     if user is None:
         raise JSONRPCError("user `%s` does not exist" % (userid,))
     return user
 def get_repo_or_error(repoid):
     """
     Get repo by id or name or return JsonRPCError if not found
     :param repoid:
     """
     repo = RepoModel().get_repo(repoid)
     if repo is None:
         raise JSONRPCError('repository `%s` does not exist' % (repoid,))
     return repo
 def get_repo_group_or_error(repogroupid):
     """
     Get repo group by id or name or return JsonRPCError if not found
     :param repogroupid:
     """
     repo_group = RepoGroup.guess_instance(repogroupid)
     if repo_group is None:
         raise JSONRPCError(
             'repository group `%s` does not exist' % (repogroupid,))
     return repo_group
 def get_user_group_or_error(usergroupid):
     """
     Get user group by id or name or return JsonRPCError if not found
     :param usergroupid:
     """
     user_group = UserGroupModel().get_group(usergroupid)
     if user_group is None:
         raise JSONRPCError('user group `%s` does not exist' % (usergroupid,))
     return user_group
 def get_perm_or_error(permid, prefix=None):
     """
     Get permission by id or name or return JsonRPCError if not found
     :param permid:
     """
     perm = Permission.get_by_key(permid)
     if perm is None:
         raise JSONRPCError('permission `%s` does not exist' % (permid,))
     if prefix:
         if not perm.permission_name.startswith(prefix):
             raise JSONRPCError('permission `%s` is invalid, '
                                'should start with %s' % (permid, prefix))
     return perm
 def get_gist_or_error(gistid):
     """
     Get gist by id or gist_access_id or return JsonRPCError if not found
     :param gistid:
     """
     gist = GistModel().get_gist(gistid)
     if gist is None:
         raise JSONRPCError('gist `%s` does not exist' % (gistid,))
     return gist
 class ApiController(JSONRPCController):
     """
     API Controller
     The authenticated user can be found as request.authuser.
     Example function::
         def func(arg1, arg2,...):
             pass
     Each function should also **raise** JSONRPCError for any
     errors that happens.
     """
     @HasPermissionAnyDecorator('hg.admin')
     def test(self, args):
         return args
     @HasPermissionAnyDecorator('hg.admin')
     def pull(self, repoid, clone_uri=Optional(None)):
         """
         Triggers a pull from remote location on given repo. Can be used to
         automatically keep remote repos up to date. This command can be executed
         only using api_key belonging to user with admin rights
         :param repoid: repository name or repository id
         :type repoid: str or int
         :param clone_uri: repository URI to pull from (optional)
         :type clone_uri: str
         OUTPUT::
           id : <id_given_in_input>
           result : {
             "msg": "Pulled from `<repository name>`"
             "repository": "<repository name>"
+          }
           error :  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "Unable to pull changes from `<reponame>`"
+          }
         """
         repo = get_repo_or_error(repoid)
         try:
             ScmModel().pull_changes(repo.repo_name,
                                     request.authuser.username,
                                     request.ip_addr,
                                     clone_uri=Optional.extract(clone_uri))
             return dict(
                 msg='Pulled from `%s`' % repo.repo_name,
                 repository=repo.repo_name
+            )
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError(
                 'Unable to pull changes from `%s`' % repo.repo_name
+            )
     @HasPermissionAnyDecorator('hg.admin')
     def rescan_repos(self, remove_obsolete=Optional(False)):
         """
         Triggers rescan repositories action. If remove_obsolete is set
         than also delete repos that are in database but not in the filesystem.
         aka "clean zombies". This command can be executed only using api_key
         belonging to user with admin rights.
         :param remove_obsolete: deletes repositories from
             database that are not found on the filesystem
         :type remove_obsolete: Optional(bool)
         OUTPUT::
           id : <id_given_in_input>
           result : {
             'added': [<added repository name>,...]
             'removed': [<removed repository name>,...]
+          }
           error :  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             'Error occurred during rescan repositories action'
+          }
         """
         try:
             rm_obsolete = Optional.extract(remove_obsolete)
             added, removed = repo2db_mapper(ScmModel().repo_scan(),
                                             remove_obsolete=rm_obsolete)
             return {'added': added, 'removed': removed}
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError(
                 'Error occurred during rescan repositories action'
+            )
     def invalidate_cache(self, repoid):
         """
         Invalidate cache for repository.
         This command can be executed only using api_key belonging to user with admin
         rights or regular user that have write or admin or write access to repository.
         :param repoid: repository name or repository id
         :type repoid: str or int
         OUTPUT::
           id : <id_given_in_input>
           result : {
             'msg': Cache for repository `<repository name>` was invalidated,
             'repository': <repository name>
+          }
           error :  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             'Error occurred during cache invalidation action'
+          }
         """
         repo = get_repo_or_error(repoid)
         if not HasPermissionAny('hg.admin')():
             if not HasRepoPermissionLevel('write')(repo.repo_name):
                 raise JSONRPCError('repository `%s` does not exist' % (repoid,))
         try:
             ScmModel().mark_for_invalidation(repo.repo_name)
             return dict(
                 msg='Cache for repository `%s` was invalidated' % (repoid,),
                 repository=repo.repo_name
+            )
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError(
                 'Error occurred during cache invalidation action'
+            )
     @HasPermissionAnyDecorator('hg.admin')
     def get_ip(self, userid=Optional(OAttr('apiuser'))):
         """
         Shows IP address as seen from Kallithea server, together with all
         defined IP addresses for given user. If userid is not passed data is
         returned for user who's calling this function.
         This command can be executed only using api_key belonging to user with
         admin rights.
         :param userid: username to show ips for
         :type userid: Optional(str or int)
         OUTPUT::
             id : <id_given_in_input>
             result : {
                          "server_ip_addr": "<ip_from_clien>",
                          "user_ips": [
+                                        {
                                            "ip_addr": "<ip_with_mask>",
                                            "ip_range": ["<start_ip>", "<end_ip>"],
                                         },
                                         ...
+                                     ]
+            }
         """
         if isinstance(userid, Optional):
             userid = request.authuser.user_id
         user = get_user_or_error(userid)
         ips = UserIpMap.query().filter(UserIpMap.user == user).all()
         return dict(
             server_ip_addr=request.ip_addr,
             user_ips=ips
+        )
     # alias for old
     show_ip = get_ip
     @HasPermissionAnyDecorator('hg.admin')
     def get_server_info(self):
         """
         return server info, including Kallithea version and installed packages
         OUTPUT::
           id : <id_given_in_input>
           result : {
             'modules': [<module name>,...]
             'py_version': <python version>,
             'platform': <platform type>,
             'kallithea_version': <kallithea version>
+          }
           error :  null
         """
         return Setting.get_server_info()
     def get_user(self, userid=Optional(OAttr('apiuser'))):
         """
         Gets a user by username or user_id, Returns empty result if user is
         not found. If userid param is skipped it is set to id of user who is
         calling this method. This command can be executed only using api_key
         belonging to user with admin rights, or regular users that cannot
         specify different userid than theirs
         :param userid: user to get data for
         :type userid: Optional(str or int)
         OUTPUT::
             id : <id_given_in_input>
             result: None if user does not exist or
+                    {
                         "user_id" :     "<user_id>",
                         "api_key" :     "<api_key>",
                         "api_keys":     "[<list of all API keys including additional ones>]"
                         "username" :    "<username>",
                         "firstname":    "<firstname>",
                         "lastname" :    "<lastname>",
                         "email" :       "<email>",
                         "emails":       "[<list of all emails including additional ones>]",
                         "ip_addresses": "[<ip_address_for_user>,...]",
                         "active" :      "<bool: user active>",
                         "admin" :       "<bool: user is admin>",
                         "extern_name" : "<extern_name>",
                         "extern_type" : "<extern type>
                         "last_login":   "<last_login>",
                         "permissions": {
                             "global": ["hg.create.repository",
                                        "repository.read",
                                        "hg.register.manual_activate"],
                             "repositories": {"repo1": "repository.none"},
                             "repositories_groups": {"Group1": "group.read"}
                          },
+                    }
             error:  null
         """
         if not HasPermissionAny('hg.admin')():
             # make sure normal user does not pass someone else userid,
             # he is not allowed to do that
             if not isinstance(userid, Optional) and userid != request.authuser.user_id:
                 raise JSONRPCError(
                     'userid is not the same as your user'
+                )
         if isinstance(userid, Optional):
             userid = request.authuser.user_id
         user = get_user_or_error(userid)
         data = user.get_api_data()
         data['permissions'] = AuthUser(user_id=user.user_id).permissions
         return data
     @HasPermissionAnyDecorator('hg.admin')
     def get_users(self):
         """
         Lists all existing users. This command can be executed only using api_key
         belonging to user with admin rights.
         OUTPUT::
             id : <id_given_in_input>
             result: [<user_object>, ...]
             error:  null
         """
         return [
             user.get_api_data()
             for user in User.query()
                 .order_by(User.username)
                 .filter_by(is_default_user=False)
+        ]
     @HasPermissionAnyDecorator('hg.admin')
     def create_user(self, username, email, password=Optional(''),
                     firstname=Optional(''), lastname=Optional(''),
                     active=Optional(True), admin=Optional(False),
                     extern_type=Optional(User.DEFAULT_AUTH_TYPE),
                     extern_name=Optional('')):
         """
         Creates new user. Returns new user object. This command can
         be executed only using api_key belonging to user with admin rights.
         :param username: new username
         :type username: str or int
         :param email: email
         :type email: str
         :param password: password
         :type password: Optional(str)
         :param firstname: firstname
         :type firstname: Optional(str)
         :param lastname: lastname
         :type lastname: Optional(str)
         :param active: active
         :type active: Optional(bool)
         :param admin: admin
         :type admin: Optional(bool)
         :param extern_name: name of extern
         :type extern_name: Optional(str)
         :param extern_type: extern_type
         :type extern_type: Optional(str)
         OUTPUT::
             id : <id_given_in_input>
             result: {
                       "msg" : "created new user `<username>`",
                       "user": <user_obj>
+                    }
             error:  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "user `<username>` already exist"
             or
             "email `<email>` already exist"
             or
             "failed to create user `<username>`"
+          }
         """
         if User.get_by_username(username):
             raise JSONRPCError("user `%s` already exist" % (username,))
         if User.get_by_email(email):
             raise JSONRPCError("email `%s` already exist" % (email,))
         try:
             user = UserModel().create_or_update(
                 username=Optional.extract(username),
                 password=Optional.extract(password),
                 email=Optional.extract(email),
                 firstname=Optional.extract(firstname),
                 lastname=Optional.extract(lastname),
                 active=Optional.extract(active),
                 admin=Optional.extract(admin),
                 extern_type=Optional.extract(extern_type),
                 extern_name=Optional.extract(extern_name)
+            )
             Session().commit()
             return dict(
                 msg='created new user `%s`' % username,
                 user=user.get_api_data()
+            )
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError('failed to create user `%s`' % (username,))
     @HasPermissionAnyDecorator('hg.admin')
     def update_user(self, userid, username=Optional(None),
                     email=Optional(None), password=Optional(None),
                     firstname=Optional(None), lastname=Optional(None),
                     active=Optional(None), admin=Optional(None),
                     extern_type=Optional(None), extern_name=Optional(None)):
         """
         updates given user if such user exists. This command can
         be executed only using api_key belonging to user with admin rights.
         :param userid: userid to update
         :type userid: str or int
         :param username: new username
         :type username: str or int
         :param email: email
         :type email: str
         :param password: password
         :type password: Optional(str)
         :param firstname: firstname
         :type firstname: Optional(str)
         :param lastname: lastname
         :type lastname: Optional(str)
         :param active: active
         :type active: Optional(bool)
         :param admin: admin
         :type admin: Optional(bool)
         :param extern_name:
         :type extern_name: Optional(str)
         :param extern_type:
         :type extern_type: Optional(str)
         OUTPUT::
             id : <id_given_in_input>
             result: {
                       "msg" : "updated user ID:<userid> <username>",
                       "user": <user_object>,
+                    }
             error:  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "failed to update user `<username>`"
+          }
         """
         user = get_user_or_error(userid)
         # only non optional arguments will be stored in updates
         updates = {}
         try:
             store_update(updates, username, 'username')
             store_update(updates, password, 'password')
             store_update(updates, email, 'email')
             store_update(updates, firstname, 'name')
             store_update(updates, lastname, 'lastname')
             store_update(updates, active, 'active')
             store_update(updates, admin, 'admin')
             store_update(updates, extern_name, 'extern_name')
             store_update(updates, extern_type, 'extern_type')
             user = UserModel().update_user(user, **updates)
             Session().commit()
             return dict(
                 msg='updated user ID:%s %s' % (user.user_id, user.username),
                 user=user.get_api_data()
+            )
         except DefaultUserException:
             log.error(traceback.format_exc())
             raise JSONRPCError('editing default user is forbidden')
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError('failed to update user `%s`' % (userid,))
     @HasPermissionAnyDecorator('hg.admin')
     def delete_user(self, userid):
         """
         deletes given user if such user exists. This command can
         be executed only using api_key belonging to user with admin rights.
         :param userid: user to delete
         :type userid: str or int
         OUTPUT::
             id : <id_given_in_input>
             result: {
                       "msg" : "deleted user ID:<userid> <username>",
                       "user": null
+                    }
             error:  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "failed to delete user ID:<userid> <username>"
+          }
         """
         user = get_user_or_error(userid)
         try:
             UserModel().delete(userid)
             Session().commit()
             return dict(
                 msg='deleted user ID:%s %s' % (user.user_id, user.username),
                 user=None
+            )
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError('failed to delete user ID:%s %s'
                                % (user.user_id, user.username))
     # permission check inside
     def get_user_group(self, usergroupid):
         """
         Gets an existing user group. This command can be executed only using api_key
         belonging to user with admin rights or user who has at least
         read access to user group.
         :param usergroupid: id of user_group to edit
         :type usergroupid: str or int
         OUTPUT::
             id : <id_given_in_input>
             result : None if group not exist
+                     {
                        "users_group_id" : "<id>",
                        "group_name" :     "<groupname>",
                        "active":          "<bool>",
                        "members" :  [<user_obj>,...]
+                     }
             error : null
         """
         user_group = get_user_group_or_error(usergroupid)
         if not HasPermissionAny('hg.admin')():
             if not HasUserGroupPermissionLevel('read')(user_group.users_group_name):
                 raise JSONRPCError('user group `%s` does not exist' % (usergroupid,))
         data = user_group.get_api_data()
         return data
     # permission check inside
     def get_user_groups(self):
         """
         Lists all existing user groups. This command can be executed only using
         api_key belonging to user with admin rights or user who has at least
         read access to user group.
         OUTPUT::
             id : <id_given_in_input>
             result : [<user_group_obj>,...]
             error : null
         """
         return [
             user_group.get_api_data()
             for user_group in UserGroupList(UserGroup.query().all(), perm_level='read')
+        ]
     @HasPermissionAnyDecorator('hg.admin', 'hg.usergroup.create.true')
     def create_user_group(self, group_name, description=Optional(''),
                           owner=Optional(OAttr('apiuser')), active=Optional(True)):
         """
         Creates new user group. This command can be executed only using api_key
         belonging to user with admin rights or an user who has create user group
         permission
         :param group_name: name of new user group
         :type group_name: str
         :param description: group description
         :type description: str
         :param owner: owner of group. If not passed apiuser is the owner
         :type owner: Optional(str or int)
         :param active: group is active
         :type active: Optional(bool)
         OUTPUT::
             id : <id_given_in_input>
             result: {
                       "msg": "created new user group `<groupname>`",
                       "user_group": <user_group_object>
+                    }
             error:  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "user group `<group name>` already exist"
             or
             "failed to create group `<group name>`"
+          }
         """
         if UserGroupModel().get_by_name(group_name):
             raise JSONRPCError("user group `%s` already exist" % (group_name,))
         try:
             if isinstance(owner, Optional):
                 owner = request.authuser.user_id
             owner = get_user_or_error(owner)
             active = Optional.extract(active)
             description = Optional.extract(description)
             ug = UserGroupModel().create(name=group_name, description=description,
                                          owner=owner, active=active)
             Session().commit()
             return dict(
                 msg='created new user group `%s`' % group_name,
                 user_group=ug.get_api_data()
+            )
         except Exception:
             log.error(traceback.format_exc())
             raise JSONRPCError('failed to create group `%s`' % (group_name,))
     # permission check inside
     def update_user_group(self, usergroupid, group_name=Optional(''),
                           description=Optional(''), owner=Optional(None),
                           active=Optional(True)):
         """
         Updates given usergroup.  This command can be executed only using api_key
         belonging to user with admin rights or an admin of given user group
         :param usergroupid: id of user group to update
         :type usergroupid: str or int
         :param group_name: name of new user group
         :type group_name: str
         :param description: group description
         :type description: str
         :param owner: owner of group.
         :type owner: Optional(str or int)
         :param active: group is active
         :type active: Optional(bool)
         OUTPUT::
           id : <id_given_in_input>
           result : {
             "msg": 'updated user group ID:<user group id> <user group name>',
             "user_group": <user_group_object>
+          }
           error :  null
         ERROR OUTPUT::
           id : <id_given_in_input>
           result : null
           error :  {
             "failed to update user group `<user group name>`"
+          }
         """
         user_group = get_user_group_or_error(usergroupid)
         if not HasPermissionAny('hg.admin')():
             if not HasUserGroupPermissionLevel('admin')(user_group.users_group_name):
                 raise JSONRPCError('user group `%s` does not exist' % (usergroupid,))
         if not isinstance(owner, Optional):
             owner = get_user_or_error(owner)
         updates = {}
         store_update(updates, group_name, 'users_group_name')
         store_update(updates, description, 'user_group_description')
         store_update(updates, owner, 'owner')
         store_update(updates, active, 'users_group_active')
         try:
             UserGroupModel().update(user_group, updates)
             Session().commit()
             return dict(
                 msg='updated user group ID:%s %s' % (user_group.users_group_id,
                                                      user_group.users_group_name),
                 user_group=user_group.get_api_data()
+            )

kallithea/controllers/files.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.controllers.files
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Files controller for Kallithea
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Apr 21, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 import os
 import posixpath
 import shutil
 import tempfile
 import traceback
 from collections import OrderedDict
 from tg import request, response
 from tg import tmpl_context as c
 from tg.i18n import ugettext as _
 from webob.exc import HTTPFound, HTTPNotFound
 from kallithea.config.routing import url
 from kallithea.controllers.changeset import _context_url, _ignorews_url, anchor_url, get_ignore_ws, get_line_ctx
 from kallithea.lib import diffs
 from kallithea.lib import helpers as h
 from kallithea.lib.auth import HasRepoPermissionLevelDecorator, LoginRequired
 from kallithea.lib.base import BaseRepoController, jsonify, render
 from kallithea.lib.exceptions import NonRelativePathError
 from kallithea.lib.utils import action_logger
 from kallithea.lib.utils2 import convert_line_endings, detect_mode, safe_int, safe_str, str2bool
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import (
     ChangesetDoesNotExistError, ChangesetError, EmptyRepositoryError, ImproperArchiveTypeError, NodeAlreadyExistsError, NodeDoesNotExistError, NodeError, RepositoryError, VCSError)
 from kallithea.lib.vcs.exceptions import (ChangesetDoesNotExistError, ChangesetError, EmptyRepositoryError, ImproperArchiveTypeError, NodeAlreadyExistsError,
                                           NodeDoesNotExistError, NodeError, RepositoryError, VCSError)
 from kallithea.lib.vcs.nodes import FileNode
 from kallithea.model.db import Repository
 from kallithea.model.repo import RepoModel
 from kallithea.model.scm import ScmModel
 log = logging.getLogger(__name__)
 class FilesController(BaseRepoController):
     def _before(self, *args, **kwargs):
         super(FilesController, self)._before(*args, **kwargs)
     def __get_cs(self, rev, silent_empty=False):
         """
         Safe way to get changeset if error occur it redirects to tip with
         proper message
         :param rev: revision to fetch
         :silent_empty: return None if repository is empty
         """
         try:
             return c.db_repo_scm_instance.get_changeset(rev)
         except EmptyRepositoryError as e:
             if silent_empty:
                 return None
             url_ = url('files_add_home',
                        repo_name=c.repo_name,
                        revision=0, f_path='', anchor='edit')
             add_new = h.link_to(_('Click here to add new file'), url_, class_="alert-link")
             h.flash(_('There are no files yet.') + ' ' + add_new, category='warning')
             raise HTTPNotFound()
         except (ChangesetDoesNotExistError, LookupError):
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         except RepositoryError as e:
             h.flash(e, category='error')
             raise HTTPNotFound()
     def __get_filenode(self, cs, path):
         """
         Returns file_node or raise HTTP error.
         :param cs: given changeset
         :param path: path to lookup
         """
         try:
             file_node = cs.get_node(path)
             if file_node.is_dir():
                 raise RepositoryError('given path is a directory')
         except ChangesetDoesNotExistError:
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         except RepositoryError as e:
             h.flash(e, category='error')
             raise HTTPNotFound()
         return file_node
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def index(self, repo_name, revision, f_path, annotate=False):
         # redirect to given revision from form if given
         post_revision = request.POST.get('at_rev', None)
         if post_revision:
             cs = self.__get_cs(post_revision) # FIXME - unused!
         c.revision = revision
         c.changeset = self.__get_cs(revision)
         c.branch = request.GET.get('branch', None)
         c.f_path = f_path
         c.annotate = annotate
         cur_rev = c.changeset.revision
         # used in files_source.html:
         c.cut_off_limit = self.cut_off_limit
         c.fulldiff = request.GET.get('fulldiff')
         # prev link
         try:
             prev_rev = c.db_repo_scm_instance.get_changeset(cur_rev).prev(c.branch)
             c.url_prev = url('files_home', repo_name=c.repo_name,
                          revision=prev_rev.raw_id, f_path=f_path)
             if c.branch:
                 c.url_prev += '?branch=%s' % c.branch
         except (ChangesetDoesNotExistError, VCSError):
             c.url_prev = '#'
         # next link
         try:
             next_rev = c.db_repo_scm_instance.get_changeset(cur_rev).next(c.branch)
             c.url_next = url('files_home', repo_name=c.repo_name,
                      revision=next_rev.raw_id, f_path=f_path)
             if c.branch:
                 c.url_next += '?branch=%s' % c.branch
         except (ChangesetDoesNotExistError, VCSError):
             c.url_next = '#'
         # files or dirs
         try:
             c.file = c.changeset.get_node(f_path)
             if c.file.is_submodule():
                 raise HTTPFound(location=c.file.url)
             elif c.file.is_file():
                 c.load_full_history = False
                 # determine if we're on branch head
                 _branches = c.db_repo_scm_instance.branches
                 c.on_branch_head = revision in _branches or revision in _branches.values()
                 _hist = []
                 c.file_history = []
                 if c.load_full_history:
                     c.file_history, _hist = self._get_node_history(c.changeset, f_path)
                 c.authors = []
                 for a in set([x.author for x in _hist]):
                     c.authors.append((h.email(a), h.person(a)))
             else:
                 c.authors = c.file_history = []
         except RepositoryError as e:
             h.flash(e, category='error')
             raise HTTPNotFound()
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             return render('files/files_ypjax.html')
         # TODO: tags and bookmarks?
         c.revision_options = [(c.changeset.raw_id,
                               _('%s at %s') % (b, h.short_id(c.changeset.raw_id))) for b in c.changeset.branches] + \
             [(n, b) for b, n in c.db_repo_scm_instance.branches.items()]
         if c.db_repo_scm_instance.closed_branches:
             prefix = _('(closed)') + ' '
             c.revision_options += [('-', '-')] + \
                 [(n, prefix + b) for b, n in c.db_repo_scm_instance.closed_branches.items()]
         return render('files/files.html')
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def history(self, repo_name, revision, f_path):
         changeset = self.__get_cs(revision)
         _file = changeset.get_node(f_path)
         if _file.is_file():
             file_history, _hist = self._get_node_history(changeset, f_path)
             res = []
             for obj in file_history:
                 res.append({
                     'text': obj[1],
                     'children': [{'id': o[0], 'text': o[1]} for o in obj[0]]
                 })
             data = {
                 'more': False,
                 'results': res
+            }
             return data
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def authors(self, repo_name, revision, f_path):
         changeset = self.__get_cs(revision)
         _file = changeset.get_node(f_path)
         if _file.is_file():
             file_history, _hist = self._get_node_history(changeset, f_path)
             c.authors = []
             for a in set([x.author for x in _hist]):
                 c.authors.append((h.email(a), h.person(a)))
             return render('files/files_history_box.html')
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def rawfile(self, repo_name, revision, f_path):
         cs = self.__get_cs(revision)
         file_node = self.__get_filenode(cs, f_path)
         response.content_disposition = \
             'attachment; filename=%s' % f_path.split(Repository.url_sep())[-1]
         response.content_type = file_node.mimetype
         return file_node.content
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def raw(self, repo_name, revision, f_path):
         cs = self.__get_cs(revision)
         file_node = self.__get_filenode(cs, f_path)
         raw_mimetype_mapping = {
             # map original mimetype to a mimetype used for "show as raw"
             # you can also provide a content-disposition to override the
             # default "attachment" disposition.
             # orig_type: (new_type, new_dispo)
             # show images inline:
             'image/x-icon': ('image/x-icon', 'inline'),
             'image/png': ('image/png', 'inline'),
             'image/gif': ('image/gif', 'inline'),
             'image/jpeg': ('image/jpeg', 'inline'),
             'image/svg+xml': ('image/svg+xml', 'inline'),
+        }
         mimetype = file_node.mimetype
         try:
             mimetype, dispo = raw_mimetype_mapping[mimetype]
         except KeyError:
             # we don't know anything special about this, handle it safely
             if file_node.is_binary:
                 # do same as download raw for binary files
                 mimetype, dispo = 'application/octet-stream', 'attachment'
             else:
                 # do not just use the original mimetype, but force text/plain,
                 # otherwise it would serve text/html and that might be unsafe.
                 # Note: underlying vcs library fakes text/plain mimetype if the
                 # mimetype can not be determined and it thinks it is not
                 # binary.This might lead to erroneous text display in some
                 # cases, but helps in other cases, like with text files
                 # without extension.
                 mimetype, dispo = 'text/plain', 'inline'
         if dispo == 'attachment':
             dispo = 'attachment; filename=%s' % f_path.split(os.sep)[-1]
         response.content_disposition = dispo
         response.content_type = mimetype
         return file_node.content
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def delete(self, repo_name, revision, f_path):
         repo = c.db_repo
         # check if revision is a branch identifier- basically we cannot
         # create multiple heads via file editing
         _branches = repo.scm_instance.branches
         # check if revision is a branch name or branch hash
         if revision not in _branches and revision not in _branches.values():
             h.flash(_('You can only delete files with revision '
                       'being a valid branch'), category='warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip',
                                   f_path=f_path))
         r_post = request.POST
         c.cs = self.__get_cs(revision)
         c.file = self.__get_filenode(c.cs, f_path)
         c.default_message = _('Deleted file %s via Kallithea') % (f_path)
         c.f_path = f_path
         node_path = f_path
         author = request.authuser.full_contact
         if r_post:
             message = r_post.get('message') or c.default_message
             try:
                 nodes = {
                     node_path: {
                         'content': ''
+                    }
+                }
                 self.scm_model.delete_nodes(
                     user=request.authuser.user_id,
                     ip_addr=request.ip_addr,
                     repo=c.db_repo,
                     message=message,
                     nodes=nodes,
                     parent_cs=c.cs,
                     author=author,
+                )
                 h.flash(_('Successfully deleted file %s') % f_path,
                         category='success')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
         return render('files/files_delete.html')
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def edit(self, repo_name, revision, f_path):
         repo = c.db_repo
         # check if revision is a branch identifier- basically we cannot
         # create multiple heads via file editing
         _branches = repo.scm_instance.branches
         # check if revision is a branch name or branch hash
         if revision not in _branches and revision not in _branches.values():
             h.flash(_('You can only edit files with revision '
                       'being a valid branch'), category='warning')
             raise HTTPFound(location=h.url('files_home',
                                   repo_name=repo_name, revision='tip',
                                   f_path=f_path))
         r_post = request.POST
         c.cs = self.__get_cs(revision)
         c.file = self.__get_filenode(c.cs, f_path)
         if c.file.is_binary:
             raise HTTPFound(location=url('files_home', repo_name=c.repo_name,
                             revision=c.cs.raw_id, f_path=f_path))
         c.default_message = _('Edited file %s via Kallithea') % (f_path)
         c.f_path = f_path
         if r_post:
             old_content = safe_str(c.file.content)
             sl = old_content.splitlines(1)
             first_line = sl[0] if sl else ''
             # modes:  0 - Unix, 1 - Mac, 2 - DOS
             mode = detect_mode(first_line, 0)
             content = convert_line_endings(r_post.get('content', ''), mode)
             message = r_post.get('message') or c.default_message
             author = request.authuser.full_contact
             if content == old_content:
                 h.flash(_('No changes'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             try:
                 self.scm_model.commit_change(repo=c.db_repo_scm_instance,
                                              repo_name=repo_name, cs=c.cs,
                                              user=request.authuser.user_id,
                                              ip_addr=request.ip_addr,
                                              author=author, message=message,
                                              content=content, f_path=f_path)
                 h.flash(_('Successfully committed to %s') % f_path,
                         category='success')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
         return render('files/files_edit.html')
     @LoginRequired()
     @HasRepoPermissionLevelDecorator('write')
     def add(self, repo_name, revision, f_path):
         repo = c.db_repo
         r_post = request.POST
         c.cs = self.__get_cs(revision, silent_empty=True)
         if c.cs is None:
             c.cs = EmptyChangeset(alias=c.db_repo_scm_instance.alias)
         c.default_message = (_('Added file via Kallithea'))
         c.f_path = f_path
         if r_post:
             unix_mode = 0
             content = convert_line_endings(r_post.get('content', ''), unix_mode)
             message = r_post.get('message') or c.default_message
             filename = r_post.get('filename')
             location = r_post.get('location', '')
             file_obj = r_post.get('upload_file', None)
             if file_obj is not None and hasattr(file_obj, 'filename'):
                 filename = file_obj.filename
                 content = file_obj.file
                 if hasattr(content, 'file'):
                     # non posix systems store real file under file attr
                     content = content.file
             if not content:
                 h.flash(_('No content'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             if not filename:
                 h.flash(_('No filename'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             # strip all crap out of file, just leave the basename
             filename = os.path.basename(filename)
             node_path = posixpath.join(location, filename)
             author = request.authuser.full_contact
             try:
                 nodes = {
                     node_path: {
                         'content': content
+                    }
+                }
                 self.scm_model.create_nodes(
                     user=request.authuser.user_id,
                     ip_addr=request.ip_addr,
                     repo=c.db_repo,
                     message=message,
                     nodes=nodes,
                     parent_cs=c.cs,
                     author=author,
+                )
                 h.flash(_('Successfully committed to %s') % node_path,
                         category='success')
             except NonRelativePathError as e:
                 h.flash(_('Location must be relative path and must not '
                           'contain .. in path'), category='warning')
                 raise HTTPFound(location=url('changeset_home', repo_name=c.repo_name,
                                     revision='tip'))
             except (NodeError, NodeAlreadyExistsError) as e:
                 h.flash(_(e), category='error')
             except Exception:
                 log.error(traceback.format_exc())
                 h.flash(_('Error occurred during commit'), category='error')
             raise HTTPFound(location=url('changeset_home',
                                 repo_name=c.repo_name, revision='tip'))
         return render('files/files_add.html')
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def archivefile(self, repo_name, fname):
         fileformat = None
         revision = None
         ext = None
         subrepos = request.GET.get('subrepos') == 'true'
         for a_type, ext_data in settings.ARCHIVE_SPECS.items():
             archive_spec = fname.split(ext_data[1])
             if len(archive_spec) == 2 and archive_spec[1] == '':
                 fileformat = a_type or ext_data[1]
                 revision = archive_spec[0]
                 ext = ext_data[1]
         try:
             dbrepo = RepoModel().get_by_repo_name(repo_name)
             if not dbrepo.enable_downloads:
                 return _('Downloads disabled') # TODO: do something else?
             if c.db_repo_scm_instance.alias == 'hg':
                 # patch and reset hooks section of UI config to not run any
                 # hooks on fetching archives with subrepos
                 for k, v in c.db_repo_scm_instance._repo.ui.configitems('hooks'):
                     c.db_repo_scm_instance._repo.ui.setconfig('hooks', k, None)
             cs = c.db_repo_scm_instance.get_changeset(revision)
             content_type = settings.ARCHIVE_SPECS[fileformat][0]
         except ChangesetDoesNotExistError:
             return _('Unknown revision %s') % revision
         except EmptyRepositoryError:
             return _('Empty repository')
         except (ImproperArchiveTypeError, KeyError):
             return _('Unknown archive type')
         from kallithea import CONFIG
         rev_name = cs.raw_id[:12]
         archive_name = '%s-%s%s' % (repo_name.replace('/', '_'), rev_name, ext)
         archive_path = None
         cached_archive_path = None
         archive_cache_dir = CONFIG.get('archive_cache_dir')
         if archive_cache_dir and not subrepos: # TODO: subrepo caching?
             if not os.path.isdir(archive_cache_dir):
                 os.makedirs(archive_cache_dir)
             cached_archive_path = os.path.join(archive_cache_dir, archive_name)
             if os.path.isfile(cached_archive_path):
                 log.debug('Found cached archive in %s', cached_archive_path)
                 archive_path = cached_archive_path
             else:
                 log.debug('Archive %s is not yet cached', archive_name)
         if archive_path is None:
             # generate new archive
             fd, archive_path = tempfile.mkstemp()
             log.debug('Creating new temp archive in %s', archive_path)
             with os.fdopen(fd, 'wb') as stream:
                 cs.fill_archive(stream=stream, kind=fileformat, subrepos=subrepos)
                 # stream (and thus fd) has been closed by cs.fill_archive
             if cached_archive_path is not None:
                 # we generated the archive - move it to cache
                 log.debug('Storing new archive in %s', cached_archive_path)
                 shutil.move(archive_path, cached_archive_path)
                 archive_path = cached_archive_path
         def get_chunked_archive(archive_path):
             stream = open(archive_path, 'rb')
             while True:
                 data = stream.read(16 * 1024)
                 if not data:
                     break
                 yield data
             stream.close()
             if archive_path != cached_archive_path:
                 log.debug('Destroying temp archive %s', archive_path)
                 os.remove(archive_path)
         action_logger(user=request.authuser,
                       action='user_downloaded_archive:%s' % (archive_name),
                       repo=repo_name, ipaddr=request.ip_addr, commit=True)
         response.content_disposition = str('attachment; filename=%s' % (archive_name))
         response.content_type = str(content_type)
         return get_chunked_archive(archive_path)
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def diff(self, repo_name, f_path):
         ignore_whitespace = request.GET.get('ignorews') == '1'
         line_context = safe_int(request.GET.get('context'), 3)
         diff2 = request.GET.get('diff2', '')
         diff1 = request.GET.get('diff1', '') or diff2
         c.action = request.GET.get('diff')
         c.no_changes = diff1 == diff2
         c.f_path = f_path
         c.big_diff = False
         fulldiff = request.GET.get('fulldiff')
         c.anchor_url = anchor_url
         c.ignorews_url = _ignorews_url
         c.context_url = _context_url
         c.changes = OrderedDict()
         c.changes[diff2] = []
         # special case if we want a show rev only, it's impl here
         # to reduce JS and callbacks
         if request.GET.get('show_rev'):
             if str2bool(request.GET.get('annotate', 'False')):
                 _url = url('files_annotate_home', repo_name=c.repo_name,
                            revision=diff1, f_path=c.f_path)
             else:
                 _url = url('files_home', repo_name=c.repo_name,
                            revision=diff1, f_path=c.f_path)
             raise HTTPFound(location=_url)
         try:
             if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_1 = c.db_repo_scm_instance.get_changeset(diff1)
                 try:
                     node1 = c.changeset_1.get_node(f_path)
                     if node1.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node1, type(node1)))
                 except NodeDoesNotExistError:
                     c.changeset_1 = EmptyChangeset(cs=diff1,
                                                    revision=c.changeset_1.revision,
                                                    repo=c.db_repo_scm_instance)
                     node1 = FileNode(f_path, '', changeset=c.changeset_1)
             else:
                 c.changeset_1 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node1 = FileNode(f_path, '', changeset=c.changeset_1)
             if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_2 = c.db_repo_scm_instance.get_changeset(diff2)
                 try:
                     node2 = c.changeset_2.get_node(f_path)
                     if node2.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node2, type(node2)))
                 except NodeDoesNotExistError:
                     c.changeset_2 = EmptyChangeset(cs=diff2,
                                                    revision=c.changeset_2.revision,
                                                    repo=c.db_repo_scm_instance)
                     node2 = FileNode(f_path, '', changeset=c.changeset_2)
             else:
                 c.changeset_2 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node2 = FileNode(f_path, '', changeset=c.changeset_2)
         except (RepositoryError, NodeError):
             log.error(traceback.format_exc())
             raise HTTPFound(location=url('files_home', repo_name=c.repo_name,
                                 f_path=f_path))
         if c.action == 'download':
             raw_diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
             diff_name = '%s_vs_%s.diff' % (diff1, diff2)
             response.content_type = 'text/plain'
             response.content_disposition = (
                 'attachment; filename=%s' % diff_name
+            )
             return raw_diff
         elif c.action == 'raw':
             raw_diff = diffs.get_gitdiff(node1, node2,
                                       ignore_whitespace=ignore_whitespace,
                                       context=line_context)
             response.content_type = 'text/plain'
             return raw_diff
         else:
             fid = h.FID(diff2, node2.path)
             line_context_lcl = get_line_ctx(fid, request.GET)
             ign_whitespace_lcl = get_ignore_ws(fid, request.GET)
             diff_limit = None if fulldiff else self.cut_off_limit
             c.a_rev, c.cs_rev, a_path, diff, st, op = diffs.wrapped_diff(filenode_old=node1,
                                          filenode_new=node2,
                                          diff_limit=diff_limit,
                                          ignore_whitespace=ign_whitespace_lcl,
                                          line_context=line_context_lcl,
                                          enable_comments=False)
             c.file_diff_data = [(fid, fid, op, a_path, node2.path, diff, st)]
             return render('files/file_diff.html')
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     def diff_2way(self, repo_name, f_path):
         diff1 = request.GET.get('diff1', '')
         diff2 = request.GET.get('diff2', '')
         try:
             if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_1 = c.db_repo_scm_instance.get_changeset(diff1)
                 try:
                     node1 = c.changeset_1.get_node(f_path)
                     if node1.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node1, type(node1)))
                 except NodeDoesNotExistError:
                     c.changeset_1 = EmptyChangeset(cs=diff1,
                                                    revision=c.changeset_1.revision,
                                                    repo=c.db_repo_scm_instance)
                     node1 = FileNode(f_path, '', changeset=c.changeset_1)
             else:
                 c.changeset_1 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node1 = FileNode(f_path, '', changeset=c.changeset_1)
             if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:
                 c.changeset_2 = c.db_repo_scm_instance.get_changeset(diff2)
                 try:
                     node2 = c.changeset_2.get_node(f_path)
                     if node2.is_dir():
                         raise NodeError('%s path is a %s not a file'
                                         % (node2, type(node2)))
                 except NodeDoesNotExistError:
                     c.changeset_2 = EmptyChangeset(cs=diff2,
                                                    revision=c.changeset_2.revision,
                                                    repo=c.db_repo_scm_instance)
                     node2 = FileNode(f_path, '', changeset=c.changeset_2)
             else:
                 c.changeset_2 = EmptyChangeset(repo=c.db_repo_scm_instance)
                 node2 = FileNode(f_path, '', changeset=c.changeset_2)
         except ChangesetDoesNotExistError as e:
             msg = _('Such revision does not exist for this repository')
             h.flash(msg, category='error')
             raise HTTPNotFound()
         c.node1 = node1
         c.node2 = node2
         c.cs1 = c.changeset_1
         c.cs2 = c.changeset_2
         return render('files/diff_2way.html')
     def _get_node_history(self, cs, f_path, changesets=None):
         """
         get changesets history for given node
         :param cs: changeset to calculate history
         :param f_path: path for node to calculate history for
         :param changesets: if passed don't calculate history and take
             changesets defined in this list
         """
         # calculate history based on tip
         tip_cs = c.db_repo_scm_instance.get_changeset()
         if changesets is None:
             try:
                 changesets = tip_cs.get_file_history(f_path)
             except (NodeDoesNotExistError, ChangesetError):
                 # this node is not present at tip !
                 changesets = cs.get_file_history(f_path)
         hist_l = []
         changesets_group = ([], _("Changesets"))
         branches_group = ([], _("Branches"))
         tags_group = ([], _("Tags"))
         for chs in changesets:
             # TODO: loop over chs.branches ... but that will not give all the bogus None branches for Git ...
             _branch = chs.branch
             n_desc = '%s (%s)' % (h.show_id(chs), _branch)
             changesets_group[0].append((chs.raw_id, n_desc,))
         hist_l.append(changesets_group)
         for name, chs in c.db_repo_scm_instance.branches.items():
             branches_group[0].append((chs, name),)
         hist_l.append(branches_group)
         for name, chs in c.db_repo_scm_instance.tags.items():
             tags_group[0].append((chs, name),)
         hist_l.append(tags_group)
         return hist_l, changesets
     @LoginRequired(allow_default_user=True)
     @HasRepoPermissionLevelDecorator('read')
     @jsonify
     def nodelist(self, repo_name, revision, f_path):
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             cs = self.__get_cs(revision)
             _d, _f = ScmModel().get_nodes(repo_name, cs.raw_id, f_path,
                                           flat=False)
             return {'nodes': _d + _f}

kallithea/lib/auth.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.lib.auth
 ~~~~~~~~~~~~~~~~~~
 authentication and permission libraries
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Apr 4, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import hashlib
 import itertools
 import logging
 import os
 import string
 import bcrypt
 import ipaddr
 from decorator import decorator
 from sqlalchemy.orm import joinedload
 from sqlalchemy.orm.exc import ObjectDeletedError
 from tg import request
 from tg.i18n import ugettext as _
 from webob.exc import HTTPForbidden, HTTPFound
 from kallithea.config.routing import url
 from kallithea.lib.caching_query import FromCache
 from kallithea.lib.utils import conditional_cache, get_repo_group_slug, get_repo_slug, get_user_group_slug
 from kallithea.lib.utils2 import ascii_bytes, ascii_str, safe_bytes
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.model.db import (
     Permission, RepoGroup, Repository, User, UserApiKeys, UserGroup, UserGroupMember, UserGroupRepoGroupToPerm, UserGroupRepoToPerm, UserGroupToPerm, UserGroupUserGroupToPerm, UserIpMap, UserToPerm)
 from kallithea.model.db import (Permission, RepoGroup, Repository, User, UserApiKeys, UserGroup, UserGroupMember, UserGroupRepoGroupToPerm, UserGroupRepoToPerm,
                                 UserGroupToPerm, UserGroupUserGroupToPerm, UserIpMap, UserToPerm)
 from kallithea.model.meta import Session
 from kallithea.model.user import UserModel
 log = logging.getLogger(__name__)
 class PasswordGenerator(object):
     """
     This is a simple class for generating password from different sets of
     characters
     usage::
         passwd_gen = PasswordGenerator()
         #print 8-letter password containing only big and small letters
             of alphabet
         passwd_gen.gen_password(8, passwd_gen.ALPHABETS_BIG_SMALL)
     """
     ALPHABETS_NUM = r'''1234567890'''
     ALPHABETS_SMALL = r'''qwertyuiopasdfghjklzxcvbnm'''
     ALPHABETS_BIG = r'''QWERTYUIOPASDFGHJKLZXCVBNM'''
     ALPHABETS_SPECIAL = r'''`-=[]\;',./~!@#$%^&*()_+{}|:"<>?'''
     ALPHABETS_FULL = ALPHABETS_BIG + ALPHABETS_SMALL \
         + ALPHABETS_NUM + ALPHABETS_SPECIAL
     ALPHABETS_ALPHANUM = ALPHABETS_BIG + ALPHABETS_SMALL + ALPHABETS_NUM
     ALPHABETS_BIG_SMALL = ALPHABETS_BIG + ALPHABETS_SMALL
     ALPHABETS_ALPHANUM_BIG = ALPHABETS_BIG + ALPHABETS_NUM
     ALPHABETS_ALPHANUM_SMALL = ALPHABETS_SMALL + ALPHABETS_NUM
     def gen_password(self, length, alphabet=ALPHABETS_FULL):
         assert len(alphabet) <= 256, alphabet
         l = []
         while len(l) < length:
             i = ord(os.urandom(1))
             if i < len(alphabet):
                 l.append(alphabet[i])
         return ''.join(l)
 def get_crypt_password(password):
     """
     Cryptographic function used for bcrypt password hashing.
     :param password: password to hash
     """
     return ascii_str(bcrypt.hashpw(safe_bytes(password), bcrypt.gensalt(10)))
 def check_password(password, hashed):
     """
     Checks password match the hashed value using bcrypt.
     Remains backwards compatible and accept plain sha256 hashes which used to
     be used on Windows.
     :param password: password
     :param hashed: password in hashed form
     """
     # sha256 hashes will always be 64 hex chars
     # bcrypt hashes will always contain $ (and be shorter)
     if len(hashed) == 64 and all(x in string.hexdigits for x in hashed):
         return hashlib.sha256(password).hexdigest() == hashed
     try:
         return bcrypt.checkpw(safe_bytes(password), ascii_bytes(hashed))
     except ValueError as e:
         # bcrypt will throw ValueError 'Invalid hashed_password salt' on all password errors
         log.error('error from bcrypt checking password: %s', e)
         return False
     log.error('check_password failed - no method found for hash length %s', len(hashed))
     return False
 def _cached_perms_data(user_id, user_is_admin):
     RK = 'repositories'
     GK = 'repositories_groups'
     UK = 'user_groups'
     GLOBAL = 'global'
     PERM_WEIGHTS = Permission.PERM_WEIGHTS
     permissions = {RK: {}, GK: {}, UK: {}, GLOBAL: set()}
     def bump_permission(kind, key, new_perm):
         """Add a new permission for kind and key.
         Assuming the permissions are comparable, set the new permission if it
         has higher weight, else drop it and keep the old permission.
         """
         cur_perm = permissions[kind][key]
         new_perm_val = PERM_WEIGHTS[new_perm]
         cur_perm_val = PERM_WEIGHTS[cur_perm]
         if new_perm_val > cur_perm_val:
             permissions[kind][key] = new_perm
     #======================================================================
     # fetch default permissions
     #======================================================================
     default_user = User.get_by_username('default', cache=True)
     default_user_id = default_user.user_id
     default_repo_perms = Permission.get_default_perms(default_user_id)
     default_repo_groups_perms = Permission.get_default_group_perms(default_user_id)
     default_user_group_perms = Permission.get_default_user_group_perms(default_user_id)
     if user_is_admin:
         #==================================================================
         # admin users have all rights;
         # based on default permissions, just set everything to admin
         #==================================================================
         permissions[GLOBAL].add('hg.admin')
         permissions[GLOBAL].add('hg.create.write_on_repogroup.true')
         # repositories
         for perm in default_repo_perms:
             r_k = perm.UserRepoToPerm.repository.repo_name
             p = 'repository.admin'
             permissions[RK][r_k] = p
         # repository groups
         for perm in default_repo_groups_perms:
             rg_k = perm.UserRepoGroupToPerm.group.group_name
             p = 'group.admin'
             permissions[GK][rg_k] = p
         # user groups
         for perm in default_user_group_perms:
             u_k = perm.UserUserGroupToPerm.user_group.users_group_name
             p = 'usergroup.admin'
             permissions[UK][u_k] = p
         return permissions
     #==================================================================
     # SET DEFAULTS GLOBAL, REPOS, REPOSITORY GROUPS
     #==================================================================
     # default global permissions taken from the default user
     default_global_perms = UserToPerm.query() \
         .filter(UserToPerm.user_id == default_user_id) \
         .options(joinedload(UserToPerm.permission))
     for perm in default_global_perms:
         permissions[GLOBAL].add(perm.permission.permission_name)
     # defaults for repositories, taken from default user
     for perm in default_repo_perms:
         r_k = perm.UserRepoToPerm.repository.repo_name
         if perm.Repository.owner_id == user_id:
             p = 'repository.admin'
         elif perm.Repository.private:
             p = 'repository.none'
         else:
             p = perm.Permission.permission_name
         permissions[RK][r_k] = p
     # defaults for repository groups taken from default user permission
     # on given group
     for perm in default_repo_groups_perms:
         rg_k = perm.UserRepoGroupToPerm.group.group_name
         p = perm.Permission.permission_name
         permissions[GK][rg_k] = p
     # defaults for user groups taken from default user permission
     # on given user group
     for perm in default_user_group_perms:
         u_k = perm.UserUserGroupToPerm.user_group.users_group_name
         p = perm.Permission.permission_name
         permissions[UK][u_k] = p
     #======================================================================
     # !! Augment GLOBALS with user permissions if any found !!
     #======================================================================
     # USER GROUPS comes first
     # user group global permissions
     user_perms_from_users_groups = Session().query(UserGroupToPerm) \
         .options(joinedload(UserGroupToPerm.permission)) \
         .join((UserGroupMember, UserGroupToPerm.users_group_id ==
                UserGroupMember.users_group_id)) \
         .filter(UserGroupMember.user_id == user_id) \
         .join((UserGroup, UserGroupMember.users_group_id ==
                UserGroup.users_group_id)) \
         .filter(UserGroup.users_group_active == True) \
         .order_by(UserGroupToPerm.users_group_id) \
         .all()
     # need to group here by groups since user can be in more than
     # one group
     _grouped = [[x, list(y)] for x, y in
                 itertools.groupby(user_perms_from_users_groups,
                                   lambda x:x.users_group)]
     for gr, perms in _grouped:
         for perm in perms:
             permissions[GLOBAL].add(perm.permission.permission_name)
     # user specific global permissions
     user_perms = Session().query(UserToPerm) \
             .options(joinedload(UserToPerm.permission)) \
             .filter(UserToPerm.user_id == user_id).all()
     for perm in user_perms:
         permissions[GLOBAL].add(perm.permission.permission_name)
     # for each kind of global permissions, only keep the one with heighest weight
     kind_max_perm = {}
     for perm in sorted(permissions[GLOBAL], key=lambda n: PERM_WEIGHTS[n]):
         kind = perm.rsplit('.', 1)[0]
         kind_max_perm[kind] = perm
     permissions[GLOBAL] = set(kind_max_perm.values())
     ## END GLOBAL PERMISSIONS
     #======================================================================
     # !! PERMISSIONS FOR REPOSITORIES !!
     #======================================================================
     #======================================================================
     # check if user is part of user groups for this repository and
     # fill in his permission from it.
     #======================================================================
     # user group for repositories permissions
     user_repo_perms_from_users_groups = \
      Session().query(UserGroupRepoToPerm, Permission, Repository,) \
         .join((Repository, UserGroupRepoToPerm.repository_id ==
                Repository.repo_id)) \
         .join((Permission, UserGroupRepoToPerm.permission_id ==
                Permission.permission_id)) \
         .join((UserGroup, UserGroupRepoToPerm.users_group_id ==
                UserGroup.users_group_id)) \
         .filter(UserGroup.users_group_active == True) \
         .join((UserGroupMember, UserGroupRepoToPerm.users_group_id ==
                UserGroupMember.users_group_id)) \
         .filter(UserGroupMember.user_id == user_id) \
         .all()
     for perm in user_repo_perms_from_users_groups:
         bump_permission(RK,
             perm.UserGroupRepoToPerm.repository.repo_name,
             perm.Permission.permission_name)
     # user permissions for repositories
     user_repo_perms = Permission.get_default_perms(user_id)
     for perm in user_repo_perms:
         bump_permission(RK,
             perm.UserRepoToPerm.repository.repo_name,
             perm.Permission.permission_name)
     #======================================================================
     # !! PERMISSIONS FOR REPOSITORY GROUPS !!
     #======================================================================
     #======================================================================
     # check if user is part of user groups for this repository groups and
     # fill in his permission from it.
     #======================================================================
     # user group for repo groups permissions
     user_repo_group_perms_from_users_groups = \
      Session().query(UserGroupRepoGroupToPerm, Permission, RepoGroup) \
      .join((RepoGroup, UserGroupRepoGroupToPerm.group_id == RepoGroup.group_id)) \
      .join((Permission, UserGroupRepoGroupToPerm.permission_id
             == Permission.permission_id)) \
      .join((UserGroup, UserGroupRepoGroupToPerm.users_group_id ==
             UserGroup.users_group_id)) \
      .filter(UserGroup.users_group_active == True) \
      .join((UserGroupMember, UserGroupRepoGroupToPerm.users_group_id
             == UserGroupMember.users_group_id)) \
      .filter(UserGroupMember.user_id == user_id) \
      .all()
     for perm in user_repo_group_perms_from_users_groups:
         bump_permission(GK,
             perm.UserGroupRepoGroupToPerm.group.group_name,
             perm.Permission.permission_name)
     # user explicit permissions for repository groups
     user_repo_groups_perms = Permission.get_default_group_perms(user_id)
     for perm in user_repo_groups_perms:
         bump_permission(GK,
             perm.UserRepoGroupToPerm.group.group_name,
             perm.Permission.permission_name)
     #======================================================================
     # !! PERMISSIONS FOR USER GROUPS !!
     #======================================================================
     # user group for user group permissions
     user_group_user_groups_perms = \
      Session().query(UserGroupUserGroupToPerm, Permission, UserGroup) \
      .join((UserGroup, UserGroupUserGroupToPerm.target_user_group_id
             == UserGroup.users_group_id)) \
      .join((Permission, UserGroupUserGroupToPerm.permission_id
             == Permission.permission_id)) \
      .join((UserGroupMember, UserGroupUserGroupToPerm.user_group_id
             == UserGroupMember.users_group_id)) \
      .filter(UserGroupMember.user_id == user_id) \
      .join((UserGroup, UserGroupMember.users_group_id ==
             UserGroup.users_group_id), aliased=True, from_joinpoint=True) \
      .filter(UserGroup.users_group_active == True) \
      .all()
     for perm in user_group_user_groups_perms:
         bump_permission(UK,
             perm.UserGroupUserGroupToPerm.target_user_group.users_group_name,
             perm.Permission.permission_name)
     # user explicit permission for user groups
     user_user_groups_perms = Permission.get_default_user_group_perms(user_id)
     for perm in user_user_groups_perms:
         bump_permission(UK,
             perm.UserUserGroupToPerm.user_group.users_group_name,
             perm.Permission.permission_name)
     return permissions
 class AuthUser(object):
     """
     Represents a Kallithea user, including various authentication and
     authorization information. Typically used to store the current user,
     but is also used as a generic user information data structure in
     parts of the code, e.g. user management.
     Constructed from a database `User` object, a user ID or cookie dict,
     it looks up the user (if needed) and copies all attributes to itself,
     adding various non-persistent data. If lookup fails but anonymous
     access to Kallithea is enabled, the default user is loaded instead.
     `AuthUser` does not by itself authenticate users. It's up to other parts of
     the code to check e.g. if a supplied password is correct, and if so, trust
     the AuthUser object as an authenticated user.
     However, `AuthUser` does refuse to load a user that is not `active`.
     Note that Kallithea distinguishes between the default user (an actual
     user in the database with username "default") and "no user" (no actual
     User object, AuthUser filled with blank values and username "None").
     If the default user is active, that will always be used instead of
     "no user". On the other hand, if the default user is disabled (and
     there is no login information), we instead get "no user"; this should
     only happen on the login page (as all other requests are redirected).
     `is_default_user` specifically checks if the AuthUser is the user named
     "default". Use `is_anonymous` to check for both "default" and "no user".
     """
     @classmethod
     def make(cls, dbuser=None, is_external_auth=False, ip_addr=None):
         """Create an AuthUser to be authenticated ... or return None if user for some reason can't be authenticated.
         Checks that a non-None dbuser is provided, is active, and that the IP address is ok.
         """
         assert ip_addr is not None
         if dbuser is None:
             log.info('No db user for authentication')
             return None
         if not dbuser.active:
             log.info('Db user %s not active', dbuser.username)
             return None
         allowed_ips = AuthUser.get_allowed_ips(dbuser.user_id, cache=True)
         if not check_ip_access(source_ip=ip_addr, allowed_ips=allowed_ips):
             log.info('Access for %s from %s forbidden - not in %s', dbuser.username, ip_addr, allowed_ips)
             return None
         return cls(dbuser=dbuser, is_external_auth=is_external_auth)
     def __init__(self, user_id=None, dbuser=None, is_external_auth=False):
         self.is_external_auth = is_external_auth # container auth - don't show logout option
         # These attributes will be overridden below if the requested user is
         # found or anonymous access (using the default user) is enabled.
         self.user_id = None
         self.username = None
         self.api_key = None
         self.name = ''
         self.lastname = ''
         self.email = ''
         self.admin = False
         # Look up database user, if necessary.
         if user_id is not None:
             assert dbuser is None
             log.debug('Auth User lookup by USER ID %s', user_id)
             dbuser = UserModel().get(user_id)
             assert dbuser is not None
         else:
             assert dbuser is not None
             log.debug('Auth User lookup by database user %s', dbuser)
         log.debug('filling %s data', dbuser)
         self.is_anonymous = dbuser.is_default_user
         if dbuser.is_default_user and not dbuser.active:
             self.username = 'None'
             self.is_default_user = False
         else:
             # copy non-confidential database fields from a `db.User` to this `AuthUser`.
             for k, v in dbuser.get_dict().items():
                 assert k not in ['api_keys', 'permissions']
                 setattr(self, k, v)
             self.is_default_user = dbuser.is_default_user
         log.debug('Auth User is now %s', self)
     @LazyProperty
     def permissions(self):
         return self.__get_perms(user=self, cache=False)
     def has_repository_permission_level(self, repo_name, level, purpose=None):
         required_perms = {
             'read': ['repository.read', 'repository.write', 'repository.admin'],
             'write': ['repository.write', 'repository.admin'],
             'admin': ['repository.admin'],
         }[level]
         actual_perm = self.permissions['repositories'].get(repo_name)
         ok = actual_perm in required_perms
         log.debug('Checking if user %r can %r repo %r (%s): %s (has %r)',
             self.username, level, repo_name, purpose, ok, actual_perm)
         return ok
     def has_repository_group_permission_level(self, repo_group_name, level, purpose=None):
         required_perms = {
             'read': ['group.read', 'group.write', 'group.admin'],
             'write': ['group.write', 'group.admin'],
             'admin': ['group.admin'],
         }[level]
         actual_perm = self.permissions['repositories_groups'].get(repo_group_name)
         ok = actual_perm in required_perms
         log.debug('Checking if user %r can %r repo group %r (%s): %s (has %r)',
             self.username, level, repo_group_name, purpose, ok, actual_perm)
         return ok
     def has_user_group_permission_level(self, user_group_name, level, purpose=None):
         required_perms = {
             'read': ['usergroup.read', 'usergroup.write', 'usergroup.admin'],
             'write': ['usergroup.write', 'usergroup.admin'],
             'admin': ['usergroup.admin'],
         }[level]
         actual_perm = self.permissions['user_groups'].get(user_group_name)
         ok = actual_perm in required_perms
         log.debug('Checking if user %r can %r user group %r (%s): %s (has %r)',
             self.username, level, user_group_name, purpose, ok, actual_perm)
         return ok
     @property
     def api_keys(self):
         return self._get_api_keys()
     def __get_perms(self, user, cache=False):
         """
         Fills user permission attribute with permissions taken from database
         works for permissions given for repositories, and for permissions that
         are granted to groups
         :param user: `AuthUser` instance
         """
         user_id = user.user_id
         user_is_admin = user.is_admin
         log.debug('Getting PERMISSION tree')
         compute = conditional_cache('short_term', 'cache_desc',
                                     condition=cache, func=_cached_perms_data)
         return compute(user_id, user_is_admin)
     def _get_api_keys(self):
         api_keys = [self.api_key]
         for api_key in UserApiKeys.query() \
                 .filter_by(user_id=self.user_id, is_expired=False):
             api_keys.append(api_key.api_key)
         return api_keys
     @property
     def is_admin(self):
         return self.admin
     @property
     def repositories_admin(self):
         """
         Returns list of repositories you're an admin of
         """
         return [x[0] for x in self.permissions['repositories'].items()
                 if x[1] == 'repository.admin']
     @property
     def repository_groups_admin(self):
         """
         Returns list of repository groups you're an admin of
         """
         return [x[0] for x in self.permissions['repositories_groups'].items()
                 if x[1] == 'group.admin']
     @property
     def user_groups_admin(self):
         """
         Returns list of user groups you're an admin of
         """
         return [x[0] for x in self.permissions['user_groups'].items()
                 if x[1] == 'usergroup.admin']
     def __repr__(self):
         return "<%s %s: %r>" % (self.__class__.__name__, self.user_id, self.username)
     def to_cookie(self):
         """ Serializes this login session to a cookie `dict`. """
         return {
             'user_id': self.user_id,
             'is_external_auth': self.is_external_auth,
+        }
     @staticmethod
     def from_cookie(cookie, ip_addr):
         """
         Deserializes an `AuthUser` from a cookie `dict` ... or return None.
         """
         return AuthUser.make(
             dbuser=UserModel().get(cookie.get('user_id')),
             is_external_auth=cookie.get('is_external_auth', False),
             ip_addr=ip_addr,
+        )
     @classmethod
     def get_allowed_ips(cls, user_id, cache=False):
         _set = set()
         default_ips = UserIpMap.query().filter(UserIpMap.user_id ==
                                         User.get_default_user(cache=True).user_id)
         if cache:
             default_ips = default_ips.options(FromCache("sql_cache_short",
                                               "get_user_ips_default"))
         for ip in default_ips:
             try:
                 _set.add(ip.ip_addr)
             except ObjectDeletedError:
                 # since we use heavy caching sometimes it happens that we get
                 # deleted objects here, we just skip them
                 pass
         user_ips = UserIpMap.query().filter(UserIpMap.user_id == user_id)
         if cache:
             user_ips = user_ips.options(FromCache("sql_cache_short",
                                                   "get_user_ips_%s" % user_id))
         for ip in user_ips:
             try:
                 _set.add(ip.ip_addr)
             except ObjectDeletedError:
                 # since we use heavy caching sometimes it happens that we get
                 # deleted objects here, we just skip them
                 pass
         return _set or set(['0.0.0.0/0', '::/0'])
 #==============================================================================
 # CHECK DECORATORS
 #==============================================================================
 def _redirect_to_login(message=None):
     """Return an exception that must be raised. It will redirect to the login
     page which will redirect back to the current URL after authentication.
     The optional message will be shown in a flash message."""
     from kallithea.lib import helpers as h
     if message:
         h.flash(message, category='warning')
     p = request.path_qs
     log.debug('Redirecting to login page, origin: %s', p)
     return HTTPFound(location=url('login_home', came_from=p))
 # Use as decorator
 class LoginRequired(object):
     """Client must be logged in as a valid User, or we'll redirect to the login
     page.
     If the "default" user is enabled and allow_default_user is true, that is
     considered valid too.
     Also checks that IP address is allowed.
     """
     def __init__(self, allow_default_user=False):
         self.allow_default_user = allow_default_user
     def __call__(self, func):
         return decorator(self.__wrapper, func)
     def __wrapper(self, func, *fargs, **fkwargs):
         controller = fargs[0]
         user = request.authuser
         loc = "%s:%s" % (controller.__class__.__name__, func.__name__)
         log.debug('Checking access for user %s @ %s', user, loc)
         # regular user authentication
         if user.is_default_user:
             if self.allow_default_user:
                 log.info('default user @ %s', loc)
                 return func(*fargs, **fkwargs)
             log.info('default user is not accepted here @ %s', loc)
         elif user.is_anonymous: # default user is disabled and no proper authentication
             log.warning('user is anonymous and NOT authenticated with regular auth @ %s', loc)
         else: # regular authentication
             log.info('user %s authenticated with regular auth @ %s', user, loc)
             return func(*fargs, **fkwargs)
         raise _redirect_to_login()
 # Use as decorator
 class NotAnonymous(object):
     """Ensures that client is not logged in as the "default" user, and
     redirects to the login page otherwise. Must be used together with
     LoginRequired."""
     def __call__(self, func):
         return decorator(self.__wrapper, func)
     def __wrapper(self, func, *fargs, **fkwargs):
         cls = fargs[0]
         user = request.authuser
         log.debug('Checking that user %s is not anonymous @%s', user.username, cls)
         if user.is_default_user:
             raise _redirect_to_login(_('You need to be a registered user to '
                                        'perform this action'))
         else:
             return func(*fargs, **fkwargs)
 class _PermsDecorator(object):
     """Base class for controller decorators with multiple permissions"""
     def __init__(self, *required_perms):
         self.required_perms = required_perms # usually very short - a list is thus fine
     def __call__(self, func):
         return decorator(self.__wrapper, func)
     def __wrapper(self, func, *fargs, **fkwargs):
         cls = fargs[0]
         user = request.authuser
         log.debug('checking %s permissions %s for %s %s',
           self.__class__.__name__, self.required_perms, cls, user)
         if self.check_permissions(user):
             log.debug('Permission granted for %s %s', cls, user)
             return func(*fargs, **fkwargs)
         else:
             log.info('Permission denied for %s %s', cls, user)
             if user.is_default_user:
                 raise _redirect_to_login(_('You need to be signed in to view this page'))
             else:
                 raise HTTPForbidden()
     def check_permissions(self, user):
         raise NotImplementedError()
 class HasPermissionAnyDecorator(_PermsDecorator):
     """
     Checks the user has any of the given global permissions.
     """
     def check_permissions(self, user):
         global_permissions = user.permissions['global'] # usually very short
         return any(p in global_permissions for p in self.required_perms)
 class _PermDecorator(_PermsDecorator):
     """Base class for controller decorators with a single permission"""
     def __init__(self, required_perm):
         _PermsDecorator.__init__(self, [required_perm])
         self.required_perm = required_perm
 class HasRepoPermissionLevelDecorator(_PermDecorator):
     """
     Checks the user has at least the specified permission level for the requested repository.
     """
     def check_permissions(self, user):
         repo_name = get_repo_slug(request)
         return user.has_repository_permission_level(repo_name, self.required_perm)
 class HasRepoGroupPermissionLevelDecorator(_PermDecorator):
     """
     Checks the user has any of given permissions for the requested repository group.
     """
     def check_permissions(self, user):
         repo_group_name = get_repo_group_slug(request)
         return user.has_repository_group_permission_level(repo_group_name, self.required_perm)
 class HasUserGroupPermissionLevelDecorator(_PermDecorator):
     """
     Checks for access permission for any of given predicates for specific
     user group. In order to fulfill the request any of predicates must be meet
     """
     def check_permissions(self, user):
         user_group_name = get_user_group_slug(request)
         return user.has_user_group_permission_level(user_group_name, self.required_perm)
 #==============================================================================
 # CHECK FUNCTIONS
 #==============================================================================
 class _PermsFunction(object):
     """Base function for other check functions with multiple permissions"""
     def __init__(self, *required_perms):
         self.required_perms = required_perms # usually very short - a list is thus fine
     def __bool__(self):
         """ Defend against accidentally forgetting to call the object
             and instead evaluating it directly in a boolean context,
             which could have security implications.
         """
         raise AssertionError(self.__class__.__name__ + ' is not a bool and must be called!')
     def __call__(self, *a, **b):
         raise NotImplementedError()
 class HasPermissionAny(_PermsFunction):
     def __call__(self, purpose=None):
         global_permissions = request.authuser.permissions['global'] # usually very short
         ok = any(p in global_permissions for p in self.required_perms)
         log.debug('Check %s for global %s (%s): %s',
             request.authuser.username, self.required_perms, purpose, ok)
         return ok
 class _PermFunction(_PermsFunction):
     """Base function for other check functions with a single permission"""
     def __init__(self, required_perm):
         _PermsFunction.__init__(self, [required_perm])
         self.required_perm = required_perm
 class HasRepoPermissionLevel(_PermFunction):
     def __call__(self, repo_name, purpose=None):
         return request.authuser.has_repository_permission_level(repo_name, self.required_perm, purpose)
 class HasRepoGroupPermissionLevel(_PermFunction):
     def __call__(self, group_name, purpose=None):
         return request.authuser.has_repository_group_permission_level(group_name, self.required_perm, purpose)
 class HasUserGroupPermissionLevel(_PermFunction):
     def __call__(self, user_group_name, purpose=None):
         return request.authuser.has_user_group_permission_level(user_group_name, self.required_perm, purpose)
 #==============================================================================
 # SPECIAL VERSION TO HANDLE MIDDLEWARE AUTH
 #==============================================================================
 class HasPermissionAnyMiddleware(object):
     def __init__(self, *perms):
         self.required_perms = set(perms)
     def __call__(self, authuser, repo_name, purpose=None):
         try:
             ok = authuser.permissions['repositories'][repo_name] in self.required_perms
         except KeyError:
             ok = False
         log.debug('Middleware check %s for %s for repo %s (%s): %s', authuser.username, self.required_perms, repo_name, purpose, ok)
         return ok

kallithea/lib/vcs/backends/base.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.backends.base
     ~~~~~~~~~~~~~~~~~
     Base for all available scm backends
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import datetime
 import itertools
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import (
     ChangesetError, EmptyRepositoryError, NodeAlreadyAddedError, NodeAlreadyChangedError, NodeAlreadyExistsError, NodeAlreadyRemovedError, NodeDoesNotExistError, NodeNotChangedError, RepositoryError)
 from kallithea.lib.vcs.exceptions import (ChangesetError, EmptyRepositoryError, NodeAlreadyAddedError, NodeAlreadyChangedError, NodeAlreadyExistsError,
                                           NodeAlreadyRemovedError, NodeDoesNotExistError, NodeNotChangedError, RepositoryError)
 from kallithea.lib.vcs.utils import author_email, author_name
 from kallithea.lib.vcs.utils.helpers import get_dict_for_attrs
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 class BaseRepository(object):
     """
     Base Repository for final backends
     **Attributes**
         ``DEFAULT_BRANCH_NAME``
             name of default branch (i.e. "trunk" for svn, "master" for git etc.
         ``scm``
             alias of scm, i.e. *git* or *hg*
         ``repo``
             object from external api
         ``revisions``
             list of all available revisions' ids, in ascending order
         ``changesets``
             storage dict caching returned changesets
         ``path``
             absolute path to the repository
         ``branches``
             branches as list of changesets
         ``tags``
             tags as list of changesets
     """
     scm = None
     DEFAULT_BRANCH_NAME = None
     EMPTY_CHANGESET = '0' * 40
     def __init__(self, repo_path, create=False, **kwargs):
         """
         Initializes repository. Raises RepositoryError if repository could
         not be find at the given ``repo_path`` or directory at ``repo_path``
         exists and ``create`` is set to True.
         :param repo_path: local path of the repository
         :param create=False: if set to True, would try to create repository.
         :param src_url=None: if set, should be proper url from which repository
           would be cloned; requires ``create`` parameter to be set to True -
           raises RepositoryError if src_url is set and create evaluates to
           False
         """
         raise NotImplementedError
     def __str__(self):
         return '<%s at %s>' % (self.__class__.__name__, self.path)
     def __repr__(self):
         return self.__str__()
     def __len__(self):
         return self.count()
     def __eq__(self, other):
         same_instance = isinstance(other, self.__class__)
         return same_instance and getattr(other, 'path', None) == self.path
     def __ne__(self, other):
         return not self.__eq__(other)
     @LazyProperty
     def alias(self):
         for k, v in settings.BACKENDS.items():
             if v.split('.')[-1] == str(self.__class__.__name__):
                 return k
     @LazyProperty
     def name(self):
         """
         Return repository name (without group name)
         """
         raise NotImplementedError
     @LazyProperty
     def owner(self):
         raise NotImplementedError
     @LazyProperty
     def description(self):
         raise NotImplementedError
     @LazyProperty
     def size(self):
         """
         Returns combined size in bytes for all repository files
         """
         size = 0
         try:
             tip = self.get_changeset()
             for topnode, dirs, files in tip.walk('/'):
                 for f in files:
                     size += tip.get_file_size(f.path)
         except RepositoryError as e:
             pass
         return size
     def is_valid(self):
         """
         Validates repository.
         """
         raise NotImplementedError
     def is_empty(self):
         return self._empty
     #==========================================================================
     # CHANGESETS
     #==========================================================================
     def get_changeset(self, revision=None):
         """
         Returns instance of ``Changeset`` class. If ``revision`` is None, most
         recent changeset is returned.
         :raises ``EmptyRepositoryError``: if there are no revisions
         """
         raise NotImplementedError
     def __iter__(self):
         """
         Allows Repository objects to be iterated.
         *Requires* implementation of ``__getitem__`` method.
         """
         for revision in self.revisions:
             yield self.get_changeset(revision)
     def get_changesets(self, start=None, end=None, start_date=None,
                        end_date=None, branch_name=None, reverse=False, max_revisions=None):
         """
         Returns iterator of ``BaseChangeset`` objects from start to end,
         both inclusive.
         :param start: None or str
         :param end: None or str
         :param start_date:
         :param end_date:
         :param branch_name:
         :param reversed:
         """
         raise NotImplementedError
     def __getitem__(self, key):
         if isinstance(key, slice):
             return (self.get_changeset(rev) for rev in self.revisions[key])
         return self.get_changeset(key)
     def count(self):
         return len(self.revisions)
     def tag(self, name, user, revision=None, message=None, date=None, **opts):
         """
         Creates and returns a tag for the given ``revision``.
         :param name: name for new tag
         :param user: full username, i.e.: "Joe Doe <joe.doe@example.com>"
         :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
         """
         raise NotImplementedError
     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 <joe.doe@example.com>"
         :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
         """
         raise NotImplementedError
     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``.
         """
         raise NotImplementedError
     # ========== #
     # COMMIT API #
     # ========== #
     @LazyProperty
     def in_memory_changeset(self):
         """
         Returns ``InMemoryChangeset`` object for this repository.
         """
         raise NotImplementedError
     def add(self, filenode, **kwargs):
         """
         Commit api function that will add given ``FileNode`` into this
         repository.
         :raises ``NodeAlreadyExistsError``: if there is a file with same path
           already in repository
         :raises ``NodeAlreadyAddedError``: if given node is already marked as
           *added*
         """
         raise NotImplementedError
     def remove(self, filenode, **kwargs):
         """
         Commit api function that will remove given ``FileNode`` into this
         repository.
         :raises ``EmptyRepositoryError``: if there are no changesets yet
         :raises ``NodeDoesNotExistError``: if there is no file with given path
         """
         raise NotImplementedError
     def commit(self, message, **kwargs):
         """
         Persists current changes made on this repository and returns newly
         created changeset.
         """
         raise NotImplementedError
     def get_state(self):
         """
         Returns dictionary with ``added``, ``changed`` and ``removed`` lists
         containing ``FileNode`` objects.
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     # =========== #
     # WORKDIR API #
     # =========== #
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         raise NotImplementedError
 class BaseChangeset(object):
     """
     Each backend should implement it's changeset representation.
     **Attributes**
         ``repository``
             repository object within which changeset exists
         ``raw_id``
             raw changeset representation (i.e. full 40 length sha for git
             backend)
         ``short_id``
             shortened (if apply) version of ``raw_id``; it would be simple
             shortcut for ``raw_id[:12]`` for git/mercurial backends or same
             as ``raw_id`` for subversion
         ``revision``
             revision number as integer
         ``files``
             list of ``FileNode`` (``Node`` with NodeKind.FILE) objects
         ``dirs``
             list of ``DirNode`` (``Node`` with NodeKind.DIR) objects
         ``nodes``
             combined list of ``Node`` objects
         ``author``
             author of the changeset, as str
         ``message``
             message of the changeset, as str
         ``parents``
             list of parent changesets
         ``last``
             ``True`` if this is last changeset in repository, ``False``
             otherwise; trying to access this attribute while there is no
             changesets would raise ``EmptyRepositoryError``
     """
     def __str__(self):
         return '<%s at %s:%s>' % (self.__class__.__name__, self.revision,
             self.short_id)
     def __repr__(self):
         return self.__str__()
     def __eq__(self, other):
         if type(self) is not type(other):
             return False
         return self.raw_id == other.raw_id
     def __json__(self, with_file_list=False):
         if with_file_list:
             return dict(
                 short_id=self.short_id,
                 raw_id=self.raw_id,
                 revision=self.revision,
                 message=self.message,
                 date=self.date,
                 author=self.author,
                 added=[el.path for el in self.added],
                 changed=[el.path for el in self.changed],
                 removed=[el.path for el in self.removed],
+            )
         else:
             return dict(
                 short_id=self.short_id,
                 raw_id=self.raw_id,
                 revision=self.revision,
                 message=self.message,
                 date=self.date,
                 author=self.author,
+            )
     @LazyProperty
     def last(self):
         if self.repository is None:
             raise ChangesetError("Cannot check if it's most recent revision")
         return self.raw_id == self.repository.revisions[-1]
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         raise NotImplementedError
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         raise NotImplementedError
     @LazyProperty
     def raw_id(self):
         """
         Returns raw string identifying this changeset.
         """
         raise NotImplementedError
     @LazyProperty
     def short_id(self):
         """
         Returns shortened version of ``raw_id`` attribute, as string,
         identifying this changeset, useful for web representation.
         """
         raise NotImplementedError
     @LazyProperty
     def revision(self):
         """
         Returns integer identifying this changeset.
         """
         raise NotImplementedError
     @LazyProperty
     def committer(self):
         """
         Returns Committer for given commit
         """
         raise NotImplementedError
     @LazyProperty
     def committer_name(self):
         """
         Returns Author name for given commit
         """
         return author_name(self.committer)
     @LazyProperty
     def committer_email(self):
         """
         Returns Author email address for given commit
         """
         return author_email(self.committer)
     @LazyProperty
     def author(self):
         """
         Returns Author for given commit
         """
         raise NotImplementedError
     @LazyProperty
     def author_name(self):
         """
         Returns Author name for given commit
         """
         return author_name(self.author)
     @LazyProperty
     def author_email(self):
         """
         Returns Author email address for given commit
         """
         return author_email(self.author)
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_content(self, path):
         """
         Returns content of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_size(self, path):
         """
         Returns size of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_changeset(self, path):
         """
         Returns last commit of the file at the given ``path``.
         """
         raise NotImplementedError
     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.
         """
         raise NotImplementedError
     def get_nodes(self, path):
         """
         Returns combined ``DirNode`` and ``FileNode`` objects list representing
         state of changeset at the given ``path``.
         :raises ``ChangesetError``: if node at the given ``path`` is not
           instance of ``DirNode``
         """
         raise NotImplementedError
     def get_node(self, path):
         """
         Returns ``Node`` object from the given ``path``.
         :raises ``NodeDoesNotExistError``: if there is no node at the given
           ``path``
         """
         raise NotImplementedError
     def fill_archive(self, stream=None, kind='tgz', prefix=None):
         """
         Fills up given stream.
         :param stream: file like object.
         :param kind: one of following: ``zip``, ``tar``, ``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>
         """
         raise NotImplementedError
     def get_chunked_archive(self, **kwargs):
         """
         Returns iterable archive. Tiny wrapper around ``fill_archive`` method.
         :param chunk_size: extra parameter which controls size of returned
             chunks. Default:8k.
         """
         chunk_size = kwargs.pop('chunk_size', 8192)
         stream = kwargs.get('stream')
         self.fill_archive(**kwargs)
         while True:
             data = stream.read(chunk_size)
             if not data:
                 break
             yield data
     @LazyProperty
     def root(self):
         """
         Returns ``RootNode`` object for this changeset.
         """
         return self.get_node('')
     def next(self, branch=None):
         """
         Returns next changeset from current, if branch is gives it will return
         next changeset belonging to this branch
         :param branch: show changesets within the given named branch
         """
         raise NotImplementedError
     def prev(self, branch=None):
         """
         Returns previous changeset from current, if branch is gives it will
         return previous changeset belonging to this branch
         :param branch: show changesets within the given named branch
         """
         raise NotImplementedError
     @LazyProperty
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def size(self):
         """
         Returns total number of bytes from contents of all filenodes.
         """
         return sum((node.size for node in self.get_filenodes_generator()))
     def walk(self, topurl=''):
         """
         Similar to os.walk method. Instead of filesystem it walks through
         changeset starting at given ``topurl``.  Returns generator of tuples
         (topnode, dirnodes, filenodes).
         """
         topnode = self.get_node(topurl)
         yield (topnode, topnode.dirs, topnode.files)
         for dirnode in topnode.dirs:
             for tup in self.walk(dirnode.path):
                 yield tup
     def get_filenodes_generator(self):
         """
         Returns generator that yields *all* file nodes.
         """
         for topnode, dirs, files in self.walk():
             for node in files:
                 yield node
     def as_dict(self):
         """
         Returns dictionary with changeset's attributes and their values.
         """
         data = get_dict_for_attrs(self, ['raw_id', 'short_id',
             'revision', 'date', 'message'])
         data['author'] = {'name': self.author_name, 'email': self.author_email}
         data['added'] = [node.path for node in self.added]
         data['changed'] = [node.path for node in self.changed]
         data['removed'] = [node.path for node in self.removed]
         return data
     @LazyProperty
     def closesbranch(self):
         return False
     @LazyProperty
     def obsolete(self):
         return False
     @LazyProperty
     def bumped(self):
         return False
     @LazyProperty
     def divergent(self):
         return False
     @LazyProperty
     def extinct(self):
         return False
     @LazyProperty
     def unstable(self):
         return False
     @LazyProperty
     def phase(self):
         return ''
 class BaseWorkdir(object):
     """
     Working directory representation of single repository.
     :attribute: repository: repository object of working directory
     """
     def __init__(self, repository):
         self.repository = repository
     def get_branch(self):
         """
         Returns name of current branch.
         """
         raise NotImplementedError
     def get_changeset(self):
         """
         Returns current changeset.
         """
         raise NotImplementedError
     def get_added(self):
         """
         Returns list of ``FileNode`` objects marked as *new* in working
         directory.
         """
         raise NotImplementedError
     def get_changed(self):
         """
         Returns list of ``FileNode`` objects *changed* in working directory.
         """
         raise NotImplementedError
     def get_removed(self):
         """
         Returns list of ``RemovedFileNode`` objects marked as *removed* in
         working directory.
         """
         raise NotImplementedError
     def get_untracked(self):
         """
         Returns list of ``FileNode`` objects which are present within working
         directory however are not tracked by repository.
         """
         raise NotImplementedError
     def get_status(self):
         """
         Returns dict with ``added``, ``changed``, ``removed`` and ``untracked``
         lists.
         """
         raise NotImplementedError
     def commit(self, message, **kwargs):
         """
         Commits local (from working directory) changes and returns newly
         created
         ``Changeset``. Updates repository's ``revisions`` list.
         :raises ``CommitError``: if any error occurs while committing
         """
         raise NotImplementedError
     def update(self, revision=None):
         """
         Fetches content of the given revision and populates it within working
         directory.
         """
         raise NotImplementedError
     def checkout_branch(self, branch=None):
         """
         Checks out ``branch`` or the backend's default branch.
         Raises ``BranchDoesNotExistError`` if the branch does not exist.
         """
         raise NotImplementedError
 class BaseInMemoryChangeset(object):
     """
     Represents differences between repository's state (most recent head) and
     changes made *in place*.
     **Attributes**
         ``repository``
             repository object for this in-memory-changeset
         ``added``
             list of ``FileNode`` objects marked as *added*
         ``changed``
             list of ``FileNode`` objects marked as *changed*
         ``removed``
             list of ``FileNode`` or ``RemovedFileNode`` objects marked to be
             *removed*
         ``parents``
             list of ``Changeset`` representing parents of in-memory changeset.
             Should always be 2-element sequence.
     """

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

➞

Show inline comments

 import re
 from io import BytesIO
 from itertools import chain
 from subprocess import PIPE, Popen
 from dulwich import objects
 from dulwich.config import ConfigFile
 from kallithea.lib.vcs.backends.base import BaseChangeset, EmptyChangeset
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import ChangesetDoesNotExistError, ChangesetError, ImproperArchiveTypeError, NodeDoesNotExistError, RepositoryError, VCSError
 from kallithea.lib.vcs.nodes import (
     AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode, SubModuleNode)
 from kallithea.lib.vcs.nodes import (AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode,
                                      SubModuleNode)
 from kallithea.lib.vcs.utils import ascii_bytes, ascii_str, date_fromtimestamp, safe_int, safe_str
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 class GitChangeset(BaseChangeset):
     """
     Represents state of the repository at a revision.
     """
     def __init__(self, repository, revision):
         self._stat_modes = {}
         self.repository = repository
         try:
             commit = self.repository._repo[ascii_bytes(revision)]
             if isinstance(commit, objects.Tag):
                 revision = safe_str(commit.object[1])
                 commit = self.repository._repo.get_object(commit.object[1])
         except KeyError:
             raise RepositoryError("Cannot get object with id %s" % revision)
         self.raw_id = ascii_str(commit.id)
         self.short_id = self.raw_id[:12]
         self._commit = commit  # a Dulwich Commmit with .id
         self._tree_id = commit.tree
         self._committer_property = 'committer'
         self._author_property = 'author'
         self._date_property = 'commit_time'
         self._date_tz_property = 'commit_timezone'
         self.revision = repository.revisions.index(self.raw_id)
         self.nodes = {}
         self._paths = {}
     @LazyProperty
     def bookmarks(self):
         return ()
     @LazyProperty
     def message(self):
         return safe_str(self._commit.message)
     @LazyProperty
     def committer(self):
         return safe_str(getattr(self._commit, self._committer_property))
     @LazyProperty
     def author(self):
         return safe_str(getattr(self._commit, self._author_property))
     @LazyProperty
     def date(self):
         return date_fromtimestamp(getattr(self._commit, self._date_property),
                                   getattr(self._commit, self._date_tz_property))
     @LazyProperty
     def _timestamp(self):
         return getattr(self._commit, self._date_property)
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.changed, self.added, self.removed
     @LazyProperty
     def tags(self):
         _tags = []
         for tname, tsha in self.repository.tags.items():
             if tsha == self.raw_id:
                 _tags.append(tname)
         return _tags
     @LazyProperty
     def branch(self):
         # Note: This function will return one branch name for the changeset -
         # that might not make sense in Git where branches() is a better match
         # for the basic model
         heads = self.repository._heads(reverse=False)
         ref = heads.get(self._commit.id)
         if ref:
             return safe_str(ref)
     @LazyProperty
     def branches(self):
         heads = self.repository._heads(reverse=True)
         return [safe_str(b) for b in heads if heads[b] == self._commit.id] # FIXME: Inefficient ... and returning None!
     def _get_id_for_path(self, path):
         # FIXME: Please, spare a couple of minutes and make those codes cleaner;
         if path not in self._paths:
             path = path.strip('/')
             # set root tree
             tree = self.repository._repo[self._tree_id]
             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.items():
                 name = safe_str(item)
                 if curdir:
                     name = '/'.join((curdir, name))
                 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.items():
                     name = safe_str(item)
                     if dir == name:
                         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.items():
                     name = safe_str(item)
                     if curdir:
                         name = '/'.join((curdir, name))
                     self._paths[name] = id
                     self._stat_modes[name] = stat
             if path not in self._paths:
                 raise NodeDoesNotExistError("There is no file nor directory "
                     "at the given path '%s' at revision %s"
                     % (path, self.short_id))
         return self._paths[path]
     def _get_kind(self, path):
         obj = self.repository._repo[self._get_id_for_path(path)]
         if isinstance(obj, objects.Blob):
             return NodeKind.FILE
         elif isinstance(obj, objects.Tree):
             return NodeKind.DIR
     def _get_filectx(self, path):
         path = path.rstrip('/')
         if self._get_kind(path) != NodeKind.FILE:
             raise ChangesetError("File does not exist for revision %s at "
                 " '%s'" % (self.raw_id, path))
         return path
     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(ascii_str(parent_id))
                 for parent_id in self._commit.parents]
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         rev_filter = settings.GIT_REV_FILTER
         so = self.repository.run_git_command(
             ['rev-list', rev_filter, '--children']
+        )
         return [
             self.repository.get_changeset(cs)
             for parts in (l.split(' ') for l in so.splitlines())
             if parts[0] == self.raw_id
             for cs in parts[1:]
+        ]
     def next(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 next_ = cs.revision + 1
                 next_rev = cs.repository.revisions[next_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(next_rev)
             if not branch or branch == cs.branch:
                 return cs
     def prev(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 prev_ = cs.revision - 1
                 if prev_ < 0:
                     raise IndexError
                 prev_rev = cs.repository.revisions[prev_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(prev_rev)
             if not branch or branch == cs.branch:
                 return cs
     def diff(self, ignore_whitespace=True, context=3):
         # Only used to feed diffstat
         rev1 = self.parents[0] if self.parents else self.repository.EMPTY_CHANGESET
         rev2 = self
         return b''.join(self.repository.get_diff(rev1, rev2,
                                     ignore_whitespace=ignore_whitespace,
                                     context=context))
     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``.
         """
         return self.get_file_history(path, limit=1)[0]
     def get_file_history(self, path, limit=None):
         """
         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.
         """
         self._get_filectx(path)
         if limit is not None:
             cmd = ['log', '-n', str(safe_int(limit, 0)),
                    '--pretty=format:%H', '-s', self.raw_id, '--', path]
         else:
             cmd = ['log',
                    '--pretty=format:%H', '-s', self.raw_id, '--', path]
         so = self.repository.run_git_command(cmd)
         ids = re.findall(r'[0-9a-fA-F]{40}', so)
         return [self.repository.get_changeset(sha) for sha in ids]
     def get_file_history_2(self, path):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
         """
         self._get_filectx(path)
         from dulwich.walk import Walker
         include = [self.raw_id]
         walker = Walker(self.repository._repo.object_store, include,
                         paths=[path], max_entries=1)
         return [self.repository.get_changeset(ascii_str(x.commit.id.decode))
                 for x in walker]
     def get_file_annotate(self, path):
         """
         Returns a generator of four element tuples with
             lineno, sha, changeset lazy loader 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', self.raw_id, '--', path]
         # -l     ==> outputs long shas (and we need all 40 characters)
         # --root ==> doesn't put '^' character for boundaries
         # -r sha ==> blames for the given revision
         so = self.repository.run_git_command(cmd)
         for i, blame_line in enumerate(so.split('\n')[:-1]):
             sha, line = re.split(r' ', blame_line, 1)
             yield (i + 1, sha, lambda sha=sha: self.repository.get_changeset(sha), line)
     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
         if kind not in allowed_kinds:
             raise ImproperArchiveTypeError('Archive kind not supported use one'
                 'of %s' % ' '.join(allowed_kinds))
         if stream is None:
             raise VCSError('You need to pass in a valid stream for filling'
                            ' with archival data')
         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'
         _git_path = settings.GIT_EXECUTABLE_PATH
         cmd = '%s archive --format=%s --prefix=%s/ %s' % (_git_path,
                                                 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):
         """
         Returns combined ``DirNode`` and ``FileNode`` objects list representing
         state of changeset at the given ``path``. If node at the given ``path``
         is not instance of ``DirNode``, ChangesetError would be raised.
         """
         if self._get_kind(path) != NodeKind.DIR:
             raise ChangesetError("Directory does not exist for revision %s at "
                 " '%s'" % (self.revision, path))
         path = path.rstrip('/')
         id = self._get_id_for_path(path)
         tree = self.repository._repo[id]
         dirnodes = []
         filenodes = []
         als = self.repository.alias
         for name, stat, id in tree.items():
             obj_path = safe_str(name)
             if path != '':
                 obj_path = '/'.join((path, obj_path))
             if objects.S_ISGITLINK(stat):
                 root_tree = self.repository._repo[self._tree_id]
                 cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(root_tree[b'.gitmodules'][1]).data))
                 url = ascii_str(cf.get(('submodule', obj_path), 'url'))
                 dirnodes.append(SubModuleNode(obj_path, url=url, changeset=ascii_str(id),
                                               alias=als))
                 continue
             obj = self.repository._repo.get_object(id)
             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 node.path not in self.nodes:
                 self.nodes[node.path] = node
         nodes.sort()
         return nodes
     def get_node(self, path):
         """
         Returns ``Node`` object from the given ``path``. If there is no node at
         the given ``path``, ``ChangesetError`` would be raised.
         """
         path = path.rstrip('/')
         if path not 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)
             stat = self._stat_modes.get(path)
             if stat and objects.S_ISGITLINK(stat):
                 tree = self.repository._repo[self._tree_id]
                 cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(tree[b'.gitmodules'][1]).data))
                 url = ascii_str(cf.get(('submodule', path), 'url'))
                 node = SubModuleNode(path, url=url, changeset=ascii_str(id_),
                                      alias=self.repository.alias)
             else:
                 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: '%s' at revision %s"
                         % (path, self.short_id))
             # cache node
             self.nodes[path] = node
         return self.nodes[path]
     @LazyProperty
     def affected_files(self):
         """
         Gets a fast accessible file changes for given changeset
         """
         added, modified, deleted = self._changes_cache
         return list(added.union(modified).union(deleted))
     @LazyProperty
     def _changes_cache(self):
         added = set()
         modified = set()
         deleted = set()
         _r = self.repository._repo
         parents = self.parents
         if not self.parents:
             parents = [EmptyChangeset()]
         for parent in parents:
             if isinstance(parent, EmptyChangeset):
                 oid = None
             else:
                 oid = _r[parent._commit.id].tree
             changes = _r.object_store.tree_changes(oid, _r[self._commit.id].tree)
             for (oldpath, newpath), (_, _), (_, _) in changes:
                 if newpath and oldpath:
                     modified.add(safe_str(newpath))
                 elif newpath and not oldpath:
                     added.add(safe_str(newpath))
                 elif not newpath and oldpath:
                     deleted.add(safe_str(oldpath))
         return added, modified, deleted
     def _get_paths_for_status(self, status):
         """
         Returns sorted list of paths for given ``status``.
         :param status: one of: *added*, *modified* or *deleted*
         """
         added, modified, deleted = self._changes_cache
         return sorted({
             'added': list(added),
             'modified': list(modified),
             'deleted': list(deleted)}[status]
+        )
     @LazyProperty
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         if not self.parents:
             return list(self._get_file_nodes())
         return AddedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('added')], self)
     @LazyProperty
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return ChangedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('modified')], self)
     @LazyProperty
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return RemovedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('deleted')], self)
     extra = {}

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

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.backends.git.repository
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~
     Git repository implementation.
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import errno
 import logging
 import os
 import re
 import time
 import urllib.error
 import urllib.parse
 import urllib.request
 from collections import OrderedDict
 import mercurial.url  # import httpbasicauthhandler, httpdigestauthhandler
 import mercurial.util  # import url as hg_url
 from dulwich.config import ConfigFile
 from dulwich.objects import Tag
 from dulwich.repo import NotGitRepository, Repo
 from kallithea.lib.vcs import subprocessio
 from kallithea.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import (
     BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError, RepositoryError, TagAlreadyExistError, TagDoesNotExistError)
 from kallithea.lib.vcs.exceptions import (BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError, RepositoryError, TagAlreadyExistError,
                                           TagDoesNotExistError)
 from kallithea.lib.vcs.utils import ascii_str, date_fromtimestamp, makedate, safe_bytes, safe_str
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.lib.vcs.utils.paths import abspath, get_user_home
 from .changeset import GitChangeset
 from .inmemory import GitInMemoryChangeset
 from .workdir import GitWorkdir
 SHA_PATTERN = re.compile(r'^([0-9a-fA-F]{12}|[0-9a-fA-F]{40})$')
 log = logging.getLogger(__name__)
 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)
         self.bare = self.repo.bare
     @property
     def _config_files(self):
         return [
             self.bare and abspath(self.path, 'config')
                       or abspath(self.path, '.git', 'config'),
              abspath(get_user_home(), '.gitconfig'),
+         ]
     @property
     def _repo(self):
         return self.repo
     @property
     def head(self):
         try:
             return self._repo.head()
         except KeyError:
             return None
     @property
     def _empty(self):
         """
         Checks if repository is empty ie. without any changesets
         """
         try:
             self.revisions[0]
         except (KeyError, IndexError):
             return True
         return False
     @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()
     @classmethod
     def _run_git_command(cls, cmd, cwd=None):
         """
         Runs given ``cmd`` as git command and returns output bytes in a tuple
         (stdout, stderr) ... or raise RepositoryError.
         :param cmd: git command to be executed
         :param cwd: passed directly to subprocess
         """
         # need to clean fix GIT_DIR !
         gitenv = dict(os.environ)
         gitenv.pop('GIT_DIR', None)
         gitenv['GIT_CONFIG_NOGLOBAL'] = '1'
         assert isinstance(cmd, list), cmd
         cmd = [settings.GIT_EXECUTABLE_PATH, '-c', 'core.quotepath=false'] + cmd
         try:
             p = subprocessio.SubprocessIOChunker(cmd, cwd=cwd, env=gitenv, shell=False)
         except (EnvironmentError, OSError) as err:
             # output from the failing process is in str(EnvironmentError)
             msg = ("Couldn't run git command %s.\n"
                    "Subprocess failed with '%s': %s\n" %
                    (cmd, type(err).__name__, err)
             ).strip()
             log.error(msg)
             raise RepositoryError(msg)
         try:
             stdout = b''.join(p.output)
             stderr = b''.join(p.error)
         finally:
             p.close()
         # TODO: introduce option to make commands fail if they have any stderr output?
         if stderr:
             log.debug('stderr from %s:\n%s', cmd, stderr)
         else:
             log.debug('stderr from %s: None', cmd)
         return stdout, stderr
     def run_git_command(self, cmd):
         """
         Runs given ``cmd`` as git command with cwd set to current repo.
         Returns stdout as unicode str ... or raise RepositoryError.
         """
         cwd = None
         if os.path.isdir(self.path):
             cwd = self.path
         stdout, _stderr = self._run_git_command(cmd, cwd=cwd)
         return safe_str(stdout)
     @classmethod
     def _check_url(cls, url):
         """
         Function will check given url and try to verify if it's a valid
         link. Sometimes it may happened that git 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, exception is also thrown
         when the return code is non 200
         """
         # check first if it's not an local url
         if os.path.isdir(url) or url.startswith('file:'):
             return True
         if url.startswith('git://'):
             return True
         if '+' in url[:url.find('://')]:
             url = url[url.find('+') + 1:]
         handlers = []
         url_obj = mercurial.util.url(safe_bytes(url))
         test_uri, authinfo = url_obj.authinfo()
         if not test_uri.endswith(b'info/refs'):
             test_uri = test_uri.rstrip(b'/') + b'/info/refs'
         url_obj.passwd = b'*****'
         cleaned_uri = str(url_obj)
         if authinfo:
             # create a password manager
             passmgr = urllib.request.HTTPPasswordMgrWithDefaultRealm()
             passmgr.add_password(*authinfo)
             handlers.extend((mercurial.url.httpbasicauthhandler(passmgr),
                              mercurial.url.httpdigestauthhandler(passmgr)))
         o = urllib.request.build_opener(*handlers)
         o.addheaders = [('User-Agent', 'git/1.7.8.0')]  # fake some git
         req = urllib.request.Request(
             "%s?%s" % (
                 test_uri,
                 urllib.parse.urlencode({"service": 'git-upload-pack'})
             ))
         try:
             resp = o.open(req)
             if resp.code != 200:
                 raise Exception('Return Code is not 200')
         except Exception as e:
             # means it cannot be cloned
             raise urllib.error.URLError("[%s] org_exc: %s" % (cleaned_uri, e))
         # now detect if it's proper git repo
         gitdata = resp.read()
         if b'service=git-upload-pack' not in gitdata:
             raise urllib.error.URLError(
                 "url [%s] does not look like an git" % cleaned_uri)
         return True
     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:
                 GitRepository._check_url(src_url)
                 self.clone(src_url, update_after_clone, bare)
                 return Repo(self.path)
             elif create:
                 os.makedirs(self.path)
                 if bare:
                     return Repo.init_bare(self.path)
                 else:
                     return Repo.init(self.path)
             else:
                 return Repo(self.path)
         except (NotGitRepository, OSError) as err:
             raise RepositoryError(err)
     def _get_all_revisions(self):
         # we must check if this repo is not empty, since later command
         # fails if it is. And it's cheaper to ask than throw the subprocess
         # errors
         try:
             self._repo.head()
         except KeyError:
             return []
         rev_filter = settings.GIT_REV_FILTER
         cmd = ['rev-list', rev_filter, '--reverse', '--date-order']
         try:
             so = self.run_git_command(cmd)
         except RepositoryError:
             # Can be raised for empty repositories
             return []
         return so.splitlines()
     def _get_all_revisions2(self):
         # alternate implementation using dulwich
         includes = [ascii_str(sha) for key, (sha, type_) in self._parsed_refs.items()
                     if type_ != b'T']
         return [c.commit.id for c in self._repo.get_walker(include=includes)]
     def _get_revision(self, revision):
         """
         Given any revision identifier, returns a 40 char string with revision hash.
         """
         if self._empty:
             raise EmptyRepositoryError("There are no changesets yet")
         if revision in (None, '', 'tip', 'HEAD', 'head', -1):
             revision = -1
         if isinstance(revision, int):
             try:
                 return self.revisions[revision]
             except IndexError:
                 msg = "Revision %r does not exist for %s" % (revision, self.name)
                 raise ChangesetDoesNotExistError(msg)
         if isinstance(revision, str):
             if revision.isdigit() and (len(revision) < 12 or len(revision) == revision.count('0')):
                 try:
                     return self.revisions[int(revision)]
                 except IndexError:
                     msg = "Revision %r does not exist for %s" % (revision, self)
                     raise ChangesetDoesNotExistError(msg)
             # get by branch/tag name
             _ref_revision = self._parsed_refs.get(safe_bytes(revision))
             if _ref_revision:  # and _ref_revision[1] in [b'H', b'RH', b'T']:
                 return ascii_str(_ref_revision[0])
             if revision in self.revisions:
                 return revision
             # maybe it's a tag ? we don't have them in self.revisions
             if revision in self.tags.values():
                 return revision
             if SHA_PATTERN.match(revision):
                 msg = "Revision %r does not exist for %s" % (revision, self.name)
                 raise ChangesetDoesNotExistError(msg)
         raise ChangesetDoesNotExistError("Given revision %r not recognized" % revision)
     def get_ref_revision(self, ref_type, ref_name):
         """
         Returns ``GitChangeset`` object representing repository's
         changeset at the given ``revision``.
         """
         return self._get_revision(ref_name)
     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.
         """
         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:
             idx_loc = '' if self.bare else '.git'
             # fallback to filesystem
             in_path = os.path.join(self.path, idx_loc, "index")
             he_path = os.path.join(self.path, idx_loc, "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):
         return safe_str(self._repo.get_description() or b'unknown')
     @LazyProperty
     def contact(self):
         undefined_contact = 'Unknown'
         return undefined_contact
     @property
     def branches(self):
         if not self.revisions:
             return {}
         _branches = [(safe_str(key), ascii_str(sha))
                      for key, (sha, type_) in self._parsed_refs.items() if type_ == b'H']
         return OrderedDict(sorted(_branches, key=(lambda ctx: ctx[0]), reverse=False))
     @LazyProperty
     def closed_branches(self):
         return {}
     @LazyProperty
     def tags(self):
         return self._get_tags()
     def _get_tags(self):
         if not self.revisions:
             return {}
         _tags = [(safe_str(key), ascii_str(sha))
                  for key, (sha, type_) in self._parsed_refs.items() if type_ == b'T']
         return OrderedDict(sorted(_tags, key=(lambda ctx: ctx[0]), reverse=True))
     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 <joe.doe@example.com>"
         :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[b"refs/tags/%s" % safe_bytes(name)] = changeset._commit.id
         self._parsed_refs = self._get_parsed_refs()
         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 <joe.doe@example.com>"
         :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)
         # self._repo.refs is a DiskRefsContainer, and .path gives the full absolute path of '.git'
         tagpath = os.path.join(safe_str(self._repo.refs.path), 'refs', 'tags', name)
         try:
             os.remove(tagpath)
             self._parsed_refs = self._get_parsed_refs()
             self.tags = self._get_tags()
         except OSError as e:
             raise RepositoryError(e.strerror)
     @LazyProperty
     def bookmarks(self):
         """
         Gets bookmarks for this repository
         """
         return {}
     @LazyProperty
     def _parsed_refs(self):
         return self._get_parsed_refs()
     def _get_parsed_refs(self):
         """Return refs as a dict, like:
         { b'v0.2.0': [b'599ba911aa24d2981225f3966eb659dfae9e9f30', b'T'] }
         """
         _repo = self._repo
         refs = _repo.get_refs()
         keys = [(b'refs/heads/', b'H'),
                 (b'refs/remotes/origin/', b'RH'),
                 (b'refs/tags/', b'T')]
         _refs = {}
         for ref, sha in refs.items():
             for k, type_ in keys:
                 if ref.startswith(k):
                     _key = ref[len(k):]
                     if type_ == b'T':
                         obj = _repo.get_object(sha)
                         if isinstance(obj, Tag):
                             sha = _repo.get_object(sha).object[1]
                     _refs[_key] = [sha, type_]
                     break
         return _refs
     def _heads(self, reverse=False):
         refs = self._repo.get_refs()
         heads = {}
         for key, val in refs.items():
             for ref_key in [b'refs/heads/', b'refs/remotes/origin/']:
                 if key.startswith(ref_key):
                     n = key[len(ref_key):]
                     if n not in [b'HEAD']:
                         heads[n] = val
         return heads if reverse else dict((y, x) for x, y in heads.items())
     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
         return GitChangeset(repository=self, revision=self._get_revision(revision))
     def get_changesets(self, start=None, end=None, start_date=None,
            end_date=None, branch_name=None, reverse=False, max_revisions=None):
         """
         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)
         # actually we should check now if it's not an empty repo to not spaw
         # subprocess commands
         if self._empty:
             raise EmptyRepositoryError("There are no changesets yet")
         # %H at format means (full) commit hash, initial hashes are retrieved
         # in ascending date order
         cmd = ['log', '--date-order', '--reverse', '--pretty=format:%H']
         if max_revisions:
             cmd += ['--max-count=%s' % max_revisions]
         if start_date:
             cmd += ['--since', start_date.strftime('%m/%d/%y %H:%M:%S')]
         if end_date:
             cmd += ['--until', end_date.strftime('%m/%d/%y %H:%M:%S')]
         if branch_name:
             cmd.append(branch_name)
         else:
             cmd.append(settings.GIT_REV_FILTER)
         revs = self.run_git_command(cmd).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.reverse()
         return CollectionGenerator(self, revs)
     def get_diff(self, rev1, rev2, path=None, ignore_whitespace=False,
                  context=3):
         """
         Returns (git like) *diff*, as plain bytes 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``. Due to limitations in Git, if
           value passed-in is greater than ``2**31-1``
           (``2147483647``), it will be set to ``2147483647``
           instead. If negative value is passed-in, it will be set to
           ``0`` instead.
         """
         # Git internally uses a signed long int for storing context
         # size (number of lines to show before and after the
         # differences). This can result in integer overflow, so we
         # ensure the requested context is smaller by one than the
         # number that would cause the overflow. It is highly unlikely
         # that a single file will contain that many lines, so this
         # kind of change should not cause any realistic consequences.
         overflowed_long_int = 2**31
         if context >= overflowed_long_int:
             context = overflowed_long_int - 1
         # Negative context values make no sense, and will result in
         # errors. Ensure this does not happen.
         if context < 0:
             context = 0
         flags = ['-U%s' % context, '--full-index', '--binary', '-p', '-M', '--abbrev=40']
         if ignore_whitespace:
             flags.append('-w')
         if hasattr(rev1, 'raw_id'):
             rev1 = getattr(rev1, 'raw_id')
         if hasattr(rev2, 'raw_id'):
             rev2 = getattr(rev2, 'raw_id')
         if rev1 == self.EMPTY_CHANGESET:
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ['show'] + flags + [rev2]
         else:
             rev1 = self.get_changeset(rev1).raw_id
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ['diff'] + flags + [rev1, rev2]
         if path:
             cmd += ['--', path]
         stdout, stderr = self._run_git_command(cmd, cwd=self.path)
         # If we used 'show' command, strip first few lines (until actual diff
         # starts)
         if rev1 == self.EMPTY_CHANGESET:
             parts = stdout.split(b'\ndiff ', 1)
             if len(parts) > 1:
                 stdout = b'diff ' + parts[1]
         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', '-q']
         if bare:
             cmd.append('--bare')
         elif not update_after_clone:
             cmd.append('--no-checkout')
         cmd += ['--', url, self.path]
         # If error occurs run_git_command raises RepositoryError already
         self.run_git_command(cmd)
     def pull(self, url):
         """
         Tries to pull changes from external location.
         """
         url = self._get_url(url)
         cmd = ['pull', '--ff-only', url]
         # If error occurs run_git_command raises RepositoryError already
         self.run_git_command(cmd)
     def fetch(self, url):
         """
         Tries to pull changes from external location.
         """
         url = self._get_url(url)
         so = self.run_git_command(['ls-remote', '-h', url])
         cmd = ['fetch', url, '--']
         for line in (x for x in so.splitlines()):
             sha, ref = line.split('\t')
             cmd.append('+%s:%s' % (ref, ref))
         self.run_git_command(cmd)
     def _update_server_info(self):
         """
         runs gits update-server-info command in this repo instance
         """
         from dulwich.server import update_server_info
         try:
             update_server_info(self._repo)
         except OSError as e:
             if e.errno not in [errno.ENOENT, errno.EROFS]:
                 raise
             # Workaround for dulwich crashing on for example its own dulwich/tests/data/repos/simple_merge.git/info/refs.lock
             log.error('Ignoring %s running update-server-info: %s', type(e).__name__, e)
     @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, str):
             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:
                 value = config.get(section, name)
             except KeyError:
                 continue
             return None if value is None else safe_str(value)
         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)

kallithea/lib/vcs/backends/hg/changeset.py

➞

Show inline comments

 import os
 import posixpath
 import mercurial.archival
 import mercurial.node
 import mercurial.obsutil
 from kallithea.lib.vcs.backends.base import BaseChangeset
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import ChangesetDoesNotExistError, ChangesetError, ImproperArchiveTypeError, NodeDoesNotExistError, VCSError
 from kallithea.lib.vcs.nodes import (
     AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode, SubModuleNode)
 from kallithea.lib.vcs.nodes import (AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode,
                                      SubModuleNode)
 from kallithea.lib.vcs.utils import ascii_bytes, ascii_str, date_fromtimestamp, safe_bytes, safe_str
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.lib.vcs.utils.paths import get_dirs_for_path
 class MercurialChangeset(BaseChangeset):
     """
     Represents state of the repository at a revision.
     """
     def __init__(self, repository, revision):
         self.repository = repository
         assert isinstance(revision, str), repr(revision)
         self._ctx = repository._repo[ascii_bytes(revision)]
         self.raw_id = ascii_str(self._ctx.hex())
         self.revision = self._ctx._rev
         self.nodes = {}
     @LazyProperty
     def tags(self):
         return [safe_str(tag) for tag in self._ctx.tags()]
     @LazyProperty
     def branch(self):
         return safe_str(self._ctx.branch())
     @LazyProperty
     def branches(self):
         return [safe_str(self._ctx.branch())]
     @LazyProperty
     def closesbranch(self):
         return self._ctx.closesbranch()
     @LazyProperty
     def obsolete(self):
         return self._ctx.obsolete()
     @LazyProperty
     def bumped(self):
         return self._ctx.phasedivergent()
     @LazyProperty
     def divergent(self):
         return self._ctx.contentdivergent()
     @LazyProperty
     def extinct(self):
         return self._ctx.extinct()
     @LazyProperty
     def unstable(self):
         return self._ctx.orphan()
     @LazyProperty
     def phase(self):
         if(self._ctx.phase() == 1):
             return 'Draft'
         elif(self._ctx.phase() == 2):
             return 'Secret'
         else:
             return ''
     @LazyProperty
     def successors(self):
         successors = mercurial.obsutil.successorssets(self._ctx._repo, self._ctx.node(), closest=True)
         if successors:
             # flatten the list here handles both divergent (len > 1)
             # and the usual case (len = 1)
             successors = [mercurial.node.hex(n)[:12] for sub in successors for n in sub if n != self._ctx.node()]
         return successors
     @LazyProperty
     def predecessors(self):
         return [mercurial.node.hex(n)[:12] for n in mercurial.obsutil.closestpredecessors(self._ctx._repo, self._ctx.node())]
     @LazyProperty
     def bookmarks(self):
         return [safe_str(bookmark) for bookmark in self._ctx.bookmarks()]
     @LazyProperty
     def message(self):
         return safe_str(self._ctx.description())
     @LazyProperty
     def committer(self):
         return safe_str(self.author)
     @LazyProperty
     def author(self):
         return safe_str(self._ctx.user())
     @LazyProperty
     def date(self):
         return date_fromtimestamp(*self._ctx.date())
     @LazyProperty
     def _timestamp(self):
         return self._ctx.date()[0]
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.repository._repo.status(self._ctx.p1().node(),
                                             self._ctx.node())
     @LazyProperty
     def _file_paths(self):
         return list(safe_str(f) for f in self._ctx)
     @LazyProperty
     def _dir_paths(self):
         p = list(set(get_dirs_for_path(*self._file_paths)))
         p.insert(0, '')
         return p
     @LazyProperty
     def _paths(self):
         return self._dir_paths + self._file_paths
     @LazyProperty
     def short_id(self):
         return self.raw_id[:12]
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         return [self.repository.get_changeset(parent.rev())
                 for parent in self._ctx.parents() if parent.rev() >= 0]
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         return [self.repository.get_changeset(child.rev())
                 for child in self._ctx.children() if child.rev() >= 0]
     def next(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 next_ = cs.repository.revisions.index(cs.raw_id) + 1
                 next_rev = cs.repository.revisions[next_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(next_rev)
             if not branch or branch == cs.branch:
                 return cs
     def prev(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 prev_ = cs.repository.revisions.index(cs.raw_id) - 1
                 if prev_ < 0:
                     raise IndexError
                 prev_rev = cs.repository.revisions[prev_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(prev_rev)
             if not branch or branch == cs.branch:
                 return cs
     def diff(self):
         # Only used to feed diffstat
         return b''.join(self._ctx.diff())
     def _get_kind(self, path):
         path = path.rstrip('/')
         if path in self._file_paths:
             return NodeKind.FILE
         elif path in self._dir_paths:
             return NodeKind.DIR
         else:
             raise ChangesetError("Node does not exist at the given path '%s'"
                 % (path))
     def _get_filectx(self, path):
         path = path.rstrip('/')
         if self._get_kind(path) != NodeKind.FILE:
             raise ChangesetError("File does not exist for revision %s at "
                 " '%s'" % (self.raw_id, path))
         return self._ctx.filectx(safe_bytes(path))
     def _extract_submodules(self):
         """
         returns a dictionary with submodule information from substate file
         of hg repository
         """
         return self._ctx.substate
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         fctx = self._get_filectx(path)
         if b'x' in fctx.flags():
             return 0o100755
         else:
             return 0o100644
     def get_file_content(self, path):
         """
         Returns content of the file at given ``path``.
         """
         fctx = self._get_filectx(path)
         return fctx.data()
     def get_file_size(self, path):
         """
         Returns size of the file at given ``path``.
         """
         fctx = self._get_filectx(path)
         return fctx.size()
     def get_file_changeset(self, path):
         """
         Returns last commit of the file at the given ``path``.
         """
         return self.get_file_history(path, limit=1)[0]
     def get_file_history(self, path, limit=None):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
         """
         fctx = self._get_filectx(path)
         hist = []
         cnt = 0
         for cs in reversed([x for x in fctx.filelog()]):
             cnt += 1
             hist.append(mercurial.node.hex(fctx.filectx(cs).node()))
             if limit is not None and cnt == limit:
                 break
         return [self.repository.get_changeset(node) for node in hist]
     def get_file_annotate(self, path):
         """
         Returns a generator of four element tuples with
             lineno, sha, changeset lazy loader and line
         """
         annotations = self._get_filectx(path).annotate()
         annotation_lines = [(annotateline.fctx, annotateline.text) for annotateline in annotations]
         for i, (fctx, line) in enumerate(annotation_lines):
             sha = ascii_str(fctx.hex())
             yield (i + 1, sha, lambda sha=sha: self.repository.get_changeset(sha), line)
     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
         if kind not in allowed_kinds:
             raise ImproperArchiveTypeError('Archive kind not supported use one'
                 'of %s' % ' '.join(allowed_kinds))
         if stream is None:
             raise VCSError('You need to pass in a valid stream for filling'
                            ' with archival data')
         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")
         mercurial.archival.archive(self.repository._repo, stream, ascii_bytes(self.raw_id),
                          safe_bytes(kind), prefix=safe_bytes(prefix), subrepos=subrepos)
     def get_nodes(self, path):
         """
         Returns combined ``DirNode`` and ``FileNode`` objects list representing
         state of changeset at the given ``path``. If node at the given ``path``
         is not instance of ``DirNode``, ChangesetError would be raised.
         """
         if self._get_kind(path) != NodeKind.DIR:
             raise ChangesetError("Directory does not exist for revision %s at "
                 " '%s'" % (self.revision, path))
         path = path.rstrip('/')
         filenodes = [FileNode(f, changeset=self) for f in self._file_paths
             if os.path.dirname(f) == path]
         dirs = path == '' and '' or [d for d in self._dir_paths
             if d and posixpath.dirname(d) == path]
         dirnodes = [DirNode(d, changeset=self) for d in dirs
             if os.path.dirname(d) == path]
         als = self.repository.alias
         for k, vals in self._extract_submodules().items():
             #vals = url,rev,type
             loc = vals[0]
             cs = vals[1]
             dirnodes.append(SubModuleNode(k, url=loc, changeset=cs,
                                           alias=als))
         nodes = dirnodes + filenodes
         for node in nodes:
             self.nodes[node.path] = node
         nodes.sort()
         return nodes
     def get_node(self, path):
         """
         Returns ``Node`` object from the given ``path``. If there is no node at
         the given ``path``, ``ChangesetError`` would be raised.
         """
         path = path.rstrip('/')
         if path not in self.nodes:
             if path in self._file_paths:
                 node = FileNode(path, changeset=self)
             elif path in self._dir_paths or path in self._dir_paths:
                 if path == '':
                     node = RootNode(changeset=self)
                 else:
                     node = DirNode(path, changeset=self)
             else:
                 raise NodeDoesNotExistError("There is no file nor directory "
                     "at the given path: '%s' at revision %s"
                     % (path, self.short_id))
             # cache node
             self.nodes[path] = node
         return self.nodes[path]
     @LazyProperty
     def affected_files(self):
         """
         Gets a fast accessible file changes for given changeset
         """
         return self._ctx.files()
     @property
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         return AddedFileNodesGenerator([safe_str(n) for n in self.status.added], self)
     @property
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         return ChangedFileNodesGenerator([safe_str(n) for n in self.status.modified], self)
     @property
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         return RemovedFileNodesGenerator([safe_str(n) for n in self.status.removed], self)
     @LazyProperty
     def extra(self):
         return self._ctx.extra()

kallithea/lib/vcs/backends/hg/repository.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.backends.hg.repository
     ~~~~~~~~~~~~~~~~~~~~~~~~~~
     Mercurial repository implementation.
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import datetime
 import logging
 import os
 import time
 import urllib.error
 import urllib.parse
 import urllib.request
 from collections import OrderedDict
 import mercurial.commands
 import mercurial.error
 import mercurial.exchange
 import mercurial.hg
 import mercurial.hgweb
 import mercurial.httppeer
 import mercurial.localrepo
 import mercurial.match
 import mercurial.mdiff
 import mercurial.node
 import mercurial.patch
 import mercurial.scmutil
 import mercurial.sshpeer
 import mercurial.tags
 import mercurial.ui
 import mercurial.url
 import mercurial.util
 from kallithea.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from kallithea.lib.vcs.exceptions import (
     BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError, RepositoryError, TagAlreadyExistError, TagDoesNotExistError, VCSError)
 from kallithea.lib.vcs.exceptions import (BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError, RepositoryError, TagAlreadyExistError,
                                           TagDoesNotExistError, VCSError)
 from kallithea.lib.vcs.utils import ascii_str, author_email, author_name, date_fromtimestamp, makedate, safe_bytes, safe_str
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.lib.vcs.utils.paths import abspath
 from .changeset import MercurialChangeset
 from .inmemory import MercurialInMemoryChangeset
 from .workdir import MercurialWorkdir
 log = logging.getLogger(__name__)
 class MercurialRepository(BaseRepository):
     """
     Mercurial repository backend
     """
     DEFAULT_BRANCH_NAME = 'default'
     scm = 'hg'
     def __init__(self, repo_path, create=False, baseui=None, src_url=None,
                  update_after_clone=False):
         """
         Raises RepositoryError if repository could not be find at the given
         ``repo_path``.
         :param repo_path: local path of the repository
         :param create=False: if set to True, would try to create repository if
            it does not exist rather than raising exception
         :param baseui=None: user data
         :param src_url=None: would try to clone repository from given location
         :param update_after_clone=False: sets update of working copy after
           making a clone
         """
         if not isinstance(repo_path, str):
             raise VCSError('Mercurial backend requires repository path to '
                            'be instance of <str> got %s instead' %
                            type(repo_path))
         self.path = abspath(repo_path)
         self.baseui = baseui or mercurial.ui.ui()
         # We've set path and ui, now we can set _repo itself
         self._repo = self._get_repo(create, src_url, update_after_clone)
     @property
     def _empty(self):
         """
         Checks if repository is empty ie. without any changesets
         """
         # TODO: Following raises errors when using InMemoryChangeset...
         # return len(self._repo.changelog) == 0
         return len(self.revisions) == 0
     @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()
     @LazyProperty
     def name(self):
         return os.path.basename(self.path)
     @LazyProperty
     def branches(self):
         return self._get_branches()
     @LazyProperty
     def closed_branches(self):
         return self._get_branches(normal=False, closed=True)
     @LazyProperty
     def allbranches(self):
         """
         List all branches, including closed branches.
         """
         return self._get_branches(closed=True)
     def _get_branches(self, normal=True, closed=False):
         """
         Gets branches for this repository
         Returns only not closed branches by default
         :param closed: return also closed branches for mercurial
         :param normal: return also normal branches
         """
         if self._empty:
             return {}
         bt = OrderedDict()
         for bn, _heads, node, isclosed in sorted(self._repo.branchmap().iterbranches()):
             if isclosed:
                 if closed:
                     bt[safe_str(bn)] = ascii_str(mercurial.node.hex(node))
             else:
                 if normal:
                     bt[safe_str(bn)] = ascii_str(mercurial.node.hex(node))
         return bt
     @LazyProperty
     def tags(self):
         """
         Gets tags for this repository
         """
         return self._get_tags()
     def _get_tags(self):
         if self._empty:
             return {}
         return OrderedDict(sorted(
             ((safe_str(n), ascii_str(mercurial.node.hex(h))) for n, h in self._repo.tags().items()),
             reverse=True,
             key=lambda x: x[0],  # sort by name
         ))
     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 <joe.doe@example.com>"
         :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)
         local = kwargs.setdefault('local', False)
         if message is None:
             message = "Added tag %s for changeset %s" % (name,
                 changeset.short_id)
         if date is None:
             date = safe_bytes(datetime.datetime.now().strftime('%a, %d %b %Y %H:%M:%S'))
         try:
             mercurial.tags.tag(self._repo, safe_bytes(name), changeset._ctx.node(), safe_bytes(message), local, safe_bytes(user), date)
         except mercurial.error.Abort as e:
             raise RepositoryError(e.args[0])
         # Reinitialize tags
         self.tags = self._get_tags()
         tag_id = self.tags[name]
         return self.get_changeset(revision=tag_id)
     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 <joe.doe@example.com>"
         :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)
         if message is None:
             message = "Removed tag %s" % name
         if date is None:
             date = safe_bytes(datetime.datetime.now().strftime('%a, %d %b %Y %H:%M:%S'))
         local = False
         try:
             mercurial.tags.tag(self._repo, safe_bytes(name), mercurial.commands.nullid, safe_bytes(message), local, safe_bytes(user), date)
             self.tags = self._get_tags()
         except mercurial.error.Abort as e:
             raise RepositoryError(e.args[0])
     @LazyProperty
     def bookmarks(self):
         """
         Gets bookmarks for this repository
         """
         return self._get_bookmarks()
     def _get_bookmarks(self):
         if self._empty:
             return {}
         return OrderedDict(sorted(
             ((safe_str(n), ascii_str(h)) for n, h in self._repo._bookmarks.items()),
             reverse=True,
             key=lambda x: x[0],  # sort by name
         ))
     def _get_all_revisions(self):
         return [ascii_str(self._repo[x].hex()) for x in self._repo.filtered(b'visible').changelog.revs()]
     def get_diff(self, rev1, rev2, path='', 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``. If negative value is passed-in, it will be
           set to ``0`` instead.
         """
         # Negative context values make no sense, and will result in
         # errors. Ensure this does not happen.
         if context < 0:
             context = 0
         if hasattr(rev1, 'raw_id'):
             rev1 = getattr(rev1, 'raw_id')
         if hasattr(rev2, 'raw_id'):
             rev2 = getattr(rev2, 'raw_id')
         # Check if given revisions are present at repository (may raise
         # ChangesetDoesNotExistError)
         if rev1 != self.EMPTY_CHANGESET:
             self.get_changeset(rev1)
         self.get_changeset(rev2)
         if path:
             file_filter = mercurial.match.exact(path)
         else:
             file_filter = None
         return b''.join(mercurial.patch.diff(self._repo, rev1, rev2, match=file_filter,
                           opts=mercurial.mdiff.diffopts(git=True,
                                         showfunc=True,
                                         ignorews=ignore_whitespace,
                                         context=context)))
     @classmethod
     def _check_url(cls, url, repoui=None):
         """
         Function 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, exception is also thrown
         when the return code is non 200
         """
         # check first if it's not an local url
         url = safe_bytes(url)
         if os.path.isdir(url) or url.startswith(b'file:'):
             return True
         if url.startswith(b'ssh:'):
             # in case of invalid uri or authentication issues, sshpeer will
             # throw an exception.
             mercurial.sshpeer.instance(repoui or mercurial.ui.ui(), url, False).lookup(b'tip')
             return True
         url_prefix = None
         if b'+' in url[:url.find(b'://')]:
             url_prefix, url = url.split(b'+', 1)
         handlers = []
         url_obj = mercurial.util.url(url)
         test_uri, authinfo = url_obj.authinfo()
         url_obj.passwd = b'*****'
         cleaned_uri = str(url_obj)
         if authinfo:
             # create a password manager
             passmgr = urllib.request.HTTPPasswordMgrWithDefaultRealm()
             passmgr.add_password(*authinfo)
             handlers.extend((mercurial.url.httpbasicauthhandler(passmgr),
                              mercurial.url.httpdigestauthhandler(passmgr)))
         o = urllib.request.build_opener(*handlers)
         o.addheaders = [('Content-Type', 'application/mercurial-0.1'),
                         ('Accept', 'application/mercurial-0.1')]
         req = urllib.request.Request(
             "%s?%s" % (
                 test_uri,
                 urllib.parse.urlencode({
                     'cmd': 'between',
                     'pairs': "%s-%s" % ('0' * 40, '0' * 40),
                 })
             ))
         try:
             resp = o.open(req)
             if resp.code != 200:
                 raise Exception('Return Code is not 200')
         except Exception as e:
             # means it cannot be cloned
             raise urllib.error.URLError("[%s] org_exc: %s" % (cleaned_uri, e))
         if not url_prefix: # skip svn+http://... (and git+... too)
             # now check if it's a proper hg repo
             try:
                 mercurial.httppeer.instance(repoui or mercurial.ui.ui(), url, False).lookup(b'tip')
             except Exception as e:
                 raise urllib.error.URLError(
                     "url [%s] does not look like an hg repo org_exc: %s"
                     % (cleaned_uri, e))
         return True
     def _get_repo(self, create, src_url=None, update_after_clone=False):
         """
         Function will check for mercurial repository in given path and return
         a localrepo object. If there is no repository in that path it will
         raise an exception unless ``create`` parameter is set to True - in
         that case repository would be created and returned.
         If ``src_url`` is given, would try to clone repository from the
         location at given clone_point. Additionally it'll make update to
         working copy accordingly to ``update_after_clone`` flag
         """
         try:
             if src_url:
                 url = safe_bytes(self._get_url(src_url))
                 opts = {}
                 if not update_after_clone:
                     opts.update({'noupdate': True})
                 MercurialRepository._check_url(url, self.baseui)
                 mercurial.commands.clone(self.baseui, url, safe_bytes(self.path), **opts)
                 # Don't try to create if we've already cloned repo
                 create = False
             return mercurial.localrepo.instance(self.baseui, safe_bytes(self.path), create=create)
         except (mercurial.error.Abort, mercurial.error.RepoError) as err:
             if create:
                 msg = "Cannot create repository at %s. Original error was %s" \
                     % (self.name, err)
             else:
                 msg = "Not valid repository at %s. Original error was %s" \
                     % (self.name, err)
             raise RepositoryError(msg)
     @LazyProperty
     def in_memory_changeset(self):
         return MercurialInMemoryChangeset(self)
     @LazyProperty
     def description(self):
         _desc = self._repo.ui.config(b'web', b'description', None, untrusted=True)
         return safe_str(_desc or b'unknown')
     @LazyProperty
     def contact(self):
         return safe_str(mercurial.hgweb.common.get_contact(self._repo.ui.config)
                             or b'Unknown')
     @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
             cl_path = os.path.join(self.path, '.hg', "00changelog.i")
             st_path = os.path.join(self.path, '.hg', "store")
             if os.path.exists(cl_path):
                 return os.stat(cl_path).st_mtime
             else:
                 return os.stat(st_path).st_mtime
     def _get_revision(self, revision):
         """
         Given any revision identifier, returns a 40 char string with revision hash.
         :param revision: str or int or None
         """
         if self._empty:
             raise EmptyRepositoryError("There are no changesets yet")
         if revision in [-1, None]:
             revision = b'tip'
         elif isinstance(revision, str):
             revision = safe_bytes(revision)
         try:
             if isinstance(revision, int):
                 return ascii_str(self._repo[revision].hex())
             return ascii_str(mercurial.scmutil.revsymbol(self._repo, revision).hex())
         except (IndexError, ValueError, mercurial.error.RepoLookupError, TypeError):
             msg = "Revision %r does not exist for %s" % (safe_str(revision), self.name)
             raise ChangesetDoesNotExistError(msg)
         except (LookupError, ):
             msg = "Ambiguous identifier `%s` for %s" % (safe_str(revision), self.name)
             raise ChangesetDoesNotExistError(msg)
     def get_ref_revision(self, ref_type, ref_name):
         """
         Returns revision number for the given reference.
         """
         if ref_type == 'rev' and not ref_name.strip('0'):
             return self.EMPTY_CHANGESET
         # lookup up the exact node id
         _revset_predicates = {
                 'branch': 'branch',
                 'book': 'bookmark',
                 'tag': 'tag',
                 'rev': 'id',
+            }
         # avoid expensive branch(x) iteration over whole repo
         rev_spec = "%%s & %s(%%s)" % _revset_predicates[ref_type]
         try:
             revs = self._repo.revs(rev_spec, ref_name, ref_name)
         except LookupError:
             msg = "Ambiguous identifier %s:%s for %s" % (ref_type, ref_name, self.name)
             raise ChangesetDoesNotExistError(msg)
         except mercurial.error.RepoLookupError:
             msg = "Revision %s:%s does not exist for %s" % (ref_type, ref_name, self.name)
             raise ChangesetDoesNotExistError(msg)
         if revs:
             revision = revs.last()
         else:
             # TODO: just report 'not found'?
             revision = ref_name
         return self._get_revision(revision)
     def _get_archives(self, archive_name='tip'):
         allowed = self.baseui.configlist(b"web", b"allow_archive",
                                          untrusted=True)
         for name, ext in [(b'zip', '.zip'), (b'gz', '.tar.gz'), (b'bz2', '.tar.bz2')]:
             if name in allowed or self._repo.ui.configbool(b"web",
                                                            b"allow" + name,
                                                            untrusted=True):
                 yield {"type": safe_str(name), "extension": ext, "node": archive_name}
     def _get_url(self, url):
         """
         Returns normalized url. If schema is not given, fall back to
         filesystem (``file:///``) schema.
         """
         if url != 'default' and '://' not in url:
             url = "file:" + urllib.request.pathname2url(url)
         return url
     def get_changeset(self, revision=None):
         """
         Returns ``MercurialChangeset`` object representing repository's
         changeset at the given ``revision``.
         """
         return MercurialChangeset(repository=self, revision=self._get_revision(revision))
     def get_changesets(self, start=None, end=None, start_date=None,
                        end_date=None, branch_name=None, reverse=False, max_revisions=None):
         """
         Returns iterator of ``MercurialChangeset`` objects from start to end
         (both are inclusive)
         :param start: None, str, int or mercurial lookup format
         :param end:  None, str, int or mercurial lookup format
         :param start_date:
         :param end_date:
         :param branch_name:
         :param reversed: return changesets in reversed order
         """
         start_raw_id = self._get_revision(start)
         start_pos = None if start is None else self.revisions.index(start_raw_id)
         end_raw_id = self._get_revision(end)
         end_pos = None if end is None else self.revisions.index(end_raw_id)
         if start_pos is not None and end_pos is not None and start_pos > end_pos:
             raise RepositoryError("Start revision '%s' cannot be "
                                   "after end revision '%s'" % (start, end))
         if branch_name and branch_name not in self.allbranches:
             msg = "Branch %r not found in %s" % (branch_name, self.name)
             raise BranchDoesNotExistError(msg)
         if end_pos is not None:
             end_pos += 1
         # filter branches
         filter_ = []
         if branch_name:
             filter_.append(b'branch("%s")' % safe_bytes(branch_name))
         if start_date:
             filter_.append(b'date(">%s")' % safe_bytes(str(start_date)))
         if end_date:
             filter_.append(b'date("<%s")' % safe_bytes(str(end_date)))
         if filter_ or max_revisions:
             if filter_:
                 revspec = b' and '.join(filter_)
             else:
                 revspec = b'all()'
             if max_revisions:
                 revspec = b'limit(%s, %d)' % (revspec, max_revisions)
             revisions = mercurial.scmutil.revrange(self._repo, [revspec])
         else:
             revisions = self.revisions
         # this is very much a hack to turn this into a list; a better solution
         # would be to get rid of this function entirely and use revsets
         revs = list(revisions)[start_pos:end_pos]
         if reverse:
             revs.reverse()
         return CollectionGenerator(self, revs)
     def pull(self, url):
         """
         Tries to pull changes from external location.
         """
         other = mercurial.hg.peer(self._repo, {}, safe_bytes(self._get_url(url)))
         try:
             mercurial.exchange.pull(self._repo, other, heads=None, force=None)
         except mercurial.error.Abort as err:
             # Propagate error but with vcs's type
             raise RepositoryError(str(err))
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         return MercurialWorkdir(self)
     def get_config_value(self, section, name=None, 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, str):
             config_file = [config_file]
         config = self._repo.ui
         if config_file:
             config = mercurial.ui.ui()
             for path in config_file:
                 config.readconfig(safe_bytes(path))
         value = config.config(safe_bytes(section), safe_bytes(name))
         return value if value is None else safe_str(value)
     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)
         """
         username = self.get_config_value('ui', 'username', config_file=config_file)
         if username:
             return author_name(username)
         return None
     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)
         """
         username = self.get_config_value('ui', 'username', config_file=config_file)
         if username:
             return author_email(username)
         return None

kallithea/model/db.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.model.db
 ~~~~~~~~~~~~~~~~~~
 Database Models for Kallithea
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Apr 08, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import base64
 import collections
 import datetime
 import functools
 import hashlib
 import logging
 import os
 import time
 import traceback
 import ipaddr
 import sqlalchemy
 from beaker.cache import cache_region, region_invalidate
 from sqlalchemy import Boolean, Column, DateTime, Float, ForeignKey, Index, Integer, LargeBinary, String, Unicode, UnicodeText, UniqueConstraint
 from sqlalchemy.ext.hybrid import hybrid_property
 from sqlalchemy.orm import class_mapper, joinedload, relationship, validates
 from tg.i18n import lazy_ugettext as _
 from webob.exc import HTTPNotFound
 import kallithea
 from kallithea.lib import ext_json
 from kallithea.lib.caching_query import FromCache
 from kallithea.lib.exceptions import DefaultUserException
 from kallithea.lib.utils2 import (
     Optional, ascii_bytes, aslist, get_changeset_safe, get_clone_url, remove_prefix, safe_bytes, safe_int, safe_str, str2bool, urlreadable)
 from kallithea.lib.utils2 import (Optional, ascii_bytes, aslist, get_changeset_safe, get_clone_url, remove_prefix, safe_bytes, safe_int, safe_str, str2bool,
                                   urlreadable)
 from kallithea.lib.vcs import get_backend
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.vcs.utils.helpers import get_scm
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.model.meta import Base, Session
 URL_SEP = '/'
 log = logging.getLogger(__name__)
 #==============================================================================
 # BASE CLASSES
 #==============================================================================
 def _hash_key(k):
     return hashlib.md5(safe_bytes(k)).hexdigest()
 class BaseDbModel(object):
     """
     Base Model for all classes
     """
     @classmethod
     def _get_keys(cls):
         """return column names for this model """
         # Note: not a normal dict - iterator gives "users.firstname", but keys gives "firstname"
         return class_mapper(cls).c.keys()
     def get_dict(self):
         """
         return dict with keys and values corresponding
         to this model data """
         d = {}
         for k in self._get_keys():
             d[k] = getattr(self, k)
         # also use __json__() if present to get additional fields
         _json_attr = getattr(self, '__json__', None)
         if _json_attr:
             # update with attributes from __json__
             if callable(_json_attr):
                 _json_attr = _json_attr()
             for k, val in _json_attr.items():
                 d[k] = val
         return d
     def get_appstruct(self):
         """return list with keys and values tuples corresponding
         to this model data """
         return [
             (k, getattr(self, k))
             for k in self._get_keys()
+        ]
     def populate_obj(self, populate_dict):
         """populate model with data from given populate_dict"""
         for k in self._get_keys():
             if k in populate_dict:
                 setattr(self, k, populate_dict[k])
     @classmethod
     def query(cls):
         return Session().query(cls)
     @classmethod
     def get(cls, id_):
         if id_:
             return cls.query().get(id_)
     @classmethod
     def guess_instance(cls, value, callback=None):
         """Haphazardly attempt to convert `value` to a `cls` instance.
         If `value` is None or already a `cls` instance, return it. If `value`
         is a number (or looks like one if you squint just right), assume it's
         a database primary key and let SQLAlchemy sort things out. Otherwise,
         fall back to resolving it using `callback` (if specified); this could
         e.g. be a function that looks up instances by name (though that won't
         work if the name begins with a digit). Otherwise, raise Exception.
         """
         if value is None:
             return None
         if isinstance(value, cls):
             return value
         if isinstance(value, int):
             return cls.get(value)
         if isinstance(value, str) and value.isdigit():
             return cls.get(int(value))
         if callback is not None:
             return callback(value)
         raise Exception(
             'given object must be int, long or Instance of %s '
             'got %s, no callback provided' % (cls, type(value))
+        )
     @classmethod
     def get_or_404(cls, id_):
         try:
             id_ = int(id_)
         except (TypeError, ValueError):
             raise HTTPNotFound
         res = cls.query().get(id_)
         if res is None:
             raise HTTPNotFound
         return res
     @classmethod
     def delete(cls, id_):
         obj = cls.query().get(id_)
         Session().delete(obj)
     def __repr__(self):
         return '<DB:%s>' % (self.__class__.__name__)
 _table_args_default_dict = {'extend_existing': True,
                             'mysql_engine': 'InnoDB',
                             'mysql_charset': 'utf8',
                             'sqlite_autoincrement': True,
+                           }
 class Setting(Base, BaseDbModel):
     __tablename__ = 'settings'
     __table_args__ = (
         _table_args_default_dict,
+    )
     SETTINGS_TYPES = {
         'str': safe_bytes,
         'int': safe_int,
         'unicode': safe_str,
         'bool': str2bool,
         'list': functools.partial(aslist, sep=',')
+    }
     DEFAULT_UPDATE_URL = ''
     app_settings_id = Column(Integer(), primary_key=True)
     app_settings_name = Column(String(255), nullable=False, unique=True)
     _app_settings_value = Column("app_settings_value", Unicode(4096), nullable=False)
     _app_settings_type = Column("app_settings_type", String(255), nullable=True) # FIXME: not nullable?
     def __init__(self, key='', val='', type='unicode'):
         self.app_settings_name = key
         self.app_settings_value = val
         self.app_settings_type = type
     @validates('_app_settings_value')
     def validate_settings_value(self, key, val):
         assert isinstance(val, str)
         return val
     @hybrid_property
     def app_settings_value(self):
         v = self._app_settings_value
         _type = self.app_settings_type
         converter = self.SETTINGS_TYPES.get(_type) or self.SETTINGS_TYPES['unicode']
         return converter(v)
     @app_settings_value.setter
     def app_settings_value(self, val):
         """
         Setter that will always make sure we use str in app_settings_value
         """
         self._app_settings_value = safe_str(val)
     @hybrid_property
     def app_settings_type(self):
         return self._app_settings_type
     @app_settings_type.setter
     def app_settings_type(self, val):
         if val not in self.SETTINGS_TYPES:
             raise Exception('type must be one of %s got %s'
                             % (list(self.SETTINGS_TYPES), val))
         self._app_settings_type = val
     def __repr__(self):
         return "<%s %s.%s=%r>" % (
             self.__class__.__name__,
             self.app_settings_name, self.app_settings_type, self.app_settings_value
+        )
     @classmethod
     def get_by_name(cls, key):
         return cls.query() \
             .filter(cls.app_settings_name == key).scalar()
     @classmethod
     def get_by_name_or_create(cls, key, val='', type='unicode'):
         res = cls.get_by_name(key)
         if res is None:
             res = cls(key, val, type)
         return res
     @classmethod
     def create_or_update(cls, key, val=Optional(''), type=Optional('unicode')):
         """
         Creates or updates Kallithea setting. If updates are triggered, it will only
         update parameters that are explicitly set. Optional instance will be skipped.
         :param key:
         :param val:
         :param type:
         :return:
         """
         res = cls.get_by_name(key)
         if res is None:
             val = Optional.extract(val)
             type = Optional.extract(type)
             res = cls(key, val, type)
             Session().add(res)
         else:
             res.app_settings_name = key
             if not isinstance(val, Optional):
                 # update if set
                 res.app_settings_value = val
             if not isinstance(type, Optional):
                 # update if set
                 res.app_settings_type = type
         return res
     @classmethod
     def get_app_settings(cls, cache=False):
         ret = cls.query()
         if cache:
             ret = ret.options(FromCache("sql_cache_short", "get_hg_settings"))
         if ret is None:
             raise Exception('Could not get application settings !')
         settings = {}
         for each in ret:
             settings[each.app_settings_name] = \
                 each.app_settings_value
         return settings
     @classmethod
     def get_auth_settings(cls, cache=False):
         ret = cls.query() \
                 .filter(cls.app_settings_name.startswith('auth_')).all()
         fd = {}
         for row in ret:
             fd[row.app_settings_name] = row.app_settings_value
         return fd
     @classmethod
     def get_default_repo_settings(cls, cache=False, strip_prefix=False):
         ret = cls.query() \
                 .filter(cls.app_settings_name.startswith('default_')).all()
         fd = {}
         for row in ret:
             key = row.app_settings_name
             if strip_prefix:
                 key = remove_prefix(key, prefix='default_')
             fd.update({key: row.app_settings_value})
         return fd
     @classmethod
     def get_server_info(cls):
         import pkg_resources
         import platform
         from kallithea.lib.utils import check_git_version
         mods = [(p.project_name, p.version) for p in pkg_resources.working_set]
         info = {
             'modules': sorted(mods, key=lambda k: k[0].lower()),
             'py_version': platform.python_version(),
             'platform': platform.platform(),
             'kallithea_version': kallithea.__version__,
             'git_version': str(check_git_version()),
             'git_path': kallithea.CONFIG.get('git_path')
+        }
         return info
 class Ui(Base, BaseDbModel):
     __tablename__ = 'ui'
     __table_args__ = (
         # FIXME: ui_key as key is wrong and should be removed when the corresponding
         # Ui.get_by_key has been replaced by the composite key
         UniqueConstraint('ui_key'),
         UniqueConstraint('ui_section', 'ui_key'),
         _table_args_default_dict,
+    )
     HOOK_UPDATE = 'changegroup.update'
     HOOK_REPO_SIZE = 'changegroup.repo_size'
     ui_id = Column(Integer(), primary_key=True)
     ui_section = Column(String(255), nullable=False)
     ui_key = Column(String(255), nullable=False)
     ui_value = Column(String(255), nullable=True) # FIXME: not nullable?
     ui_active = Column(Boolean(), nullable=False, default=True)
     @classmethod
     def get_by_key(cls, section, key):
         """ Return specified Ui object, or None if not found. """
         return cls.query().filter_by(ui_section=section, ui_key=key).scalar()
     @classmethod
     def get_or_create(cls, section, key):
         """ Return specified Ui object, creating it if necessary. """
         setting = cls.get_by_key(section, key)
         if setting is None:
             setting = cls(ui_section=section, ui_key=key)
             Session().add(setting)
         return setting
     @classmethod
     def get_builtin_hooks(cls):
         q = cls.query()
         q = q.filter(cls.ui_key.in_([cls.HOOK_UPDATE, cls.HOOK_REPO_SIZE]))
         q = q.filter(cls.ui_section == 'hooks')
         return q.all()
     @classmethod
     def get_custom_hooks(cls):
         q = cls.query()
         q = q.filter(~cls.ui_key.in_([cls.HOOK_UPDATE, cls.HOOK_REPO_SIZE]))
         q = q.filter(cls.ui_section == 'hooks')
         return q.all()
     @classmethod
     def get_repos_location(cls):
         return cls.get_by_key('paths', '/').ui_value
     @classmethod
     def create_or_update_hook(cls, key, val):
         new_ui = cls.get_or_create('hooks', key)
         new_ui.ui_active = True
         new_ui.ui_value = val
     def __repr__(self):
         return '<%s %s.%s=%r>' % (
             self.__class__.__name__,
             self.ui_section, self.ui_key, self.ui_value)
 class User(Base, BaseDbModel):
     __tablename__ = 'users'
     __table_args__ = (
         Index('u_username_idx', 'username'),
         Index('u_email_idx', 'email'),
         _table_args_default_dict,
+    )
     DEFAULT_USER = 'default'
     DEFAULT_GRAVATAR_URL = 'https://secure.gravatar.com/avatar/{md5email}?d=identicon&s={size}'
     # The name of the default auth type in extern_type, 'internal' lives in auth_internal.py
     DEFAULT_AUTH_TYPE = 'internal'
     user_id = Column(Integer(), primary_key=True)
     username = Column(String(255), nullable=False, unique=True)
     password = Column(String(255), nullable=False)
     active = Column(Boolean(), nullable=False, default=True)
     admin = Column(Boolean(), nullable=False, default=False)
     name = Column("firstname", Unicode(255), nullable=False)
     lastname = Column(Unicode(255), nullable=False)
     _email = Column("email", String(255), nullable=True, unique=True) # FIXME: not nullable?
     last_login = Column(DateTime(timezone=False), nullable=True)
     extern_type = Column(String(255), nullable=True) # FIXME: not nullable?
     extern_name = Column(String(255), nullable=True) # FIXME: not nullable?
     api_key = Column(String(255), nullable=False)
     created_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     _user_data = Column("user_data", LargeBinary(), nullable=True)  # JSON data # FIXME: not nullable?
     user_log = relationship('UserLog')
     user_perms = relationship('UserToPerm', primaryjoin="User.user_id==UserToPerm.user_id", cascade='all')
     repositories = relationship('Repository')
     repo_groups = relationship('RepoGroup')
     user_groups = relationship('UserGroup')
     user_followers = relationship('UserFollowing', primaryjoin='UserFollowing.follows_user_id==User.user_id', cascade='all')
     followings = relationship('UserFollowing', primaryjoin='UserFollowing.user_id==User.user_id', cascade='all')
     repo_to_perm = relationship('UserRepoToPerm', primaryjoin='UserRepoToPerm.user_id==User.user_id', cascade='all')
     repo_group_to_perm = relationship('UserRepoGroupToPerm', primaryjoin='UserRepoGroupToPerm.user_id==User.user_id', cascade='all')
     group_member = relationship('UserGroupMember', cascade='all')
     # comments created by this user
     user_comments = relationship('ChangesetComment', cascade='all')
     # extra emails for this user
     user_emails = relationship('UserEmailMap', cascade='all')
     # extra API keys
     user_api_keys = relationship('UserApiKeys', cascade='all')
     ssh_keys = relationship('UserSshKeys', cascade='all')
     @hybrid_property
     def email(self):
         return self._email
     @email.setter
     def email(self, val):
         self._email = val.lower() if val else None
     @property
     def firstname(self):
         # alias for future
         return self.name
     @property
     def emails(self):
         other = UserEmailMap.query().filter(UserEmailMap.user == self).all()
         return [self.email] + [x.email for x in other]
     @property
     def api_keys(self):
         other = UserApiKeys.query().filter(UserApiKeys.user == self).all()
         return [self.api_key] + [x.api_key for x in other]
     @property
     def ip_addresses(self):
         ret = UserIpMap.query().filter(UserIpMap.user == self).all()
         return [x.ip_addr for x in ret]
     @property
     def full_name(self):
         return '%s %s' % (self.firstname, self.lastname)
     @property
     def full_name_or_username(self):
         """
         Show full name.
         If full name is not set, fall back to username.
         """
         return ('%s %s' % (self.firstname, self.lastname)
                 if (self.firstname and self.lastname) else self.username)
     @property
     def full_name_and_username(self):
         """
         Show full name and username as 'Firstname Lastname (username)'.
         If full name is not set, fall back to username.
         """
         return ('%s %s (%s)' % (self.firstname, self.lastname, self.username)
                 if (self.firstname and self.lastname) else self.username)
     @property
     def full_contact(self):
         return '%s %s <%s>' % (self.firstname, self.lastname, self.email)
     @property
     def short_contact(self):
         return '%s %s' % (self.firstname, self.lastname)
     @property
     def is_admin(self):
         return self.admin
     @hybrid_property
     def is_default_user(self):
         return self.username == User.DEFAULT_USER
     @hybrid_property
     def user_data(self):
         if not self._user_data:
             return {}
         try:
             return ext_json.loads(self._user_data)
         except TypeError:
             return {}
     @user_data.setter
     def user_data(self, val):
         try:
             self._user_data = ascii_bytes(ext_json.dumps(val))
         except Exception:
             log.error(traceback.format_exc())
     def __repr__(self):
         return "<%s %s: %r')>" % (self.__class__.__name__, self.user_id, self.username)
     @classmethod
     def guess_instance(cls, value):
         return super(User, cls).guess_instance(value, User.get_by_username)
     @classmethod
     def get_or_404(cls, id_, allow_default=True):
         '''
         Overridden version of BaseDbModel.get_or_404, with an extra check on
         the default user.
         '''
         user = super(User, cls).get_or_404(id_)
         if not allow_default and user.is_default_user:
             raise DefaultUserException()
         return user
     @classmethod
     def get_by_username_or_email(cls, username_or_email, case_insensitive=False, cache=False):
         """
         For anything that looks like an email address, look up by the email address (matching
         case insensitively).
         For anything else, try to look up by the user name.
         This assumes no normal username can have '@' symbol.
         """
         if '@' in username_or_email:
             return User.get_by_email(username_or_email, cache=cache)
         else:
             return User.get_by_username(username_or_email, case_insensitive=case_insensitive, cache=cache)
     @classmethod
     def get_by_username(cls, username, case_insensitive=False, cache=False):
         if case_insensitive:
             q = cls.query().filter(sqlalchemy.func.lower(cls.username) == sqlalchemy.func.lower(username))
         else:
             q = cls.query().filter(cls.username == username)
         if cache:
             q = q.options(FromCache(
                             "sql_cache_short",
                             "get_user_%s" % _hash_key(username)
+                          )
+            )
         return q.scalar()
     @classmethod
     def get_by_api_key(cls, api_key, cache=False, fallback=True):
         if len(api_key) != 40 or not api_key.isalnum():
             return None
         q = cls.query().filter(cls.api_key == api_key)
         if cache:
             q = q.options(FromCache("sql_cache_short",
                                     "get_api_key_%s" % api_key))
         res = q.scalar()
         if fallback and not res:
             # fallback to additional keys
             _res = UserApiKeys.query().filter_by(api_key=api_key, is_expired=False).first()
             if _res:
                 res = _res.user
         if res is None or not res.active or res.is_default_user:
             return None
         return res
     @classmethod
     def get_by_email(cls, email, cache=False):
         q = cls.query().filter(sqlalchemy.func.lower(cls.email) == sqlalchemy.func.lower(email))
         if cache:
             q = q.options(FromCache("sql_cache_short",
                                     "get_email_key_%s" % email))
         ret = q.scalar()
         if ret is None:
             q = UserEmailMap.query()
             # try fetching in alternate email map
             q = q.filter(sqlalchemy.func.lower(UserEmailMap.email) == sqlalchemy.func.lower(email))
             q = q.options(joinedload(UserEmailMap.user))
             if cache:
                 q = q.options(FromCache("sql_cache_short",
                                         "get_email_map_key_%s" % email))
             ret = getattr(q.scalar(), 'user', None)
         return ret
     @classmethod
     def get_from_cs_author(cls, author):
         """
         Tries to get User objects out of commit author string
         :param author:
         """
         from kallithea.lib.helpers import email, author_name
         # Valid email in the attribute passed, see if they're in the system
         _email = email(author)
         if _email:
             user = cls.get_by_email(_email)
             if user is not None:
                 return user
         # Maybe we can match by username?
         _author = author_name(author)
         user = cls.get_by_username(_author, case_insensitive=True)
         if user is not None:
             return user
     def update_lastlogin(self):
         """Update user lastlogin"""
         self.last_login = datetime.datetime.now()
         log.debug('updated user %s lastlogin', self.username)
     @classmethod
     def get_first_admin(cls):
         user = User.query().filter(User.admin == True).first()
         if user is None:
             raise Exception('Missing administrative account!')
         return user
     @classmethod
     def get_default_user(cls, cache=False):
         user = User.get_by_username(User.DEFAULT_USER, cache=cache)
         if user is None:
             raise Exception('Missing default account!')
         return user
     def get_api_data(self, details=False):
         """
         Common function for generating user related data for API
         """
         user = self
         data = dict(
             user_id=user.user_id,
             username=user.username,
             firstname=user.name,
             lastname=user.lastname,
             email=user.email,
             emails=user.emails,
             active=user.active,
             admin=user.admin,
+        )
         if details:
             data.update(dict(
                 extern_type=user.extern_type,
                 extern_name=user.extern_name,
                 api_key=user.api_key,
                 api_keys=user.api_keys,
                 last_login=user.last_login,
                 ip_addresses=user.ip_addresses
                 ))
         return data
     def __json__(self):
         data = dict(
             full_name=self.full_name,
             full_name_or_username=self.full_name_or_username,
             short_contact=self.short_contact,
             full_contact=self.full_contact
+        )
         data.update(self.get_api_data())
         return data
 class UserApiKeys(Base, BaseDbModel):
     __tablename__ = 'user_api_keys'
     __table_args__ = (
         Index('uak_api_key_idx', 'api_key'),
         Index('uak_api_key_expires_idx', 'api_key', 'expires'),
         _table_args_default_dict,
+    )
     user_api_key_id = Column(Integer(), primary_key=True)
     user_id = Column(Integer(), ForeignKey('users.user_id'), nullable=False)
     api_key = Column(String(255), nullable=False, unique=True)
     description = Column(UnicodeText(), nullable=False)
     expires = Column(Float(53), nullable=False)
     created_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     user = relationship('User')
     @hybrid_property
     def is_expired(self):
         return (self.expires != -1) & (time.time() > self.expires)
 class UserEmailMap(Base, BaseDbModel):
     __tablename__ = 'user_email_map'
     __table_args__ = (
         Index('uem_email_idx', 'email'),
         _table_args_default_dict,
+    )
     email_id = Column(Integer(), primary_key=True)
     user_id = Column(Integer(), ForeignKey('users.user_id'), nullable=False)
     _email = Column("email", String(255), nullable=False, unique=True)
     user = relationship('User')
     @validates('_email')
     def validate_email(self, key, email):
         # check if this email is not main one
         main_email = Session().query(User).filter(User.email == email).scalar()
         if main_email is not None:
             raise AttributeError('email %s is present is user table' % email)
         return email
     @hybrid_property
     def email(self):
         return self._email
     @email.setter
     def email(self, val):
         self._email = val.lower() if val else None
 class UserIpMap(Base, BaseDbModel):
     __tablename__ = 'user_ip_map'
     __table_args__ = (
         UniqueConstraint('user_id', 'ip_addr'),
         _table_args_default_dict,
+    )
     ip_id = Column(Integer(), primary_key=True)
     user_id = Column(Integer(), ForeignKey('users.user_id'), nullable=False)
     ip_addr = Column(String(255), nullable=False)
     active = Column(Boolean(), nullable=False, default=True)
     user = relationship('User')
     @classmethod
     def _get_ip_range(cls, ip_addr):
         net = ipaddr.IPNetwork(address=ip_addr)
         return [str(net.network), str(net.broadcast)]
     def __json__(self):
         return dict(
           ip_addr=self.ip_addr,
           ip_range=self._get_ip_range(self.ip_addr)
+        )
     def __repr__(self):
         return "<%s %s: %s>" % (self.__class__.__name__, self.user_id, self.ip_addr)
 class UserLog(Base, BaseDbModel):
     __tablename__ = 'user_logs'
     __table_args__ = (
         _table_args_default_dict,
+    )
     user_log_id = Column(Integer(), primary_key=True)
     user_id = Column(Integer(), ForeignKey('users.user_id'), nullable=True)
     username = Column(String(255), nullable=False)
     repository_id = Column(Integer(), ForeignKey('repositories.repo_id'), nullable=True)
     repository_name = Column(Unicode(255), nullable=False)
     user_ip = Column(String(255), nullable=True)
     action = Column(UnicodeText(), nullable=False)
     action_date = Column(DateTime(timezone=False), nullable=False)
     def __repr__(self):
         return "<%s %r: %r')>" % (self.__class__.__name__,
                                   self.repository_name,
                                   self.action)
     @property
     def action_as_day(self):
         return datetime.date(*self.action_date.timetuple()[:3])
     user = relationship('User')
     repository = relationship('Repository', cascade='')
 class UserGroup(Base, BaseDbModel):
     __tablename__ = 'users_groups'
     __table_args__ = (
         _table_args_default_dict,
+    )
     users_group_id = Column(Integer(), primary_key=True)
     users_group_name = Column(Unicode(255), nullable=False, unique=True)
     user_group_description = Column(Unicode(10000), nullable=True) # FIXME: not nullable?
     users_group_active = Column(Boolean(), nullable=False)
     owner_id = Column('user_id', Integer(), ForeignKey('users.user_id'), nullable=False)
     created_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     _group_data = Column("group_data", LargeBinary(), nullable=True)  # JSON data # FIXME: not nullable?
     members = relationship('UserGroupMember', cascade="all, delete-orphan")
     users_group_to_perm = relationship('UserGroupToPerm', cascade='all')

Changeset was too big and was cut off... Show full diff anyway

0 comments (0 inline, 0 general)