kallithea Changeset - f9988201a3c4

Changeset - f9988201a3c4

Parent rev.

Child rev.

[Not reviewed]

default

0 6 0

Mads Kiilerich - 6 years ago 2019-12-21 15:14:08
mads@kiilerich.com

Grafted from: ffca1325171a

py3: some not-entirely-trivial removing of "unicode"

6 files changed with 10 insertions and 12 deletions:

kallithea/controllers/admin/admin.py

kallithea/controllers/search.py

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

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

kallithea/model/db.py

scripts/update-copyrights.py

0 comments (0 inline, 0 general)

kallithea/controllers/admin/admin.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.admin.admin
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Controller for Admin panel of 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 7, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 from sqlalchemy.orm import joinedload
 from sqlalchemy.sql.expression import and_, func, or_
 from tg import request
 from tg import tmpl_context as c
 from whoosh import query
 from whoosh.qparser.dateparse import DateParserPlugin
 from whoosh.qparser.default import QueryParser
 from kallithea.lib.auth import HasPermissionAnyDecorator, LoginRequired
 from kallithea.lib.base import BaseController, render
 from kallithea.lib.indexers import JOURNAL_SCHEMA
 from kallithea.lib.page import Page
 from kallithea.lib.utils2 import remove_prefix, remove_suffix, safe_int
 from kallithea.model.db import UserLog
 log = logging.getLogger(__name__)
 def _journal_filter(user_log, search_term):
     """
     Filters sqlalchemy user_log based on search_term with whoosh Query language
     http://packages.python.org/Whoosh/querylang.html
     :param user_log:
     :param search_term:
     """
     log.debug('Initial search term: %r', search_term)
     qry = None
     if search_term:
         qp = QueryParser('repository', schema=JOURNAL_SCHEMA)
         qp.add_plugin(DateParserPlugin())
-        qry = qp.parse(unicode(search_term))
         qry = qp.parse(search_term)
         log.debug('Filtering using parsed query %r', qry)
     def wildcard_handler(col, wc_term):
         if wc_term.startswith('*') and not wc_term.endswith('*'):
             # postfix == endswith
             wc_term = remove_prefix(wc_term, prefix='*')
             return func.lower(col).endswith(func.lower(wc_term))
         elif wc_term.startswith('*') and wc_term.endswith('*'):
             # wildcard == ilike
             wc_term = remove_prefix(wc_term, prefix='*')
             wc_term = remove_suffix(wc_term, suffix='*')
             return func.lower(col).contains(func.lower(wc_term))
     def get_filterion(field, val, term):
         if field == 'repository':
             field = getattr(UserLog, 'repository_name')
         elif field == 'ip':
             field = getattr(UserLog, 'user_ip')
         elif field == 'date':
             field = getattr(UserLog, 'action_date')
         elif field == 'username':
             field = getattr(UserLog, 'username')
         else:
             field = getattr(UserLog, field)
         log.debug('filter field: %s val=>%s', field, val)
         # sql filtering
         if isinstance(term, query.Wildcard):
             return wildcard_handler(field, val)
         elif isinstance(term, query.Prefix):
             return func.lower(field).startswith(func.lower(val))
         elif isinstance(term, query.DateRange):
             return and_(field >= val[0], field <= val[1])
         return func.lower(field) == func.lower(val)
     if isinstance(qry, (query.And, query.Term, query.Prefix, query.Wildcard,
                         query.DateRange)):
         if not isinstance(qry, query.And):
             qry = [qry]
         for term in qry:
             field = term.fieldname
             val = (term.text if not isinstance(term, query.DateRange)
                    else [term.startdate, term.enddate])
             user_log = user_log.filter(get_filterion(field, val, term))
     elif isinstance(qry, query.Or):
         filters = []
         for term in qry:
             field = term.fieldname
             val = (term.text if not isinstance(term, query.DateRange)
                    else [term.startdate, term.enddate])
             filters.append(get_filterion(field, val, term))
         user_log = user_log.filter(or_(*filters))
     return user_log
 class AdminController(BaseController):
     @LoginRequired(allow_default_user=True)
     def _before(self, *args, **kwargs):
         super(AdminController, self)._before(*args, **kwargs)
     @HasPermissionAnyDecorator('hg.admin')
     def index(self):
         users_log = UserLog.query() \
                 .options(joinedload(UserLog.user)) \
                 .options(joinedload(UserLog.repository))
         # FILTERING
         c.search_term = request.GET.get('filter')
         users_log = _journal_filter(users_log, c.search_term)
         users_log = users_log.order_by(UserLog.action_date.desc())
         p = safe_int(request.GET.get('page'), 1)
         c.users_log = Page(users_log, page=p, items_per_page=10,
                            filter=c.search_term)
         if request.environ.get('HTTP_X_PARTIAL_XHR'):
             return render('admin/admin_log.html')
         return render('admin/admin.html')

kallithea/controllers/search.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.search
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Search 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 7, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 import traceback
 from tg import config, request
 from tg import tmpl_context as c
 from tg.i18n import ugettext as _
 from whoosh.index import EmptyIndexError, exists_in, open_dir
 from whoosh.qparser import QueryParser, QueryParserError
 from whoosh.query import Phrase, Prefix
 from kallithea.lib.auth import LoginRequired
 from kallithea.lib.base import BaseRepoController, render
 from kallithea.lib.indexers import CHGSET_IDX_NAME, CHGSETS_SCHEMA, IDX_NAME, SCHEMA, WhooshResultWrapper
 from kallithea.lib.page import Page
 from kallithea.lib.utils2 import safe_int
 from kallithea.model.repo import RepoModel
 log = logging.getLogger(__name__)
 class SearchController(BaseRepoController):
     @LoginRequired(allow_default_user=True)
     def index(self, repo_name=None):
         c.repo_name = repo_name
         c.formated_results = []
         c.runtime = ''
         c.cur_query = request.GET.get('q', None)
         c.cur_type = request.GET.get('type', 'content')
         c.cur_search = search_type = {'content': 'content',
                                       'commit': 'message',
                                       'path': 'path',
                                       'repository': 'repository'
                                       }.get(c.cur_type, 'content')
         index_name = {
             'content': IDX_NAME,
             'commit': CHGSET_IDX_NAME,
             'path': IDX_NAME
         }.get(c.cur_type, IDX_NAME)
         schema_defn = {
             'content': SCHEMA,
             'commit': CHGSETS_SCHEMA,
             'path': SCHEMA
         }.get(c.cur_type, SCHEMA)
         log.debug('IDX: %s', index_name)
         log.debug('SCHEMA: %s', schema_defn)
         if c.cur_query:
             cur_query = c.cur_query.lower()
             log.debug(cur_query)
         if c.cur_query:
             p = safe_int(request.GET.get('page'), 1)
             highlight_items = set()
             index_dir = config['index_dir']
             try:
                 if not exists_in(index_dir, index_name):
                     raise EmptyIndexError
                 idx = open_dir(index_dir, indexname=index_name)
                 searcher = idx.searcher()
                 qp = QueryParser(search_type, schema=schema_defn)
                 if c.repo_name:
                     # use "repository_rawname:" instead of "repository:"
                     # for case-sensitive matching
                     cur_query = u'repository_rawname:%s %s' % (c.repo_name, cur_query)
                 try:
-                    query = qp.parse(unicode(cur_query))
                     query = qp.parse(cur_query)
                     # extract words for highlight
                     if isinstance(query, Phrase):
                         highlight_items.update(query.words)
                     elif isinstance(query, Prefix):
                         highlight_items.add(query.text)
                     else:
                         for i in query.all_terms():
                             if i[0] in ['content', 'message']:
                                 highlight_items.add(i[1])
                     matcher = query.matcher(searcher)
                     log.debug('query: %s', query)
                     log.debug('hl terms: %s', highlight_items)
                     results = searcher.search(query)
                     res_ln = len(results)
                     c.runtime = '%s results (%.3f seconds)' % (
                         res_ln, results.runtime
+                    )
                     repo_location = RepoModel().repos_path
                     c.formated_results = Page(
                         WhooshResultWrapper(search_type, searcher, matcher,
                                             highlight_items, repo_location),
                         page=p,
                         item_count=res_ln,
                         items_per_page=10,
                         type=c.cur_type,
                         q=c.cur_query,
+                    )
                 except QueryParserError:
                     c.runtime = _('Invalid search query. Try quoting it.')
                 searcher.close()
             except EmptyIndexError:
                 log.error("Empty search index - run 'kallithea-cli index-create' regularly")
                 c.runtime = _('The server has no search index.')
             except Exception:
                 log.error(traceback.format_exc())
                 c.runtime = _('An error occurred during search operation.')
         # Return a rendered template
         return render('/search/search.html')

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.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('info/refs'):
             test_uri = test_uri.rstrip('/') + '/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 '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, unicode)):
         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 = u'Unknown'
         return undefined_contact
     @property
     def branches(self):
         if not self.revisions:
             return {}
         sortkey = lambda ctx: ctx[0]
         _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=sortkey, 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 {}
         sortkey = lambda ctx: ctx[0]
         _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=sortkey, 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/inmemory.py

➞

Show inline comments

 import datetime
 import mercurial.context
 import mercurial.node
 from kallithea.lib.vcs.backends.base import BaseInMemoryChangeset
 from kallithea.lib.vcs.exceptions import RepositoryError
 from kallithea.lib.vcs.utils import ascii_str, safe_bytes, safe_str
 class MercurialInMemoryChangeset(BaseInMemoryChangeset):
     def commit(self, message, author, parents=None, branch=None, date=None,
                **kwargs):
         """
         Performs in-memory commit (doesn't check workdir in any way) and
         returns newly created ``Changeset``. Updates repository's
         ``revisions``.
         :param message: message of the commit
         :param author: full username, i.e. "Joe Doe <joe.doe@example.com>"
         :param parents: single parent or sequence of parents from which commit
           would be derived
         :param date: ``datetime.datetime`` instance. Defaults to
           ``datetime.datetime.now()``.
         :param branch: branch name, as string. If none given, default backend's
           branch would be used.
         :raises ``CommitError``: if any error occurs while committing
         """
         self.check_integrity(parents)
         if not isinstance(message, str):
             raise RepositoryError('message must be a str - got %r' % type(message))
         if not isinstance(author, str):
             raise RepositoryError('author must be a str - got %r' % type(author))
         from .repository import MercurialRepository
         if not isinstance(message, unicode) or not isinstance(author, unicode):
             raise RepositoryError('Given message and author needs to be '
                                   'an <unicode> instance got %r & %r instead'
                                   % (type(message), type(author)))
         if branch is None:
             branch = MercurialRepository.DEFAULT_BRANCH_NAME
         kwargs[b'branch'] = safe_bytes(branch)
         def filectxfn(_repo, memctx, bytes_path):
             """
             Callback from Mercurial, returning ctx to commit for the given
             path.
             """
             path = safe_str(bytes_path)
             # check if this path is removed
             if path in (node.path for node in self.removed):
                 return None
             # check if this path is added
             for node in self.added:
                 if node.path == path:
                     return mercurial.context.memfilectx(_repo, memctx, path=bytes_path,
                         data=node.content,
                         islink=False,
                         isexec=node.is_executable,
                         copysource=False)
             # or changed
             for node in self.changed:
                 if node.path == path:
                     return mercurial.context.memfilectx(_repo, memctx, path=bytes_path,
                         data=node.content,
                         islink=False,
                         isexec=node.is_executable,
                         copysource=False)
             raise RepositoryError("Given path haven't been marked as added, "
                                   "changed or removed (%s)" % path)
         parents = [None, None]
         for i, parent in enumerate(self.parents):
             if parent is not None:
                 parents[i] = parent._ctx.node()
         if date and isinstance(date, datetime.datetime):
             date = safe_bytes(date.strftime('%a, %d %b %Y %H:%M:%S'))
         commit_ctx = mercurial.context.memctx(
             repo=self.repository._repo,
             parents=parents,
             text=b'',
             files=[safe_bytes(x) for x in self.get_paths()],
             filectxfn=filectxfn,
             user=safe_bytes(author),
             date=date,
             extra=kwargs)
         # injecting given _repo params
         commit_ctx._text = safe_bytes(message)
         commit_ctx._user = safe_bytes(author)
         commit_ctx._date = date
         # TODO: Catch exceptions!
         n = self.repository._repo.commitctx(commit_ctx)
         # Returns mercurial node
         self._commit_ctx = commit_ctx  # For reference
         # Update vcs repository object & recreate mercurial _repo
         # new_ctx = self.repository._repo[node]
         # new_tip = ascii_str(self.repository.get_changeset(new_ctx.hex()))
         self.repository.revisions.append(ascii_str(mercurial.node.hex(n)))
         self._repo = self.repository._get_repo(create=False)
         self.repository.branches = self.repository._get_branches()
         tip = self.repository.get_changeset()
         self.reset()
         return tip

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.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, unicode)
         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 unicode in app_settings_value
         :param 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')
     users_group_repo_to_perm = relationship('UserGroupRepoToPerm', cascade='all')
     users_group_repo_group_to_perm = relationship('UserGroupRepoGroupToPerm', cascade='all')
     user_user_group_to_perm = relationship('UserUserGroupToPerm ', cascade='all')
     user_group_user_group_to_perm = relationship('UserGroupUserGroupToPerm ', primaryjoin="UserGroupUserGroupToPerm.target_user_group_id==UserGroup.users_group_id", cascade='all')
     owner = relationship('User')
     @hybrid_property
     def group_data(self):
         if not self._group_data:
             return {}
         try:
             return ext_json.loads(self._group_data)
         except TypeError:
             return {}
     @group_data.setter
     def group_data(self, val):
         try:
             self._group_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.users_group_id,
                                   self.users_group_name)
     @classmethod
     def guess_instance(cls, value):
         return super(UserGroup, cls).guess_instance(value, UserGroup.get_by_group_name)
     @classmethod
     def get_by_group_name(cls, group_name, cache=False,
                           case_insensitive=False):
         if case_insensitive:
             q = cls.query().filter(sqlalchemy.func.lower(cls.users_group_name) == sqlalchemy.func.lower(group_name))
         else:
             q = cls.query().filter(cls.users_group_name == group_name)
         if cache:
             q = q.options(FromCache(
                             "sql_cache_short",
                             "get_group_%s" % _hash_key(group_name)
+                          )
+            )
         return q.scalar()
     @classmethod
     def get(cls, user_group_id, cache=False):
         user_group = cls.query()
         if cache:
             user_group = user_group.options(FromCache("sql_cache_short",
                                     "get_users_group_%s" % user_group_id))
         return user_group.get(user_group_id)
     def get_api_data(self, with_members=True):
         user_group = self
         data = dict(
             users_group_id=user_group.users_group_id,
             group_name=user_group.users_group_name,
             group_description=user_group.user_group_description,
             active=user_group.users_group_active,
             owner=user_group.owner.username,
+        )
         if with_members:
             data['members'] = [
                 ugm.user.get_api_data()
                 for ugm in user_group.members
+            ]
         return data
 class UserGroupMember(Base, BaseDbModel):
     __tablename__ = 'users_groups_members'
     __table_args__ = (
         _table_args_default_dict,
+    )
     users_group_member_id = Column(Integer(), primary_key=True)
     users_group_id = Column(Integer(), ForeignKey('users_groups.users_group_id'), nullable=False)
     user_id = Column(Integer(), ForeignKey('users.user_id'), nullable=False)
     user = relationship('User')
     users_group = relationship('UserGroup')
     def __init__(self, gr_id='', u_id=''):
         self.users_group_id = gr_id
         self.user_id = u_id
 class RepositoryField(Base, BaseDbModel):
     __tablename__ = 'repositories_fields'
     __table_args__ = (
         UniqueConstraint('repository_id', 'field_key'),  # no-multi field
         _table_args_default_dict,
+    )
     PREFIX = 'ex_'  # prefix used in form to not conflict with already existing fields
     repo_field_id = Column(Integer(), primary_key=True)
     repository_id = Column(Integer(), ForeignKey('repositories.repo_id'), nullable=False)
     field_key = Column(String(250), nullable=False)
     field_label = Column(String(1024), nullable=False)
     field_value = Column(String(10000), nullable=False)
     field_desc = Column(String(1024), nullable=False)
     field_type = Column(String(255), nullable=False)
     created_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     repository = relationship('Repository')
     @property
     def field_key_prefixed(self):
         return 'ex_%s' % self.field_key
     @classmethod
     def un_prefix_key(cls, key):
         if key.startswith(cls.PREFIX):
             return key[len(cls.PREFIX):]
         return key
     @classmethod
     def get_by_key_name(cls, key, repo):
         row = cls.query() \
                 .filter(cls.repository == repo) \
                 .filter(cls.field_key == key).scalar()
         return row
 class Repository(Base, BaseDbModel):
     __tablename__ = 'repositories'
     __table_args__ = (
         Index('r_repo_name_idx', 'repo_name'),
         _table_args_default_dict,
+    )
     DEFAULT_CLONE_URI = '{scheme}://{user}@{netloc}/{repo}'
     DEFAULT_CLONE_SSH = 'ssh://{system_user}@{hostname}/{repo}'
     STATE_CREATED = u'repo_state_created'
     STATE_PENDING = u'repo_state_pending'
     STATE_ERROR = u'repo_state_error'
     repo_id = Column(Integer(), primary_key=True)
     repo_name = Column(Unicode(255), nullable=False, unique=True)
     repo_state = Column(String(255), nullable=False)
     clone_uri = Column(String(255), nullable=True) # FIXME: not nullable?
     repo_type = Column(String(255), nullable=False) # 'hg' or 'git'
     owner_id = Column('user_id', Integer(), ForeignKey('users.user_id'), nullable=False)
     private = Column(Boolean(), nullable=False)
     enable_statistics = Column("statistics", Boolean(), nullable=False, default=True)
     enable_downloads = Column("downloads", Boolean(), nullable=False, default=True)
     description = Column(Unicode(10000), nullable=False)
     created_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     updated_on = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now)
     _landing_revision = Column("landing_revision", String(255), nullable=False)
     _changeset_cache = Column("changeset_cache", LargeBinary(), nullable=True) # JSON data # FIXME: not nullable?
     fork_id = Column(Integer(), ForeignKey('repositories.repo_id'), nullable=True)
     group_id = Column(Integer(), ForeignKey('groups.group_id'), nullable=True)
     owner = relationship('User')
     fork = relationship('Repository', remote_side=repo_id)
     group = relationship('RepoGroup')
     repo_to_perm = relationship('UserRepoToPerm', cascade='all', order_by='UserRepoToPerm.repo_to_perm_id')
     users_group_to_perm = relationship('UserGroupRepoToPerm', cascade='all')
     stats = relationship('Statistics', cascade='all', uselist=False)
     followers = relationship('UserFollowing',
                              primaryjoin='UserFollowing.follows_repository_id==Repository.repo_id',
                              cascade='all')
     extra_fields = relationship('RepositoryField',
                                 cascade="all, delete-orphan")
     logs = relationship('UserLog')
     comments = relationship('ChangesetComment', cascade="all, delete-orphan")
     pull_requests_org = relationship('PullRequest',
                     primaryjoin='PullRequest.org_repo_id==Repository.repo_id',
                     cascade="all, delete-orphan")
     pull_requests_other = relationship('PullRequest',
                     primaryjoin='PullRequest.other_repo_id==Repository.repo_id',
                     cascade="all, delete-orphan")
     def __repr__(self):
         return "<%s %s: %r>" % (self.__class__.__name__,
                                   self.repo_id, self.repo_name)
     @hybrid_property
     def landing_rev(self):
         # always should return [rev_type, rev]
         if self._landing_revision:
             _rev_info = self._landing_revision.split(':')
             if len(_rev_info) < 2:
                 _rev_info.insert(0, 'rev')
             return [_rev_info[0], _rev_info[1]]
         return [None, None]
     @landing_rev.setter
     def landing_rev(self, val):
         if ':' not in val:
             raise ValueError('value must be delimited with `:` and consist '
                              'of <rev_type>:<rev>, got %s instead' % val)
         self._landing_revision = val
     @hybrid_property
     def changeset_cache(self):
         try:
             cs_cache = ext_json.loads(self._changeset_cache) # might raise on bad data
             cs_cache['raw_id'] # verify data, raise exception on error
             return cs_cache
         except (TypeError, KeyError, ValueError):
             return EmptyChangeset().__json__()
     @changeset_cache.setter
     def changeset_cache(self, val):
         try:
             self._changeset_cache = ascii_bytes(ext_json.dumps(val))
         except Exception:
             log.error(traceback.format_exc())
     @classmethod
     def query(cls, sorted=False):
         """Add Repository-specific helpers for common query constructs.
         sorted: if True, apply the default ordering (name, case insensitive).
         """
         q = super(Repository, cls).query()
         if sorted:
             q = q.order_by(sqlalchemy.func.lower(Repository.repo_name))
         return q
     @classmethod
     def url_sep(cls):
         return URL_SEP
     @classmethod
     def normalize_repo_name(cls, repo_name):
         """
         Normalizes os specific repo_name to the format internally stored inside
         database using URL_SEP
         :param cls:
         :param repo_name:
         """
         return cls.url_sep().join(repo_name.split(os.sep))
     @classmethod
     def guess_instance(cls, value):
         return super(Repository, cls).guess_instance(value, Repository.get_by_repo_name)
     @classmethod
     def get_by_repo_name(cls, repo_name, case_insensitive=False):
         """Get the repo, defaulting to database case sensitivity.
         case_insensitive will be slower and should only be specified if necessary."""
         if case_insensitive:
             q = Session().query(cls).filter(sqlalchemy.func.lower(cls.repo_name) == sqlalchemy.func.lower(repo_name))
         else:
             q = Session().query(cls).filter(cls.repo_name == repo_name)
         q = q.options(joinedload(Repository.fork)) \
                 .options(joinedload(Repository.owner)) \
                 .options(joinedload(Repository.group))
         return q.scalar()
     @classmethod
     def get_by_full_path(cls, repo_full_path):
         base_full_path = os.path.realpath(cls.base_path())
         repo_full_path = os.path.realpath(repo_full_path)
         assert repo_full_path.startswith(base_full_path + os.path.sep)
         repo_name = repo_full_path[len(base_full_path) + 1:]
         repo_name = cls.normalize_repo_name(repo_name)
         return cls.get_by_repo_name(repo_name.strip(URL_SEP))
     @classmethod
     def get_repo_forks(cls, repo_id):
         return cls.query().filter(Repository.fork_id == repo_id)
     @classmethod
     def base_path(cls):
         """
         Returns base path where all repos are stored
         :param cls:
         """
         q = Session().query(Ui) \
             .filter(Ui.ui_key == cls.url_sep())
         q = q.options(FromCache("sql_cache_short", "repository_repo_path"))
         return q.one().ui_value
     @property
     def forks(self):
         """
         Return forks of this repo
         """
         return Repository.get_repo_forks(self.repo_id)
     @property
     def parent(self):
         """
         Returns fork parent
         """
         return self.fork
     @property
     def just_name(self):
         return self.repo_name.split(Repository.url_sep())[-1]
     @property
     def groups_with_parents(self):
         groups = []
         group = self.group
         while group is not None:
             groups.append(group)
             group = group.parent_group
             assert group not in groups, group # avoid recursion on bad db content
         groups.reverse()
         return groups
     @LazyProperty
     def repo_path(self):
         """
         Returns base full path for that repository means where it actually
         exists on a filesystem
         """
         q = Session().query(Ui).filter(Ui.ui_key ==
                                               Repository.url_sep())
         q = q.options(FromCache("sql_cache_short", "repository_repo_path"))
         return q.one().ui_value
     @property
     def repo_full_path(self):
         p = [self.repo_path]
         # we need to split the name by / since this is how we store the
         # names in the database, but that eventually needs to be converted
         # into a valid system path
         p += self.repo_name.split(Repository.url_sep())
         return os.path.join(*p)
     @property
     def cache_keys(self):
         """
         Returns associated cache keys for that repo
         """
         return CacheInvalidation.query() \
             .filter(CacheInvalidation.cache_args == self.repo_name) \
             .order_by(CacheInvalidation.cache_key) \
             .all()
     def get_new_name(self, repo_name):
         """
         returns new full repository name based on assigned group and new new
         :param group_name:
         """
         path_prefix = self.group.full_path_splitted if self.group else []
         return Repository.url_sep().join(path_prefix + [repo_name])
     @property
     def _ui(self):
         """
         Creates an db based ui object for this repository
         """
         from kallithea.lib.utils import make_ui
         return make_ui()
     @classmethod
     def is_valid(cls, repo_name):
         """
         returns True if given repo name is a valid filesystem repository
         :param cls:
         :param repo_name:
         """
         from kallithea.lib.utils import is_valid_repo
         return is_valid_repo(repo_name, cls.base_path())
     def get_api_data(self, with_revision_names=False,
                            with_pullrequests=False):
         """
         Common function for generating repo api data.
         Optionally, also return tags, branches, bookmarks and PRs.
         """
         repo = self
         data = dict(
             repo_id=repo.repo_id,
             repo_name=repo.repo_name,
             repo_type=repo.repo_type,
             clone_uri=repo.clone_uri,
             private=repo.private,
             created_on=repo.created_on,
             description=repo.description,
             landing_rev=repo.landing_rev,
             owner=repo.owner.username,
             fork_of=repo.fork.repo_name if repo.fork else None,
             enable_statistics=repo.enable_statistics,
             enable_downloads=repo.enable_downloads,
             last_changeset=repo.changeset_cache,
+        )
         if with_revision_names:
             scm_repo = repo.scm_instance_no_cache()
             data.update(dict(
                 tags=scm_repo.tags,
                 branches=scm_repo.branches,
                 bookmarks=scm_repo.bookmarks,
             ))
         if with_pullrequests:
             data['pull_requests'] = repo.pull_requests_other
         rc_config = Setting.get_app_settings()
         repository_fields = str2bool(rc_config.get('repository_fields'))
         if repository_fields:
             for f in self.extra_fields:
                 data[f.field_key_prefixed] = f.field_value
         return data
     @property
     def last_db_change(self):
         return self.updated_on
     @property
     def clone_uri_hidden(self):
         clone_uri = self.clone_uri
         if clone_uri:
             import urlobject
             url_obj = urlobject.URLObject(self.clone_uri)
             if url_obj.password:
                 clone_uri = url_obj.with_password('*****')
         return clone_uri
     def clone_url(self, clone_uri_tmpl, with_id=False, username=None):
         if '{repo}' not in clone_uri_tmpl and '_{repoid}' not in clone_uri_tmpl:
             log.error("Configured clone_uri_tmpl %r has no '{repo}' or '_{repoid}' and cannot toggle to use repo id URLs", clone_uri_tmpl)
         elif with_id:
             clone_uri_tmpl = clone_uri_tmpl.replace('{repo}', '_{repoid}')
         else:
             clone_uri_tmpl = clone_uri_tmpl.replace('_{repoid}', '{repo}')
         import kallithea.lib.helpers as h
         prefix_url = h.canonical_url('home')
         return get_clone_url(clone_uri_tmpl=clone_uri_tmpl,
                              prefix_url=prefix_url,
                              repo_name=self.repo_name,
                              repo_id=self.repo_id,
                              username=username)
     def set_state(self, state):
         self.repo_state = state
     #==========================================================================
     # SCM PROPERTIES
     #==========================================================================
     def get_changeset(self, rev=None):
         return get_changeset_safe(self.scm_instance, rev)
     def get_landing_changeset(self):
         """
         Returns landing changeset, or if that doesn't exist returns the tip
         """
         _rev_type, _rev = self.landing_rev
         cs = self.get_changeset(_rev)
         if isinstance(cs, EmptyChangeset):
             return self.get_changeset()
         return cs
     def update_changeset_cache(self, cs_cache=None):
         """
         Update cache of last changeset for repository, keys should be::
             short_id
             raw_id
             revision
             message
             date
             author
         :param cs_cache:
         """
         from kallithea.lib.vcs.backends.base import BaseChangeset
         if cs_cache is None:
             cs_cache = EmptyChangeset()
             # use no-cache version here
             scm_repo = self.scm_instance_no_cache()
             if scm_repo:
                 cs_cache = scm_repo.get_changeset()
         if isinstance(cs_cache, BaseChangeset):
             cs_cache = cs_cache.__json__()
         if (not self.changeset_cache or cs_cache['raw_id'] != self.changeset_cache['raw_id']):
             _default = datetime.datetime.fromtimestamp(0)
             last_change = cs_cache.get('date') or _default
             log.debug('updated repo %s with new cs cache %s',
                       self.repo_name, cs_cache)
             self.updated_on = last_change
             self.changeset_cache = cs_cache
             Session().commit()
         else:
             log.debug('changeset_cache for %s already up to date with %s',
                       self.repo_name, cs_cache['raw_id'])
     @property
     def tip(self):
         return self.get_changeset('tip')
     @property
     def author(self):
         return self.tip.author
     @property
     def last_change(self):
         return self.scm_instance.last_change
     def get_comments(self, revisions=None):
         """
         Returns comments for this repository grouped by revisions
         :param revisions: filter query by revisions only
         """
         cmts = ChangesetComment.query() \
             .filter(ChangesetComment.repo == self)
         if revisions is not None:
             if not revisions:
                 return {} # don't use sql 'in' on empty set
             cmts = cmts.filter(ChangesetComment.revision.in_(revisions))
         grouped = collections.defaultdict(list)
         for cmt in cmts.all():
             grouped[cmt.revision].append(cmt)
         return grouped
     def statuses(self, revisions):
         """
         Returns statuses for this repository.
         PRs without any votes do _not_ show up as unreviewed.
         :param revisions: list of revisions to get statuses for
         """
         if not revisions:
             return {}
         statuses = ChangesetStatus.query() \
             .filter(ChangesetStatus.repo == self) \
             .filter(ChangesetStatus.version == 0) \
             .filter(ChangesetStatus.revision.in_(revisions))
         grouped = {}
         for stat in statuses.all():
             pr_id = pr_nice_id = pr_repo = None
             if stat.pull_request:
                 pr_id = stat.pull_request.pull_request_id
                 pr_nice_id = PullRequest.make_nice_id(pr_id)
                 pr_repo = stat.pull_request.other_repo.repo_name
             grouped[stat.revision] = [str(stat.status), stat.status_lbl,
                                       pr_id, pr_repo, pr_nice_id,
                                       stat.author]
         return grouped
     def _repo_size(self):
         from kallithea.lib import helpers as h
         log.debug('calculating repository size...')
         return h.format_byte_size(self.scm_instance.size)
     #==========================================================================
     # SCM CACHE INSTANCE
     #==========================================================================
     def set_invalidate(self):
         """
         Mark caches of this repo as invalid.
         """
         CacheInvalidation.set_invalidate(self.repo_name)
     _scm_instance = None
     @property
     def scm_instance(self):
         if self._scm_instance is None:
             self._scm_instance = self.scm_instance_cached()
         return self._scm_instance
     def scm_instance_cached(self, valid_cache_keys=None):
         @cache_region('long_term', 'scm_instance_cached')
         def _c(repo_name): # repo_name is just for the cache key
             log.debug('Creating new %s scm_instance and populating cache', repo_name)
             return self.scm_instance_no_cache()
         rn = self.repo_name
         valid = CacheInvalidation.test_and_set_valid(rn, None, valid_cache_keys=valid_cache_keys)
         if not valid:
             log.debug('Cache for %s invalidated, getting new object', rn)
             region_invalidate(_c, None, 'scm_instance_cached', rn)
         else:
             log.debug('Trying to get scm_instance of %s from cache', rn)
         return _c(rn)
     def scm_instance_no_cache(self):
         repo_full_path = self.repo_full_path
         alias = get_scm(repo_full_path)[0]
         log.debug('Creating instance of %s repository from %s',
                   alias, self.repo_full_path)
         backend = get_backend(alias)
         if alias == 'hg':
             repo = backend(repo_full_path, create=False,
                            baseui=self._ui)
         else:
             repo = backend(repo_full_path, create=False)
         return repo
     def __json__(self):
         return dict(
             repo_id=self.repo_id,
             repo_name=self.repo_name,
             landing_rev=self.landing_rev,
+        )
 class RepoGroup(Base, BaseDbModel):
     __tablename__ = 'groups'
     __table_args__ = (
         _table_args_default_dict,
+    )
     SEP = ' &raquo; '
     group_id = Column(Integer(), primary_key=True)
     group_name = Column(Unicode(255), nullable=False, unique=True) # full path
     parent_group_id = Column('group_parent_id', Integer(), ForeignKey('groups.group_id'), nullable=True)
     group_description = Column(Unicode(10000), 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)
     repo_group_to_perm = relationship('UserRepoGroupToPerm', cascade='all', order_by='UserRepoGroupToPerm.group_to_perm_id')
     users_group_to_perm = relationship('UserGroupRepoGroupToPerm', cascade='all')
     parent_group = relationship('RepoGroup', remote_side=group_id)
     owner = relationship('User')
     @classmethod
     def query(cls, sorted=False):
         """Add RepoGroup-specific helpers for common query constructs.
         sorted: if True, apply the default ordering (name, case insensitive).
         """
         q = super(RepoGroup, cls).query()
         if sorted:
             q = q.order_by(sqlalchemy.func.lower(RepoGroup.group_name))
         return q
     def __init__(self, group_name='', parent_group=None):
         self.group_name = group_name
         self.parent_group = parent_group
     def __repr__(self):
         return "<%s %s: %s>" % (self.__class__.__name__,
                                 self.group_id, self.group_name)
     @classmethod
     def _generate_choice(cls, repo_group):
         """Return tuple with group_id and name as html literal"""
         from webhelpers2.html import literal
         if repo_group is None:
             return (-1, u'-- %s --' % _('top level'))
         return repo_group.group_id, literal(cls.SEP.join(repo_group.full_path_splitted))
     @classmethod
     def groups_choices(cls, groups):
         """Return tuples with group_id and name as html literal."""
         return sorted((cls._generate_choice(g) for g in groups),
                       key=lambda c: c[1].split(cls.SEP))
     @classmethod
     def url_sep(cls):
         return URL_SEP
     @classmethod
     def guess_instance(cls, value):
         return super(RepoGroup, cls).guess_instance(value, RepoGroup.get_by_group_name)
     @classmethod
     def get_by_group_name(cls, group_name, cache=False, case_insensitive=False):
         group_name = group_name.rstrip('/')
         if case_insensitive:
             gr = cls.query() \
                 .filter(sqlalchemy.func.lower(cls.group_name) == sqlalchemy.func.lower(group_name))
         else:
             gr = cls.query() \
                 .filter(cls.group_name == group_name)
         if cache:
             gr = gr.options(FromCache(
                             "sql_cache_short",
                             "get_group_%s" % _hash_key(group_name)
+                            )
+            )
         return gr.scalar()
     @property
     def parents(self):
         groups = []
         group = self.parent_group
         while group is not None:
             groups.append(group)
             group = group.parent_group
             assert group not in groups, group # avoid recursion on bad db content
         groups.reverse()
         return groups
     @property
     def children(self):
         return RepoGroup.query().filter(RepoGroup.parent_group == self)
     @property
     def name(self):
         return self.group_name.split(RepoGroup.url_sep())[-1]
     @property
     def full_path(self):
         return self.group_name
     @property
     def full_path_splitted(self):
         return self.group_name.split(RepoGroup.url_sep())
     @property
     def repositories(self):
         return Repository.query(sorted=True).filter_by(group=self)
     @property
     def repositories_recursive_count(self):
         cnt = self.repositories.count()
         def children_count(group):
             cnt = 0
             for child in group.children:
                 cnt += child.repositories.count()
                 cnt += children_count(child)
             return cnt
         return cnt + children_count(self)
     def _recursive_objects(self, include_repos=True):
         all_ = []
         def _get_members(root_gr):
             if include_repos:
                 for r in root_gr.repositories:
                     all_.append(r)
             childs = root_gr.children.all()
             if childs:
                 for gr in childs:
                     all_.append(gr)
                     _get_members(gr)
         _get_members(self)
         return [self] + all_
     def recursive_groups_and_repos(self):
         """
         Recursive return all groups, with repositories in those groups
         """
         return self._recursive_objects()
     def recursive_groups(self):
         """
         Returns all children groups for this group including children of children
         """
         return self._recursive_objects(include_repos=False)
     def get_new_name(self, group_name):
         """
         returns new full group name based on parent and new name
         :param group_name:
         """
         path_prefix = (self.parent_group.full_path_splitted if
                        self.parent_group else [])
         return RepoGroup.url_sep().join(path_prefix + [group_name])
     def get_api_data(self):
         """
         Common function for generating api data
         """
         group = self
         data = dict(
             group_id=group.group_id,
             group_name=group.group_name,
             group_description=group.group_description,
             parent_group=group.parent_group.group_name if group.parent_group else None,
             repositories=[x.repo_name for x in group.repositories],
             owner=group.owner.username
+        )
         return data
 class Permission(Base, BaseDbModel):
     __tablename__ = 'permissions'
     __table_args__ = (
         Index('p_perm_name_idx', 'permission_name'),
         _table_args_default_dict,
+    )
     PERMS = (
         ('hg.admin', _('Kallithea Administrator')),
         ('repository.none', _('Default user has no access to new repositories')),
         ('repository.read', _('Default user has read access to new repositories')),
         ('repository.write', _('Default user has write access to new repositories')),
         ('repository.admin', _('Default user has admin access to new repositories')),
         ('group.none', _('Default user has no access to new repository groups')),
         ('group.read', _('Default user has read access to new repository groups')),
         ('group.write', _('Default user has write access to new repository groups')),
         ('group.admin', _('Default user has admin access to new repository groups')),
         ('usergroup.none', _('Default user has no access to new user groups')),
         ('usergroup.read', _('Default user has read access to new user groups')),
         ('usergroup.write', _('Default user has write access to new user groups')),
         ('usergroup.admin', _('Default user has admin access to new user groups')),
         ('hg.repogroup.create.false', _('Only admins can create repository groups')),
         ('hg.repogroup.create.true', _('Non-admins can create repository groups')),
         ('hg.usergroup.create.false', _('Only admins can create user groups')),
         ('hg.usergroup.create.true', _('Non-admins can create user groups')),
         ('hg.create.none', _('Only admins can create top level repositories')),
         ('hg.create.repository', _('Non-admins can create top level repositories')),
         ('hg.create.write_on_repogroup.true', _('Repository creation enabled with write permission to a repository group')),
         ('hg.create.write_on_repogroup.false', _('Repository creation disabled with write permission to a repository group')),
         ('hg.fork.none', _('Only admins can fork repositories')),
         ('hg.fork.repository', _('Non-admins can fork repositories')),
         ('hg.register.none', _('Registration disabled')),
         ('hg.register.manual_activate', _('User registration with manual account activation')),
         ('hg.register.auto_activate', _('User registration with automatic account activation')),
         ('hg.extern_activate.manual', _('Manual activation of external account')),
         ('hg.extern_activate.auto', _('Automatic activation of external account')),
+    )
     # definition of system default permissions for DEFAULT user
     DEFAULT_USER_PERMISSIONS = (
         'repository.read',
         'group.read',
         'usergroup.read',
         'hg.create.repository',
         'hg.create.write_on_repogroup.true',
         'hg.fork.repository',
         'hg.register.manual_activate',
         'hg.extern_activate.auto',
+    )
     # defines which permissions are more important higher the more important
     # Weight defines which permissions are more important.
     # The higher number the more important.
     PERM_WEIGHTS = {
         'repository.none': 0,
         'repository.read': 1,
         'repository.write': 3,
         'repository.admin': 4,
         'group.none': 0,
         'group.read': 1,
         'group.write': 3,
         'group.admin': 4,
         'usergroup.none': 0,
         'usergroup.read': 1,
         'usergroup.write': 3,
         'usergroup.admin': 4,
         'hg.repogroup.create.false': 0,
         'hg.repogroup.create.true': 1,
         'hg.usergroup.create.false': 0,
         'hg.usergroup.create.true': 1,
         'hg.fork.none': 0,
         'hg.fork.repository': 1,
         'hg.create.none': 0,
         'hg.create.repository': 1,
         'hg.create.write_on_repogroup.false': 0,
         'hg.create.write_on_repogroup.true': 1,
         'hg.register.none': 0,
         'hg.register.manual_activate': 1,
         'hg.register.auto_activate': 2,
         'hg.extern_activate.manual': 0,
         'hg.extern_activate.auto': 1,
+    }
     permission_id = Column(Integer(), primary_key=True)
     permission_name = Column(String(255), nullable=False)
     def __repr__(self):
         return "<%s %s: %r>" % (
             self.__class__.__name__, self.permission_id, self.permission_name
+        )
     @classmethod
     def guess_instance(cls, value):
         return super(Permission, cls).guess_instance(value, Permission.get_by_key)
     @classmethod
     def get_by_key(cls, key):
         return cls.query().filter(cls.permission_name == key).scalar()
     @classmethod
     def get_default_perms(cls, default_user_id):
         q = Session().query(UserRepoToPerm, Repository, cls) \
          .join((Repository, UserRepoToPerm.repository_id == Repository.repo_id)) \
          .join((cls, UserRepoToPerm.permission_id == cls.permission_id)) \
          .filter(UserRepoToPerm.user_id == default_user_id)
         return q.all()
     @classmethod
     def get_default_group_perms(cls, default_user_id):
         q = Session().query(UserRepoGroupToPerm, RepoGroup, cls) \
          .join((RepoGroup, UserRepoGroupToPerm.group_id == RepoGroup.group_id)) \
          .join((cls, UserRepoGroupToPerm.permission_id == cls.permission_id)) \

scripts/update-copyrights.py

➞

Show inline comments

 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
 Kallithea script for maintaining contributor lists from version control
 history.
 This script and the data in it is a best effort attempt at reverse engineering
 previous attributions and correlate that with version control history while
 preserving all existing copyright statements and attribution. This script is
 processing and summarizing information found elsewhere - it is not by itself
 making any claims. Comments in the script are an attempt at reverse engineering
 possible explanations - they are not showing any intent or confirming it is
 correct.
 Three files are generated / modified by this script:
 kallithea/templates/about.html claims to show copyright holders, and the GPL
 license requires such existing "legal notices" to be preserved. We also try to
 keep it updated with copyright holders, but do not claim it is a correct list.
 CONTRIBUTORS has the purpose of giving credit where credit is due and list all
 the contributor names in the source.
 kallithea/templates/base/base.html contains the copyright years in the page
 footer.
 Both make a best effort of listing all copyright holders, but revision control
 history might be a better and more definitive source.
 Contributors are sorted "fairly" by copyright year and amount of
 contribution.
 New contributors are listed, without considering if the contribution contains
 copyrightable work.
 When the copyright might belong to a different legal entity than the
 contributor, the legal entity is given credit too.
 """
 import os
 import re
 from collections import defaultdict
 from . import contributor_data
 def sortkey(x):
     """Return key for sorting contributors "fairly":
     * latest contribution
     * first contribution
     * number of contribution years
     * name (with some unicode normalization)
-    The entries must be 2-tuples of a list of string years and the unicode name"""
     The entries must be 2-tuples of a list of string years and the name"""
     return (x[0] and -int(x[0][-1]),
             x[0] and int(x[0][0]),
             -len(x[0]),
             x[1].decode('utf-8').lower().replace(u'\xe9', u'e').replace(u'\u0142', u'l')
+        )
 def nice_years(l, dash='-', join=' '):
     """Convert a list of years into brief range like '1900-1901, 1921'."""
     if not l:
         return ''
     start = end = int(l[0])
     ranges = []
     for year in l[1:] + [0]:
         year = int(year)
         if year == end + 1:
             end = year
             continue
         if start == end:
             ranges.append('%s' % start)
         else:
             ranges.append('%s%s%s' % (start, dash, end))
         start = end = year
     assert start == 0 and end == 0, (start, end)
     return join.join(ranges)
 def insert_entries(
         filename,
         all_entries,
         no_entries,
         domain_extra,
         split_re,
         normalize_name,
         format_f):
     """Update file with contributor information.
     all_entries: list of tuples with year and name
     no_entries: set of names or name and year tuples to ignore
     domain_extra: map domain name to extra credit name
     split_re: regexp matching the part of file to rewrite
     normalize_name: function to normalize names for grouping and display
     format_f: function formatting year list and name to a string
     """
     name_years = defaultdict(set)
     for year, name in all_entries:
         if name in no_entries or (name, year) in no_entries:
             continue
         parts = name.split(' <', 1)
         if len(parts) == 2:
             name = parts[0] + ' <' + parts[1].lower()
         domain = name.split('@', 1)[-1].rstrip('>')
         if domain in domain_extra:
             name_years[domain_extra[domain]].add(year)
         name_years[normalize_name(name)].add(year)
     l = [(list(sorted(year for year in years if year)), name)
          for name, years in name_years.items()]
     l.sort(key=sortkey)
     with open(filename) as f:
         pre, post = re.split(split_re, f.read())
     with open(filename, 'w') as f:
         f.write(pre +
                 ''.join(format_f(years, name) for years, name in l) +
                 post)
 def main():
     repo_entries = [
         (year, contributor_data.name_fixes.get(name) or contributor_data.name_fixes.get(name.rsplit('<', 1)[0].strip()) or name)
         for year, name in
         (line.strip().split(' ', 1)
          for line in os.popen("""hg log -r '::.' -T '{date(date,"%Y")} {author}\n'""").readlines())
+        ]
     insert_entries(
         filename='kallithea/templates/about.html',
         all_entries=repo_entries + contributor_data.other_about + contributor_data.other,
         no_entries=contributor_data.no_about,
         domain_extra=contributor_data.domain_extra,
         split_re=r'(?:  <li>Copyright &copy; [^\n]*</li>\n)*',
         normalize_name=lambda name: name.split('<', 1)[0].strip(),
         format_f=lambda years, name: '  <li>Copyright &copy; %s, %s</li>\n' % (nice_years(years, '&ndash;', ', '), name),
+        )
     insert_entries(
         filename='CONTRIBUTORS',
         all_entries=repo_entries + contributor_data.other_contributors + contributor_data.other,
         no_entries=contributor_data.total_ignore,
         domain_extra=contributor_data.domain_extra,
         split_re=r'(?:    [^\n]*\n)*',
         normalize_name=lambda name: name,
         format_f=lambda years, name: ('    %s%s%s\n' % (name, ' ' if years else '', nice_years(years))),
+        )
     insert_entries(
         filename='kallithea/templates/base/base.html',
         all_entries=repo_entries,
         no_entries=contributor_data.total_ignore,
         domain_extra={},
         split_re=r'(?<=&copy;) .* (?=by various authors)',
         normalize_name=lambda name: '',
         format_f=lambda years, name: ' ' + nice_years(years, '&ndash;', ', ') + ' ',
+        )
     #docs/conf.py:copyright = u'2010-2016 by various authors, licensed as GPLv3.'
     insert_entries(
         filename='docs/conf.py',
         all_entries=repo_entries,
         no_entries=contributor_data.total_ignore,
         domain_extra={},
         split_re=r"(?<=copyright = u').*(?= by various authors)",
         normalize_name=lambda name: '',
         format_f=lambda years, name: nice_years(years, '-', ', '),
+        )
 if __name__ == '__main__':
     main()
 # To list new contributors since last tagging:
 # { hg log -r '::tagged()' -T '    {author}\n    {author}\n'; hg log -r '::.' -T '    {author}\n' | sort | uniq; } | sort | uniq -u

0 comments (0 inline, 0 general)