kallithea Changeset - fb4b72c1c0f1

Changeset - fb4b72c1c0f1

Parent rev.

Child rev.

[Not reviewed]

default

0 14 0

Mads Kiilerich - 6 years ago 2020-02-05 23:03:55
mads@kiilerich.com

py3: automatic migration with 2to3 -f basestring

14 files changed with 18 insertions and 18 deletions:

kallithea/lib/celerypylons/__init__.py

kallithea/lib/helpers.py

kallithea/lib/rcmail/message.py

kallithea/lib/rcmail/smtp_mailer.py

kallithea/lib/utils.py

kallithea/lib/utils2.py

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

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

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

kallithea/lib/vcs/utils/__init__.py

kallithea/model/db.py

kallithea/model/scm.py

kallithea/model/user_group.py

kallithea/tests/scripts/manual_test_crawler.py

0 comments (0 inline, 0 general)

kallithea/lib/celerypylons/__init__.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
 Kallithea wrapper of Celery
 The Celery configuration is in the Kallithea ini file but must be converted to an
 entirely different format before Celery can use it.
 We read the configuration from tg.config at module import time. This module can
 thus not be imported in global scope but must be imported on demand in function
 scope after tg.config has been initialized.
 To make sure that the config really has been initialized, we check one of the
 mandatory settings.
 """
 import celery
 import celery.result as result
 import tg
 from celery.bin import worker
 from celery.task import task
 def celery_config(config):
     """Return Celery config object populated from relevant settings in a config dict, such as tg.config"""
     # Verify .ini file configuration has been loaded
     assert config['celery.imports'] == 'kallithea.lib.celerylib.tasks', 'Kallithea Celery configuration has not been loaded'
     class CeleryConfig(object):
         pass
     celery_config = CeleryConfig()
     PREFIXES = """ADMINS BROKER CASSANDRA CELERYBEAT CELERYD CELERYMON CELERY EMAIL SERVER""".split()
     LIST_PARAMS = """CELERY_IMPORTS ADMINS ROUTES CELERY_ACCEPT_CONTENT""".split()
     for config_key, config_value in sorted(config.items()):
         celery_key = config_key.replace('.', '_').upper()
         if celery_key.split('_', 1)[0] not in PREFIXES:
             continue
-        if not isinstance(config_value, basestring):
+        if not isinstance(config_value, str):
             continue
         if celery_key in LIST_PARAMS:
             celery_value = config_value.split()
         elif config_value.isdigit():
             celery_value = int(config_value)
         elif config_value.lower() in ['true', 'false']:
             celery_value = config_value.lower() == 'true'
         else:
             celery_value = config_value
         setattr(celery_config, celery_key, celery_value)
     return celery_config
 def make_app():
     """Create celery app from the TurboGears configuration file"""
     app = celery.Celery()
     app.config_from_object(celery_config(tg.config))
     return app

kallithea/lib/helpers.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/>.
 """
 Helper functions
 Consists of functions to typically be used within templates, but also
 available to Controllers. This module is available to both as 'h'.
 """
 import hashlib
 import json
 import logging
 import random
 import re
 import textwrap
 import urlparse
 from beaker.cache import cache_region
 from pygments import highlight as code_highlight
 from pygments.formatters.html import HtmlFormatter
 from tg.i18n import ugettext as _
 from webhelpers2.html import HTML, escape, literal
 from webhelpers2.html.tags import NotGiven, Option, Options, _input, _make_safe_id_component, checkbox, end_form
 from webhelpers2.html.tags import form as insecure_form
 from webhelpers2.html.tags import hidden, link_to, password, radio
 from webhelpers2.html.tags import select as webhelpers2_select
 from webhelpers2.html.tags import submit, text, textarea
 from webhelpers2.number import format_byte_size
 from webhelpers2.text import chop_at, truncate, wrap_paragraphs
 from kallithea.config.routing import url
 from kallithea.lib.annotate import annotate_highlight
 #==============================================================================
 # PERMS
 #==============================================================================
 from kallithea.lib.auth import HasPermissionAny, HasRepoGroupPermissionLevel, HasRepoPermissionLevel
 from kallithea.lib.markup_renderer import url_re
 from kallithea.lib.pygmentsutils import get_custom_lexer
 from kallithea.lib.utils2 import MENTIONS_REGEX, AttributeDict
 from kallithea.lib.utils2 import age as _age
 from kallithea.lib.utils2 import credentials_filter, safe_bytes, safe_int, safe_unicode, str2bool, time_to_datetime
 from kallithea.lib.vcs.backends.base import BaseChangeset, EmptyChangeset
 from kallithea.lib.vcs.exceptions import ChangesetDoesNotExistError
 #==============================================================================
 # SCM FILTERS available via h.
 #==============================================================================
 from kallithea.lib.vcs.utils import author_email, author_name
 log = logging.getLogger(__name__)
 def canonical_url(*args, **kargs):
     '''Like url(x, qualified=True), but returns url that not only is qualified
     but also canonical, as configured in canonical_url'''
     from kallithea import CONFIG
     try:
         parts = CONFIG.get('canonical_url', '').split('://', 1)
         kargs['host'] = parts[1]
         kargs['protocol'] = parts[0]
     except IndexError:
         kargs['qualified'] = True
     return url(*args, **kargs)
 def canonical_hostname():
     '''Return canonical hostname of system'''
     from kallithea import CONFIG
     try:
         parts = CONFIG.get('canonical_url', '').split('://', 1)
         return parts[1].split('/', 1)[0]
     except IndexError:
         parts = url('home', qualified=True).split('://', 1)
         return parts[1].split('/', 1)[0]
 def html_escape(s):
     """Return string with all html escaped.
     This is also safe for javascript in html but not necessarily correct.
     """
     return (s
         .replace('&', '&amp;')
         .replace(">", "&gt;")
         .replace("<", "&lt;")
         .replace('"', "&quot;")
         .replace("'", "&apos;") # Note: this is HTML5 not HTML4 and might not work in mails
+        )
 def js(value):
     """Convert Python value to the corresponding JavaScript representation.
     This is necessary to safely insert arbitrary values into HTML <script>
     sections e.g. using Mako template expression substitution.
     Note: Rather than using this function, it's preferable to avoid the
     insertion of values into HTML <script> sections altogether. Instead,
     data should (to the extent possible) be passed to JavaScript using
     data attributes or AJAX calls, eliminating the need for JS specific
     escaping.
     Note: This is not safe for use in attributes (e.g. onclick), because
     quotes are not escaped.
     Because the rules for parsing <script> varies between XHTML (where
     normal rules apply for any special characters) and HTML (where
     entities are not interpreted, but the literal string "</script>"
     is forbidden), the function ensures that the result never contains
     '&', '<' and '>', thus making it safe in both those contexts (but
     not in attributes).
     """
     return literal(
         ('(' + json.dumps(value) + ')')
         # In JSON, the following can only appear in string literals.
         .replace('&', r'\x26')
         .replace('<', r'\x3c')
         .replace('>', r'\x3e')
+    )
 def jshtml(val):
     """HTML escapes a string value, then converts the resulting string
     to its corresponding JavaScript representation (see `js`).
     This is used when a plain-text string (possibly containing special
     HTML characters) will be used by a script in an HTML context (e.g.
     element.innerHTML or jQuery's 'html' method).
     If in doubt, err on the side of using `jshtml` over `js`, since it's
     better to escape too much than too little.
     """
     return js(escape(val))
 def shorter(s, size=20, firstline=False, postfix='...'):
     """Truncate s to size, including the postfix string if truncating.
     If firstline, truncate at newline.
     """
     if firstline:
         s = s.split('\n', 1)[0].rstrip()
     if len(s) > size:
         return s[:size - len(postfix)] + postfix
     return s
 def reset(name, value, id=NotGiven, **attrs):
     """Create a reset button, similar to webhelpers2.html.tags.submit ."""
     return _input("reset", name, value, id, attrs)
 def select(name, selected_values, options, id=NotGiven, **attrs):
     """Convenient wrapper of webhelpers2 to let it accept options as a tuple list"""
     if isinstance(options, list):
         option_list = options
         # Handle old value,label lists ... where value also can be value,label lists
         options = Options()
         for x in option_list:
             if isinstance(x, tuple) and len(x) == 2:
                 value, label = x
-            elif isinstance(x, basestring):
+            elif isinstance(x, str):
                 value = label = x
             else:
                 log.error('invalid select option %r', x)
                 raise
             if isinstance(value, list):
                 og = options.add_optgroup(label)
                 for x in value:
                     if isinstance(x, tuple) and len(x) == 2:
                         group_value, group_label = x
-                    elif isinstance(x, basestring):
+                    elif isinstance(x, str):
                         group_value = group_label = x
                     else:
                         log.error('invalid select option %r', x)
                         raise
                     og.add_option(group_label, group_value)
             else:
                 options.add_option(label, value)
     return webhelpers2_select(name, selected_values, options, id=id, **attrs)
 safeid = _make_safe_id_component
 def FID(raw_id, path):
     """
     Creates a unique ID for filenode based on it's hash of path and revision
     it's safe to use in urls
     :param raw_id:
     :param path:
     """
     return 'C-%s-%s' % (short_id(raw_id), hashlib.md5(safe_bytes(path)).hexdigest()[:12])
 class _FilesBreadCrumbs(object):
     def __call__(self, repo_name, rev, paths):
         if isinstance(paths, str):
             paths = safe_unicode(paths)
         url_l = [link_to(repo_name, url('files_home',
                                         repo_name=repo_name,
                                         revision=rev, f_path=''),
                          class_='ypjax-link')]
         paths_l = paths.split('/')
         for cnt, p in enumerate(paths_l):
             if p != '':
                 url_l.append(link_to(p,
                                      url('files_home',
                                          repo_name=repo_name,
                                          revision=rev,
                                          f_path='/'.join(paths_l[:cnt + 1])
                                          ),
                                      class_='ypjax-link'
+                                     )
+                             )
         return literal('/'.join(url_l))
 files_breadcrumbs = _FilesBreadCrumbs()
 class CodeHtmlFormatter(HtmlFormatter):
     """
     My code Html Formatter for source codes
     """
     def wrap(self, source, outfile):
         return self._wrap_div(self._wrap_pre(self._wrap_code(source)))
     def _wrap_code(self, source):
         for cnt, it in enumerate(source):
             i, t = it
             t = '<span id="L%s">%s</span>' % (cnt + 1, t)
             yield i, t
     def _wrap_tablelinenos(self, inner):
         inner_lines = []
         lncount = 0
         for t, line in inner:
             if t:
                 lncount += 1
             inner_lines.append(line)
         fl = self.linenostart
         mw = len(str(lncount + fl - 1))
         sp = self.linenospecial
         st = self.linenostep
         la = self.lineanchors
         aln = self.anchorlinenos
         nocls = self.noclasses
         if sp:
             lines = []
             for i in range(fl, fl + lncount):
                 if i % st == 0:
                     if i % sp == 0:
                         if aln:
                             lines.append('<a href="#%s%d" class="special">%*d</a>' %
                                          (la, i, mw, i))
                         else:
                             lines.append('<span class="special">%*d</span>' % (mw, i))
                     else:
                         if aln:
                             lines.append('<a href="#%s%d">%*d</a>' % (la, i, mw, i))
                         else:
                             lines.append('%*d' % (mw, i))
                 else:
                     lines.append('')
             ls = '\n'.join(lines)
         else:
             lines = []
             for i in range(fl, fl + lncount):
                 if i % st == 0:
                     if aln:
                         lines.append('<a href="#%s%d">%*d</a>' % (la, i, mw, i))
                     else:
                         lines.append('%*d' % (mw, i))
                 else:
                     lines.append('')
             ls = '\n'.join(lines)
         # in case you wonder about the seemingly redundant <div> here: since the
         # content in the other cell also is wrapped in a div, some browsers in
         # some configurations seem to mess up the formatting...
         if nocls:
             yield 0, ('<table class="%stable">' % self.cssclass +
                       '<tr><td><div class="linenodiv">'
                       '<pre>' + ls + '</pre></div></td>'
                       '<td id="hlcode" class="code">')
         else:
             yield 0, ('<table class="%stable">' % self.cssclass +
                       '<tr><td class="linenos"><div class="linenodiv">'
                       '<pre>' + ls + '</pre></div></td>'
                       '<td id="hlcode" class="code">')
         yield 0, ''.join(inner_lines)
         yield 0, '</td></tr></table>'
 _whitespace_re = re.compile(r'(\t)|( )(?=\n|</div>)')
 def _markup_whitespace(m):
     groups = m.groups()
     if groups[0]:
         return '<u>\t</u>'
     if groups[1]:
         return ' <i></i>'
 def markup_whitespace(s):
     return _whitespace_re.sub(_markup_whitespace, s)
 def pygmentize(filenode, **kwargs):
     """
     pygmentize function using pygments
     :param filenode:
     """
     lexer = get_custom_lexer(filenode.extension) or filenode.lexer
     return literal(markup_whitespace(
         code_highlight(safe_unicode(filenode.content), lexer, CodeHtmlFormatter(**kwargs))))
 def hsv_to_rgb(h, s, v):
     if s == 0.0:
         return v, v, v
     i = int(h * 6.0)  # XXX assume int() truncates!
     f = (h * 6.0) - i
     p = v * (1.0 - s)
     q = v * (1.0 - s * f)
     t = v * (1.0 - s * (1.0 - f))
     i = i % 6
     if i == 0:
         return v, t, p
     if i == 1:
         return q, v, p
     if i == 2:
         return p, v, t
     if i == 3:
         return p, q, v
     if i == 4:
         return t, p, v
     if i == 5:
         return v, p, q
 def gen_color(n=10000):
     """generator for getting n of evenly distributed colors using
     hsv color and golden ratio. It always return same order of colors
     :returns: RGB tuple
     """
     golden_ratio = 0.618033988749895
     h = 0.22717784590367374
     for _unused in range(n):
         h += golden_ratio
         h %= 1

kallithea/lib/rcmail/message.py

➞

Show inline comments

 from kallithea.lib.rcmail.exceptions import BadHeaders, InvalidMessage
 from kallithea.lib.rcmail.response import MailResponse
 class Attachment(object):
     """
     Encapsulates file attachment information.
     :param filename: filename of attachment
     :param content_type: file mimetype
     :param data: the raw file data, either as string or file obj
     :param disposition: content-disposition (if any)
     """
     def __init__(self,
                  filename=None,
                  content_type=None,
                  data=None,
                  disposition=None):
         self.filename = filename
         self.content_type = content_type
         self.disposition = disposition or 'attachment'
         self._data = data
     @property
     def data(self):
-        if isinstance(self._data, basestring):
+        if isinstance(self._data, str):
             return self._data
         self._data = self._data.read()
         return self._data
 class Message(object):
     """
     Encapsulates an email message.
     :param subject: email subject header
     :param recipients: list of email addresses
     :param body: plain text message
     :param html: HTML message
     :param sender: email sender address
     :param cc: CC list
     :param bcc: BCC list
     :param extra_headers: dict of extra email headers
     :param attachments: list of Attachment instances
     :param recipients_separator: alternative separator for any of
         'From', 'To', 'Delivered-To', 'Cc', 'Bcc' fields
     """
     def __init__(self,
                  subject=None,
                  recipients=None,
                  body=None,
                  html=None,
                  sender=None,
                  cc=None,
                  bcc=None,
                  extra_headers=None,
                  attachments=None,
                  recipients_separator="; "):
         self.subject = subject or ''
         self.sender = sender
         self.body = body
         self.html = html
         self.recipients = recipients or []
         self.attachments = attachments or []
         self.cc = cc or []
         self.bcc = bcc or []
         self.extra_headers = extra_headers or {}
         self.recipients_separator = recipients_separator
     @property
     def send_to(self):
         return set(self.recipients) | set(self.bcc or ()) | set(self.cc or ())
     def to_message(self):
         """
         Returns raw email.Message instance.Validates message first.
         """
         self.validate()
         return self.get_response().to_message()
     def get_response(self):
         """
         Creates a Lamson MailResponse instance
         """
         response = MailResponse(Subject=self.subject,
                                 To=self.recipients,
                                 From=self.sender,
                                 Body=self.body,
                                 Html=self.html,
                                 separator=self.recipients_separator)
         if self.cc:
             response.base['Cc'] = self.cc
         for attachment in self.attachments:
             response.attach(attachment.filename,
                             attachment.content_type,
                             attachment.data,
                             attachment.disposition)
         response.update(self.extra_headers)
         return response
     def is_bad_headers(self):
         """
         Checks for bad headers i.e. newlines in subject, sender or recipients.
         """
         headers = [self.subject, self.sender]
         headers += list(self.send_to)
         headers += self.extra_headers.values()
         for val in headers:
             for c in '\r\n':
                 if c in val:
                     return True
         return False
     def validate(self):
         """
         Checks if message is valid and raises appropriate exception.
         """
         if not self.recipients:
             raise InvalidMessage("No recipients have been added")
         if not self.body and not self.html:
             raise InvalidMessage("No body has been set")
         if not self.sender:
             raise InvalidMessage("No sender address has been set")
         if self.is_bad_headers():
             raise BadHeaders
     def add_recipient(self, recipient):
         """
         Adds another recipient to the message.
         :param recipient: email address of recipient.
         """
         self.recipients.append(recipient)
     def add_cc(self, recipient):
         """
         Adds an email address to the CC list.
         :param recipient: email address of recipient.
         """
         self.cc.append(recipient)
     def add_bcc(self, recipient):
         """
         Adds an email address to the BCC list.
         :param recipient: email address of recipient.
         """
         self.bcc.append(recipient)
     def attach(self, attachment):
         """
         Adds an attachment to the message.
         :param attachment: an **Attachment** instance.
         """
         self.attachments.append(attachment)

kallithea/lib/rcmail/smtp_mailer.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.lib.rcmail.smtp_mailer
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Simple smtp mailer used in 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: Sep 13, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 import smtplib
 import time
 from email.utils import formatdate
 from ssl import SSLError
 from kallithea.lib.rcmail.message import Message
 from kallithea.lib.rcmail.utils import DNS_NAME
 class SmtpMailer(object):
     """SMTP mailer class
     mailer = SmtpMailer(mail_from, user, passwd, mail_server, smtp_auth
                         mail_port, ssl, tls)
     mailer.send(recipients, subject, body, attachment_files)
     :param recipients might be a list of string or single string
     :param attachment_files is a dict of {filename:location}
         it tries to guess the mimetype and attach the file
     """
     def __init__(self, mail_from, user, passwd, mail_server, smtp_auth=None,
                  mail_port=None, ssl=False, tls=False, debug=False):
         self.mail_from = mail_from
         self.mail_server = mail_server
         self.mail_port = mail_port
         self.user = user
         self.passwd = passwd
         self.ssl = ssl
         self.tls = tls
         self.debug = debug
         self.auth = smtp_auth
     def send(self, recipients=None, subject='', body='', html='',
              attachment_files=None, headers=None):
         recipients = recipients or []
-        if isinstance(recipients, basestring):
+        if isinstance(recipients, str):
             recipients = [recipients]
         if headers is None:
             headers = {}
         headers.setdefault('Date', formatdate(time.time()))
         msg = Message(subject, recipients, body, html, self.mail_from,
                       recipients_separator=", ", extra_headers=headers)
         raw_msg = msg.to_message()
         if self.ssl:
             smtp_serv = smtplib.SMTP_SSL(self.mail_server, self.mail_port,
                                          local_hostname=DNS_NAME.get_fqdn())
         else:
             smtp_serv = smtplib.SMTP(self.mail_server, self.mail_port,
                                      local_hostname=DNS_NAME.get_fqdn())
         if self.tls:
             smtp_serv.ehlo()
             smtp_serv.starttls()
         if self.debug:
             smtp_serv.set_debuglevel(1)
         smtp_serv.ehlo()
         if self.auth:
             smtp_serv.esmtp_features["auth"] = self.auth
         # if server requires authorization you must provide login and password
         # but only if we have them
         if self.user and self.passwd:
             smtp_serv.login(self.user, self.passwd)
         smtp_serv.sendmail(msg.sender, msg.send_to, raw_msg.as_string())
         logging.info('MAIL SENT TO: %s' % recipients)
         try:
             smtp_serv.quit()
         except SSLError:
             # SSL error might sometimes be raised in tls connections on closing
             smtp_serv.close()

kallithea/lib/utils.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.lib.utils
 ~~~~~~~~~~~~~~~~~~~
 Utilities library 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 18, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import datetime
 import logging
 import os
 import re
 import sys
 import traceback
 from distutils.version import StrictVersion
 import beaker.cache
 import mercurial.config
 import mercurial.ui
 from tg.i18n import ugettext as _
 import kallithea.config.conf
 from kallithea.lib.exceptions import HgsubversionImportError
 from kallithea.lib.utils2 import ascii_bytes, aslist, get_current_authuser, safe_bytes, safe_str, safe_unicode
 from kallithea.lib.vcs.backends.git.repository import GitRepository
 from kallithea.lib.vcs.backends.hg.repository import MercurialRepository
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import RepositoryError, VCSError
 from kallithea.lib.vcs.utils.fakemod import create_module
 from kallithea.lib.vcs.utils.helpers import get_scm
 from kallithea.model import meta
 from kallithea.model.db import RepoGroup, Repository, Setting, Ui, User, UserGroup, UserLog
 log = logging.getLogger(__name__)
 REMOVED_REPO_PAT = re.compile(r'rm__\d{8}_\d{6}_\d{6}_.*')
 #==============================================================================
 # PERM DECORATOR HELPERS FOR EXTRACTING NAMES FOR PERM CHECKS
 #==============================================================================
 def get_repo_slug(request):
     _repo = request.environ['pylons.routes_dict'].get('repo_name')
     if _repo:
         _repo = _repo.rstrip('/')
     return _repo
 def get_repo_group_slug(request):
     _group = request.environ['pylons.routes_dict'].get('group_name')
     if _group:
         _group = _group.rstrip('/')
     return _group
 def get_user_group_slug(request):
     _group = request.environ['pylons.routes_dict'].get('id')
     _group = UserGroup.get(_group)
     if _group:
         return _group.users_group_name
     return None
 def _get_permanent_id(s):
     """Helper for decoding stable URLs with repo ID. For a string like '_123'
     return 123.
     """
     by_id_match = re.match(r'^_(\d+)$', s)
     if by_id_match is None:
         return None
     return int(by_id_match.group(1))
 def fix_repo_id_name(path):
     """
     Rewrite repo_name for _<ID> permanent URLs.
     Given a path, if the first path element is like _<ID>, return the path with
     this part expanded to the corresponding full repo name, else return the
     provided path.
     """
     first, rest = path, ''
     if '/' in path:
         first, rest_ = path.split('/', 1)
         rest = '/' + rest_
     repo_id = _get_permanent_id(first)
     if repo_id is not None:
         repo = Repository.get(repo_id)
         if repo is not None:
             return repo.repo_name + rest
     return path
 def action_logger(user, action, repo, ipaddr='', commit=False):
     """
     Action logger for various actions made by users
     :param user: user that made this action, can be a unique username string or
         object containing user_id attribute
     :param action: action to log, should be on of predefined unique actions for
         easy translations
     :param repo: string name of repository or object containing repo_id,
         that action was made on
     :param ipaddr: optional IP address from what the action was made
     """
     # if we don't get explicit IP address try to get one from registered user
     # in tmpl context var
     if not ipaddr:
         ipaddr = getattr(get_current_authuser(), 'ip_addr', '')
     if getattr(user, 'user_id', None):
         user_obj = User.get(user.user_id)
-    elif isinstance(user, basestring):
+    elif isinstance(user, str):
         user_obj = User.get_by_username(user)
     else:
         raise Exception('You have to provide a user object or a username')
     if getattr(repo, 'repo_id', None):
         repo_obj = Repository.get(repo.repo_id)
         repo_name = repo_obj.repo_name
-    elif isinstance(repo, basestring):
+    elif isinstance(repo, str):
         repo_name = repo.lstrip('/')
         repo_obj = Repository.get_by_repo_name(repo_name)
     else:
         repo_obj = None
         repo_name = u''
     user_log = UserLog()
     user_log.user_id = user_obj.user_id
     user_log.username = user_obj.username
     user_log.action = safe_unicode(action)
     user_log.repository = repo_obj
     user_log.repository_name = repo_name
     user_log.action_date = datetime.datetime.now()
     user_log.user_ip = ipaddr
     meta.Session().add(user_log)
     log.info('Logging action:%s on %s by user:%s ip:%s',
              action, safe_unicode(repo), user_obj, ipaddr)
     if commit:
         meta.Session().commit()
 def get_filesystem_repos(path):
     """
     Scans given path for repos and return (name,(type,path)) tuple
     :param path: path to scan for repositories
     :param recursive: recursive search and return names with subdirs in front
     """
     # remove ending slash for better results
     path = safe_str(path.rstrip(os.sep))
     log.debug('now scanning in %s', path)
     def isdir(*n):
         return os.path.isdir(os.path.join(*n))
     for root, dirs, _files in os.walk(path):
         recurse_dirs = []
         for subdir in dirs:
             # skip removed repos
             if REMOVED_REPO_PAT.match(subdir):
                 continue
             # skip .<something> dirs TODO: rly? then we should prevent creating them ...
             if subdir.startswith('.'):
                 continue
             cur_path = os.path.join(root, subdir)
             if isdir(cur_path, '.git'):
                 log.warning('ignoring non-bare Git repo: %s', cur_path)
                 continue
             if (isdir(cur_path, '.hg') or
                 isdir(cur_path, '.svn') or
                 isdir(cur_path, 'objects') and (isdir(cur_path, 'refs') or
                                                 os.path.isfile(os.path.join(cur_path, 'packed-refs')))):
                 if not os.access(cur_path, os.R_OK) or not os.access(cur_path, os.X_OK):
                     log.warning('ignoring repo path without access: %s', cur_path)
                     continue
                 if not os.access(cur_path, os.W_OK):
                     log.warning('repo path without write access: %s', cur_path)
                 try:
                     scm_info = get_scm(cur_path)
                     assert cur_path.startswith(path)
                     repo_path = cur_path[len(path) + 1:]
                     yield repo_path, scm_info
                     continue # no recursion
                 except VCSError:
                     # We should perhaps ignore such broken repos, but especially
                     # the bare git detection is unreliable so we dive into it
                     pass
             recurse_dirs.append(subdir)
         dirs[:] = recurse_dirs
 def is_valid_repo_uri(repo_type, url, ui):
     """Check if the url seems like a valid remote repo location - raise an Exception if any problems"""
     if repo_type == 'hg':
         if url.startswith('http') or url.startswith('ssh'):
             # initially check if it's at least the proper URL
             # or does it pass basic auth
             MercurialRepository._check_url(url, ui)
         elif url.startswith('svn+http'):
             try:
                 from hgsubversion.svnrepo import svnremoterepo
             except ImportError:
                 raise HgsubversionImportError(_('Unable to activate hgsubversion support. '
                                                 'The "hgsubversion" library is missing'))
             svnremoterepo(ui, url).svn.uuid
         elif url.startswith('git+http'):
             raise NotImplementedError()
         else:
             raise Exception('URI %s not allowed' % (url,))
     elif repo_type == 'git':
         if url.startswith('http') or url.startswith('git'):
             # initially check if it's at least the proper URL
             # or does it pass basic auth
             GitRepository._check_url(url)
         elif url.startswith('svn+http'):
             raise NotImplementedError()
         elif url.startswith('hg+http'):
             raise NotImplementedError()
         else:
             raise Exception('URI %s not allowed' % (url))
 def is_valid_repo(repo_name, base_path, scm=None):
     """
     Returns True if given path is a valid repository False otherwise.
     If scm param is given also compare if given scm is the same as expected
     from scm parameter
     :param repo_name:
     :param base_path:
     :param scm:
     :return True: if given path is a valid repository
     """
     # TODO: paranoid security checks?
     full_path = os.path.join(safe_str(base_path), safe_str(repo_name))
     try:
         scm_ = get_scm(full_path)
         if scm:
             return scm_[0] == scm
         return True
     except VCSError:
         return False
 def is_valid_repo_group(repo_group_name, base_path, skip_path_check=False):
     """
     Returns True if given path is a repository group False otherwise
     :param repo_name:
     :param base_path:
     """
     full_path = os.path.join(safe_str(base_path), safe_str(repo_group_name))
     # check if it's not a repo
     if is_valid_repo(repo_group_name, base_path):
         return False
     try:
         # we need to check bare git repos at higher level
         # since we might match branches/hooks/info/objects or possible
         # other things inside bare git repo
         get_scm(os.path.dirname(full_path))
         return False
     except VCSError:
         pass
     # check if it's a valid path
     if skip_path_check or os.path.isdir(full_path):
         return True
     return False
 # propagated from mercurial documentation
 ui_sections = ['alias', 'auth',
                 'decode/encode', 'defaults',
                 'diff', 'email',
                 'extensions', 'format',
                 'merge-patterns', 'merge-tools',
                 'hooks', 'http_proxy',
                 'smtp', 'patch',
                 'paths', 'profiling',
                 'server', 'trusted',
                 'ui', 'web', ]
 def make_ui(repo_path=None):
     """
     Create an Mercurial 'ui' object based on database Ui settings, possibly
     augmenting with content from a hgrc file.
     """
     baseui = mercurial.ui.ui()
     # clean the baseui object
     baseui._ocfg = mercurial.config.config()
     baseui._ucfg = mercurial.config.config()
     baseui._tcfg = mercurial.config.config()

kallithea/lib/utils2.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.lib.utils2
 ~~~~~~~~~~~~~~~~~~~~
 Some simple helper functions.
 Note: all these functions should be independent of Kallithea classes, i.e.
 models, controllers, etc.  to prevent import cycles.
 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: Jan 5, 2011
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 from __future__ import print_function
 import binascii
 import datetime
 import json
 import os
 import pwd
 import re
 import time
 import urllib
 import urlobject
 from tg.i18n import ugettext as _
 from tg.i18n import ungettext
 from webhelpers2.text import collapse, remove_formatting, strip_tags
 from kallithea.lib.vcs.utils import ascii_bytes, ascii_str, safe_bytes, safe_str, safe_unicode  # re-export
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 def str2bool(_str):
     """
     returns True/False value from given string, it tries to translate the
     string into boolean
     :param _str: string value to translate into boolean
     :rtype: boolean
     :returns: boolean from given string
     """
     if _str is None:
         return False
     if _str in (True, False):
         return _str
     _str = str(_str).strip().lower()
     return _str in ('t', 'true', 'y', 'yes', 'on', '1')
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
-    if isinstance(obj, (basestring)):
+    if isinstance(obj, (str)):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
     elif isinstance(obj, (list, tuple)):
         return obj
     elif obj is None:
         return []
     else:
         return [obj]
 def convert_line_endings(line, mode):
     """
     Converts a given line  "line end" according to given mode
     Available modes are::
 - Unix
 - Mac
 - DOS
     :param line: given line to convert
     :param mode: mode to convert to
     :rtype: str
     :return: converted line according to mode
     """
     if mode == 0:
         line = line.replace('\r\n', '\n')
         line = line.replace('\r', '\n')
     elif mode == 1:
         line = line.replace('\r\n', '\r')
         line = line.replace('\n', '\r')
     elif mode == 2:
         line = re.sub("\r(?!\n)|(?<!\r)\n", "\r\n", line)
     return line
 def detect_mode(line, default):
     """
     Detects line break for given line, if line break couldn't be found
     given default value is returned
     :param line: str line
     :param default: default
     :rtype: int
     :return: value of line end on of 0 - Unix, 1 - Mac, 2 - DOS
     """
     if line.endswith('\r\n'):
         return 2
     elif line.endswith('\n'):
         return 0
     elif line.endswith('\r'):
         return 1
     else:
         return default
 def generate_api_key():
     """
     Generates a random (presumably unique) API key.
     This value is used in URLs and "Bearer" HTTP Authorization headers,
     which in practice means it should only contain URL-safe characters
     (RFC 3986):
         unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~"
     """
     # Hexadecimal certainly qualifies as URL-safe.
     return ascii_str(binascii.hexlify(os.urandom(20)))
 def safe_int(val, default=None):
     """
     Returns int() of val if val is not convertable to int use default
     instead
     :param val:
     :param default:
     """
     try:
         val = int(val)
     except (ValueError, TypeError):
         val = default
     return val
 def remove_suffix(s, suffix):
     if s.endswith(suffix):
         s = s[:-1 * len(suffix)]
     return s
 def remove_prefix(s, prefix):
     if s.startswith(prefix):
         s = s[len(prefix):]
     return s
 def age(prevdate, show_short_version=False, now=None):
     """
     turns a datetime into an age string.
     If show_short_version is True, then it will generate a not so accurate but shorter string,
     example: 2days ago, instead of 2 days and 23 hours ago.
     :param prevdate: datetime object
     :param show_short_version: if it should approximate the date and return a shorter string
     :rtype: unicode
     :returns: unicode words describing age
     """
     now = now or datetime.datetime.now()
     order = ['year', 'month', 'day', 'hour', 'minute', 'second']
     deltas = {}
     future = False
     if prevdate > now:
         now, prevdate = prevdate, now
         future = True
     if future:
         prevdate = prevdate.replace(microsecond=0)
     # Get date parts deltas
     from dateutil import relativedelta
     for part in order:
         d = relativedelta.relativedelta(now, prevdate)
         deltas[part] = getattr(d, part + 's')
     # Fix negative offsets (there is 1 second between 10:59:59 and 11:00:00,
     # not 1 hour, -59 minutes and -59 seconds)
     for num, length in [(5, 60), (4, 60), (3, 24)]:  # seconds, minutes, hours
         part = order[num]
         carry_part = order[num - 1]
         if deltas[part] < 0:
             deltas[part] += length
             deltas[carry_part] -= 1
     # Same thing for days except that the increment depends on the (variable)
     # number of days in the month
     month_lengths = [31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
     if deltas['day'] < 0:
         if prevdate.month == 2 and (prevdate.year % 4 == 0 and
             (prevdate.year % 100 != 0 or prevdate.year % 400 == 0)
         ):
             deltas['day'] += 29
         else:
             deltas['day'] += month_lengths[prevdate.month - 1]
         deltas['month'] -= 1
     if deltas['month'] < 0:
         deltas['month'] += 12
         deltas['year'] -= 1
     # In short version, we want nicer handling of ages of more than a year
     if show_short_version:
         if deltas['year'] == 1:
             # ages between 1 and 2 years: show as months
             deltas['month'] += 12
             deltas['year'] = 0
         if deltas['year'] >= 2:
             # ages 2+ years: round
             if deltas['month'] > 6:
                 deltas['year'] += 1
                 deltas['month'] = 0
     # Format the result
     fmt_funcs = {
         'year': lambda d: ungettext(u'%d year', '%d years', d) % d,
         'month': lambda d: ungettext(u'%d month', '%d months', d) % d,
         'day': lambda d: ungettext(u'%d day', '%d days', d) % d,
         'hour': lambda d: ungettext(u'%d hour', '%d hours', d) % d,
         'minute': lambda d: ungettext(u'%d minute', '%d minutes', d) % d,
         'second': lambda d: ungettext(u'%d second', '%d seconds', d) % d,
+    }
     for i, part in enumerate(order):
         value = deltas[part]
         if value == 0:
             continue
         if i < 5:
             sub_part = order[i + 1]
             sub_value = deltas[sub_part]
         else:
             sub_value = 0
         if sub_value == 0 or show_short_version:
             if future:
                 return _('in %s') % fmt_funcs[part](value)
             else:
                 return _('%s ago') % fmt_funcs[part](value)
         if future:
             return _('in %s and %s') % (fmt_funcs[part](value),
                 fmt_funcs[sub_part](sub_value))
         else:
             return _('%s and %s ago') % (fmt_funcs[part](value),
                 fmt_funcs[sub_part](sub_value))
     return _('just now')
 def uri_filter(uri):
     """
     Removes user:password from given url string
     :param uri:
     :rtype: unicode
     :returns: filtered list of strings
     """
     if not uri:
         return []
     proto = ''
     for pat in ('https://', 'http://', 'git://'):
         if uri.startswith(pat):
             uri = uri[len(pat):]
             proto = pat
             break
     # remove passwords and username
     uri = uri[uri.find('@') + 1:]
     # get the port
     cred_pos = uri.find(':')
     if cred_pos == -1:
         host, port = uri, None
     else:
         host, port = uri[:cred_pos], uri[cred_pos + 1:]
     return [_f for _f in [proto, host, port] if _f]
 def credentials_filter(uri):
     """
     Returns a url with removed credentials
     :param uri:
     """
     uri = uri_filter(uri)
     # check if we have port
     if len(uri) > 2 and uri[2]:
         uri[2] = ':' + uri[2]
     return ''.join(uri)
 def get_clone_url(clone_uri_tmpl, prefix_url, repo_name, repo_id, username=None):
     parsed_url = urlobject.URLObject(prefix_url)
     prefix = safe_unicode(urllib.unquote(parsed_url.path.rstrip('/')))
     try:
         system_user = pwd.getpwuid(os.getuid()).pw_name
     except Exception: # TODO: support all systems - especially Windows
         system_user = 'kallithea' # hardcoded default value ...
     args = {
         'scheme': parsed_url.scheme,
         'user': safe_unicode(urllib.quote(safe_str(username or ''))),
         'netloc': parsed_url.netloc + prefix,  # like "hostname:port/prefix" (with optional ":port" and "/prefix")
         'prefix': prefix, # undocumented, empty or starting with /
         'repo': repo_name,
         'repoid': str(repo_id),
         'system_user': safe_unicode(system_user),
         'hostname': parsed_url.hostname,
+    }
     url = re.sub('{([^{}]+)}', lambda m: args.get(m.group(1), m.group(0)), clone_uri_tmpl)
     # remove leading @ sign if it's present. Case of empty user
     url_obj = urlobject.URLObject(url)
     if not url_obj.username:
         url_obj = url_obj.with_username(None)
     return safe_unicode(url_obj)
 def get_changeset_safe(repo, rev):
     """
     Safe version of get_changeset if this changeset doesn't exists for a
     repo it returns a Dummy one instead
     :param repo:
     :param rev:
     """
     from kallithea.lib.vcs.backends.base import BaseRepository
     from kallithea.lib.vcs.exceptions import RepositoryError
     from kallithea.lib.vcs.backends.base import EmptyChangeset
     if not isinstance(repo, BaseRepository):
         raise Exception('You must pass an Repository '
                         'object as first argument got %s' % type(repo))
     try:
         cs = repo.get_changeset(rev)
     except (RepositoryError, LookupError):
         cs = EmptyChangeset(requested_revision=rev)
     return cs
 def datetime_to_time(dt):
     if dt:
         return time.mktime(dt.timetuple())
 def time_to_datetime(tm):
     if tm:
-        if isinstance(tm, basestring):
+        if isinstance(tm, str):
             try:
                 tm = float(tm)
             except ValueError:
                 return
         return datetime.datetime.fromtimestamp(tm)
 # Must match regexp in kallithea/public/js/base.js MentionsAutoComplete()
 # Check char before @ - it must not look like we are in an email addresses.
 # Matching is greedy so we don't have to look beyond the end.
 MENTIONS_REGEX = re.compile(r'(?:^|(?<=[^a-zA-Z0-9]))@([a-zA-Z0-9][-_.a-zA-Z0-9]*[a-zA-Z0-9])')
 def extract_mentioned_usernames(text):
     r"""
     Returns list of (possible) usernames @mentioned in given text.
     >>> extract_mentioned_usernames('@1-2.a_X,@1234 not@not @ddd@not @n @ee @ff @gg, @gg;@hh @n\n@zz,')
     ['1-2.a_X', '1234', 'ddd', 'ee', 'ff', 'gg', 'gg', 'hh', 'zz']
     """
     return MENTIONS_REGEX.findall(text)
 def extract_mentioned_users(text):
     """ Returns set of actual database Users @mentioned in given text. """
     from kallithea.model.db import User
     result = set()
     for name in extract_mentioned_usernames(text):
         user = User.get_by_username(name, case_insensitive=True)
         if user is not None and not user.is_default_user:
             result.add(user)
     return result
 class AttributeDict(dict):
     def __getattr__(self, attr):
         return self.get(attr, None)
     __setattr__ = dict.__setitem__
     __delattr__ = dict.__delitem__
 def obfuscate_url_pw(engine):
     from sqlalchemy.engine import url as sa_url
     from sqlalchemy.exc import ArgumentError
     try:
         _url = sa_url.make_url(engine or '')
     except ArgumentError:
         return engine
     if _url.password:
         _url.password = 'XXXXX'
     return str(_url)
 class HookEnvironmentError(Exception): pass
 def get_hook_environment():
     """
     Get hook context by deserializing the global KALLITHEA_EXTRAS environment
     variable.
     Called early in Git out-of-process hooks to get .ini config path so the
     basic environment can be configured properly. Also used in all hooks to get
     information about the action that triggered it.
     """
     try:
         kallithea_extras = os.environ['KALLITHEA_EXTRAS']
     except KeyError:
         raise HookEnvironmentError("Environment variable KALLITHEA_EXTRAS not found")
     extras = json.loads(kallithea_extras)
     for k in ['username', 'repository', 'scm', 'action', 'ip', 'config']:
         try:
             extras[k]
         except KeyError:
             raise HookEnvironmentError('Missing key %s in KALLITHEA_EXTRAS %s' % (k, extras))
     return AttributeDict(extras)
 def set_hook_environment(username, ip_addr, repo_name, repo_alias, action=None):
     """Prepare global context for running hooks by serializing data in the
     global KALLITHEA_EXTRAS environment variable.
     Most importantly, this allow Git hooks to do proper logging and updating of
     caches after pushes.
     Must always be called before anything with hooks are invoked.
     """
     from kallithea import CONFIG
     extras = {
         'ip': ip_addr, # used in log_push/pull_action action_logger
         'username': username,
         'action': action or 'push_local', # used in log_push_action_raw_ids action_logger
         'repository': repo_name,
         'scm': repo_alias, # used to pick hack in log_push_action_raw_ids
         'config': CONFIG['__file__'], # used by git hook to read config
+    }
     os.environ['KALLITHEA_EXTRAS'] = json.dumps(extras)
 def get_current_authuser():
     """
     Gets kallithea user from threadlocal tmpl_context variable if it's
     defined, else returns None.
     """
     from tg import tmpl_context
     try:
         return getattr(tmpl_context, 'authuser', None)
     except TypeError:  # No object (name: context) has been registered for this thread
         return None
 class OptionalAttr(object):
     """
     Special Optional Option that defines other attribute. Example::
         def test(apiuser, userid=Optional(OAttr('apiuser')):
             user = Optional.extract(userid)
             # calls
     """
     def __init__(self, attr_name):
         self.attr_name = attr_name
     def __repr__(self):
         return '<OptionalAttr:%s>' % self.attr_name
     def __call__(self):
         return self
 # alias
 OAttr = OptionalAttr
 class Optional(object):
     """
     Defines an optional parameter::
         param = param.getval() if isinstance(param, Optional) else param
         param = param() if isinstance(param, Optional) else param
     is equivalent of::
         param = Optional.extract(param)
     """
     def __init__(self, type_):
         self.type_ = type_
     def __repr__(self):
         return '<Optional:%s>' % self.type_.__repr__()
     def __call__(self):
         return self.getval()
     def getval(self):
         """
         returns value from this Optional instance
         """
         if isinstance(self.type_, OAttr):
             # use params name
             return self.type_.attr_name
         return self.type_
     @classmethod
     def extract(cls, val):
         """
         Extracts value from Optional() instance
         :param val:
         :return: original value if it's not Optional instance else
             value of instance
         """
         if isinstance(val, cls):
             return val.getval()
         return val
 def urlreadable(s, _cleanstringsub=re.compile('[^-a-zA-Z0-9./]+').sub):
     return _cleanstringsub('_', safe_str(s)).rstrip('_')
 def recursive_replace(str_, replace=' '):
     """
     Recursive replace of given sign to just one instance
     :param str_: given string

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

➞

Show inline comments

@@ @@ -511,226 +511,226 @@ class GitRepository(BaseRepository): @@
         # 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, basestring):
+        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:
                 return config.get(section, name)
             except KeyError:
                 continue
         return None
     def get_user_name(self, config_file=None):
         """
         Returns user's name from global configuration file.
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'name', config_file)
     def get_user_email(self, config_file=None):
         """
         Returns user's email from global configuration file.
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'email', config_file)

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

➞

Show inline comments

 import os
 import posixpath
 import mercurial.archival
 import mercurial.node
 import mercurial.obsutil
 from kallithea.lib.vcs.backends.base import BaseChangeset
 from kallithea.lib.vcs.conf import settings
 from kallithea.lib.vcs.exceptions import ChangesetDoesNotExistError, ChangesetError, ImproperArchiveTypeError, NodeDoesNotExistError, VCSError
 from kallithea.lib.vcs.nodes import (
     AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, RemovedFileNodesGenerator, RootNode, SubModuleNode)
 from kallithea.lib.vcs.utils import ascii_bytes, ascii_str, date_fromtimestamp, safe_bytes, safe_str, safe_unicode
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.lib.vcs.utils.paths import get_dirs_for_path
 class MercurialChangeset(BaseChangeset):
     """
     Represents state of the repository at a revision.
     """
     def __init__(self, repository, revision):
         self.repository = repository
-        assert isinstance(revision, basestring), repr(revision)
+        assert isinstance(revision, str), repr(revision)
         self._ctx = repository._repo[ascii_bytes(revision)]
         self.raw_id = ascii_str(self._ctx.hex())
         self.revision = self._ctx._rev
         self.nodes = {}
     @LazyProperty
     def tags(self):
         return [safe_unicode(tag) for tag in self._ctx.tags()]
     @LazyProperty
     def branch(self):
         return safe_unicode(self._ctx.branch())
     @LazyProperty
     def branches(self):
         return [safe_unicode(self._ctx.branch())]
     @LazyProperty
     def closesbranch(self):
         return self._ctx.closesbranch()
     @LazyProperty
     def obsolete(self):
         return self._ctx.obsolete()
     @LazyProperty
     def bumped(self):
         return self._ctx.phasedivergent()
     @LazyProperty
     def divergent(self):
         return self._ctx.contentdivergent()
     @LazyProperty
     def extinct(self):
         return self._ctx.extinct()
     @LazyProperty
     def unstable(self):
         return self._ctx.orphan()
     @LazyProperty
     def phase(self):
         if(self._ctx.phase() == 1):
             return 'Draft'
         elif(self._ctx.phase() == 2):
             return 'Secret'
         else:
             return ''
     @LazyProperty
     def successors(self):
         successors = mercurial.obsutil.successorssets(self._ctx._repo, self._ctx.node(), closest=True)
         if successors:
             # flatten the list here handles both divergent (len > 1)
             # and the usual case (len = 1)
             successors = [mercurial.node.hex(n)[:12] for sub in successors for n in sub if n != self._ctx.node()]
         return successors
     @LazyProperty
     def predecessors(self):
         return [mercurial.node.hex(n)[:12] for n in mercurial.obsutil.closestpredecessors(self._ctx._repo, self._ctx.node())]
     @LazyProperty
     def bookmarks(self):
         return [safe_unicode(bookmark) for bookmark in self._ctx.bookmarks()]
     @LazyProperty
     def message(self):
         return safe_unicode(self._ctx.description())
     @LazyProperty
     def committer(self):
         return safe_unicode(self.author)
     @LazyProperty
     def author(self):
         return safe_unicode(self._ctx.user())
     @LazyProperty
     def date(self):
         return date_fromtimestamp(*self._ctx.date())
     @LazyProperty
     def _timestamp(self):
         return self._ctx.date()[0]
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.repository._repo.status(self._ctx.p1().node(),
                                             self._ctx.node())
     @LazyProperty
     def _file_paths(self):
         return list(self._ctx)
     @LazyProperty
     def _dir_paths(self):
         p = list(set(get_dirs_for_path(*self._file_paths)))
         p.insert(0, '')
         return p
     @LazyProperty
     def _paths(self):
         return self._dir_paths + self._file_paths
     @LazyProperty
     def short_id(self):
         return self.raw_id[:12]
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         return [self.repository.get_changeset(parent.rev())
                 for parent in self._ctx.parents() if parent.rev() >= 0]
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         return [self.repository.get_changeset(child.rev())
                 for child in self._ctx.children() if child.rev() >= 0]
     def next(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 next_ = cs.repository.revisions.index(cs.raw_id) + 1
                 next_rev = cs.repository.revisions[next_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(next_rev)
             if not branch or branch == cs.branch:
                 return cs
     def prev(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
         cs = self
         while True:
             try:
                 prev_ = cs.repository.revisions.index(cs.raw_id) - 1
                 if prev_ < 0:
                     raise IndexError
                 prev_rev = cs.repository.revisions[prev_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = cs.repository.get_changeset(prev_rev)
             if not branch or branch == cs.branch:
                 return cs
     def diff(self):
         # Only used to feed diffstat
         return b''.join(self._ctx.diff())
     def _fix_path(self, path):
         """
         Paths are stored without trailing slash so we need to get rid off it if
         needed. Also mercurial keeps filenodes as str so we need to decode
         from unicode to str
         """
         if path.endswith('/'):
             path = path.rstrip('/')
         return safe_str(path)
     def _get_kind(self, path):
         path = self._fix_path(path)
         if path in self._file_paths:
             return NodeKind.FILE
         elif path in self._dir_paths:
             return NodeKind.DIR
         else:
             raise ChangesetError("Node does not exist at the given path '%s'"
                 % (path))
     def _get_filectx(self, path):

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

➞

Show inline comments

@@ @@ -394,225 +394,225 @@ class MercurialRepository(BaseRepository @@
     @LazyProperty
     def contact(self):
         return safe_unicode(mercurial.hgweb.common.get_contact(self._repo.ui.config)
                             or b'Unknown')
     @LazyProperty
     def last_change(self):
         """
         Returns last change made on this repository as datetime object
         """
         return date_fromtimestamp(self._get_mtime(), makedate()[1])
     def _get_mtime(self):
         try:
             return time.mktime(self.get_changeset().date.timetuple())
         except RepositoryError:
             # fallback to filesystem
             cl_path = os.path.join(self.path, '.hg', "00changelog.i")
             st_path = os.path.join(self.path, '.hg', "store")
             if os.path.exists(cl_path):
                 return os.stat(cl_path).st_mtime
             else:
                 return os.stat(st_path).st_mtime
     def _get_revision(self, revision):
         """
         Given any revision identifier, returns a 40 char string with revision hash.
         :param revision: str or int or None
         """
         if self._empty:
             raise EmptyRepositoryError("There are no changesets yet")
         if revision in [-1, None]:
             revision = b'tip'
         elif isinstance(revision, unicode):
             revision = safe_bytes(revision)
         try:
             if isinstance(revision, int):
                 return ascii_str(self._repo[revision].hex())
             return ascii_str(mercurial.scmutil.revsymbol(self._repo, revision).hex())
         except (IndexError, ValueError, mercurial.error.RepoLookupError, TypeError):
             msg = "Revision %r does not exist for %s" % (safe_unicode(revision), self.name)
             raise ChangesetDoesNotExistError(msg)
         except (LookupError, ):
             msg = "Ambiguous identifier `%s` for %s" % (safe_unicode(revision), self.name)
             raise ChangesetDoesNotExistError(msg)
     def get_ref_revision(self, ref_type, ref_name):
         """
         Returns revision number for the given reference.
         """
         ref_name = safe_str(ref_name)
         if ref_type == 'rev' and not ref_name.strip('0'):
             return self.EMPTY_CHANGESET
         # lookup up the exact node id
         _revset_predicates = {
                 'branch': 'branch',
                 'book': 'bookmark',
                 'tag': 'tag',
                 'rev': 'id',
+            }
         # avoid expensive branch(x) iteration over whole repo
         rev_spec = "%%s & %s(%%s)" % _revset_predicates[ref_type]
         try:
             revs = self._repo.revs(rev_spec, ref_name, ref_name)
         except LookupError:
             msg = "Ambiguous identifier %s:%s for %s" % (ref_type, ref_name, self.name)
             raise ChangesetDoesNotExistError(msg)
         except mercurial.error.RepoLookupError:
             msg = "Revision %s:%s does not exist for %s" % (ref_type, ref_name, self.name)
             raise ChangesetDoesNotExistError(msg)
         if revs:
             revision = revs.last()
         else:
             # TODO: just report 'not found'?
             revision = ref_name
         return self._get_revision(revision)
     def _get_archives(self, archive_name='tip'):
         allowed = self.baseui.configlist(b"web", b"allow_archive",
                                          untrusted=True)
         for name, ext in [(b'zip', '.zip'), (b'gz', '.tar.gz'), (b'bz2', '.tar.bz2')]:
             if name in allowed or self._repo.ui.configbool(b"web",
                                                            b"allow" + name,
                                                            untrusted=True):
                 yield {"type": name, "extension": ext, "node": archive_name}
     def _get_url(self, url):
         """
         Returns normalized url. If schema is not given, would fall
         to filesystem
         (``file:///``) schema.
         """
         url = safe_str(url)
         if url != 'default' and '://' not in url:
             url = "file:" + urllib.pathname2url(url)
         return url
     def get_changeset(self, revision=None):
         """
         Returns ``MercurialChangeset`` object representing repository's
         changeset at the given ``revision``.
         """
         return MercurialChangeset(repository=self, revision=self._get_revision(revision))
     def get_changesets(self, start=None, end=None, start_date=None,
                        end_date=None, branch_name=None, reverse=False, max_revisions=None):
         """
         Returns iterator of ``MercurialChangeset`` objects from start to end
         (both are inclusive)
         :param start: None, str, int or mercurial lookup format
         :param end:  None, str, int or mercurial lookup format
         :param start_date:
         :param end_date:
         :param branch_name:
         :param reversed: return changesets in reversed order
         """
         start_raw_id = self._get_revision(start)
         start_pos = None if start is None else self.revisions.index(start_raw_id)
         end_raw_id = self._get_revision(end)
         end_pos = None if end is None else self.revisions.index(end_raw_id)
         if start_pos is not None and end_pos is not None and start_pos > end_pos:
             raise RepositoryError("Start revision '%s' cannot be "
                                   "after end revision '%s'" % (start, end))
         if branch_name and branch_name not in self.allbranches:
             msg = "Branch %r not found in %s" % (branch_name, self.name)
             raise BranchDoesNotExistError(msg)
         if end_pos is not None:
             end_pos += 1
         # filter branches
         filter_ = []
         if branch_name:
             filter_.append(b'branch("%s")' % safe_bytes(branch_name))
         if start_date:
             filter_.append(b'date(">%s")' % safe_bytes(str(start_date)))
         if end_date:
             filter_.append(b'date("<%s")' % safe_bytes(str(end_date)))
         if filter_ or max_revisions:
             if filter_:
                 revspec = b' and '.join(filter_)
             else:
                 revspec = b'all()'
             if max_revisions:
                 revspec = b'limit(%s, %d)' % (revspec, max_revisions)
             revisions = mercurial.scmutil.revrange(self._repo, [revspec])
         else:
             revisions = self.revisions
         # this is very much a hack to turn this into a list; a better solution
         # would be to get rid of this function entirely and use revsets
         revs = list(revisions)[start_pos:end_pos]
         if reverse:
             revs.reverse()
         return CollectionGenerator(self, revs)
     def pull(self, url):
         """
         Tries to pull changes from external location.
         """
         other = mercurial.hg.peer(self._repo, {}, safe_bytes(self._get_url(url)))
         try:
             mercurial.exchange.pull(self._repo, other, heads=None, force=None)
         except mercurial.error.Abort as err:
             # Propagate error but with vcs's type
             raise RepositoryError(str(err))
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         return MercurialWorkdir(self)
     def get_config_value(self, section, name=None, config_file=None):
         """
         Returns configuration value for a given [``section``] and ``name``.
         :param section: Section we want to retrieve value from
         :param name: Name of configuration we want to retrieve
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         if config_file is None:
             config_file = []
-        elif isinstance(config_file, basestring):
+        elif isinstance(config_file, str):
             config_file = [config_file]
         config = self._repo.ui
         if config_file:
             config = mercurial.ui.ui()
             for path in config_file:
                 config.readconfig(safe_bytes(path))
         return config.config(safe_bytes(section), safe_bytes(name))
     def get_user_name(self, config_file=None):
         """
         Returns user's name from global configuration file.
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         username = self.get_config_value('ui', 'username', config_file=config_file)
         if username:
             return author_name(username)
         return None
     def get_user_email(self, config_file=None):
         """
         Returns user's email from global configuration file.
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         username = self.get_config_value('ui', 'username', config_file=config_file)
         if username:
             return author_email(username)
         return None

kallithea/lib/vcs/utils/__init__.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
 This module provides some useful tools for ``vcs`` like annotate/diff html
 output. It also includes some internal helpers.
 """
 import datetime
 import re
 import time
 def makedate():
     lt = time.localtime()
     if lt[8] == 1 and time.daylight:
         tz = time.altzone
     else:
         tz = time.timezone
     return time.mktime(lt), tz
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
-    if isinstance(obj, basestring):
+    if isinstance(obj, str):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
     elif isinstance(obj, (list, tuple)):
         return obj
     elif obj is None:
         return []
     else:
         return [obj]
 def date_fromtimestamp(unixts, tzoffset=0):
     """
     Makes a local datetime object out of unix timestamp
     :param unixts:
     :param tzoffset:
     """
     return datetime.datetime.fromtimestamp(float(unixts))
 def safe_int(val, default=None):
     """
     Returns int() of val if val is not convertible to int use default
     instead
     :param val:
     :param default:
     """
     try:
         val = int(val)
     except (ValueError, TypeError):
         val = default
     return val
 def safe_unicode(s):
     """
     Safe unicode function. Use a few tricks to turn s into unicode string:
     In case of UnicodeDecodeError with configured default encodings, try to
     detect encoding with chardet library, then fall back to first encoding with
     errors replaced.
     """
     if isinstance(s, unicode):
         return s
     if not isinstance(s, bytes):  # use __str__ / __unicode__ and don't expect UnicodeDecodeError
         return unicode(s)
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return unicode(s, enc)
         except UnicodeDecodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(s)['encoding']
         if encoding is not None:
             return s.decode(encoding)
     except (ImportError, UnicodeDecodeError):
         pass
     return unicode(s, settings.DEFAULT_ENCODINGS[0], 'replace')
 def safe_bytes(s):
     """
     Safe bytes function. Use a few tricks to turn s into bytes string:
     In case of UnicodeEncodeError with configured default encodings, fall back
     to first configured encoding with errors replaced.
     """
     if isinstance(s, bytes):
         return s
     assert isinstance(s, unicode), repr(s)  # bytes cannot coerse with __str__ or handle None or int
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return s.encode(enc)
         except UnicodeEncodeError:
             pass
     return s.encode(settings.DEFAULT_ENCODINGS[0], 'replace')
 safe_str = safe_bytes  # safe_str is deprecated - it will be redefined when changing to py3
 def ascii_bytes(s):
     """
     Simple conversion from unicode/str to bytes, *assuming* all codepoints are
 -bit and it thus is pure ASCII.
     Will fail badly with UnicodeError on invalid input.
     This should be used where enocding and "safe" ambiguity should be avoided.
     Where strings already have been encoded in other ways but still are unicode
     string - for example to hex, base64, json, urlencoding, or are known to be
     identifiers.
     >>> ascii_bytes('a')
     'a'
     >>> ascii_bytes(u'a')
     'a'
     >>> ascii_bytes('å')
     Traceback (most recent call last):
     UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 0: ordinal not in range(128)
     >>> ascii_bytes(u'å')
     Traceback (most recent call last):
     UnicodeEncodeError: 'ascii' codec can't encode characters in position 0-1: ordinal not in range(128)
     """
     assert isinstance(s, (unicode, str)), repr(s)
     return s.encode('ascii')
 def ascii_str(s):
     r"""
     Simple conversion from bytes to str, *assuming* all codepoints are
 -bit and it thus is pure ASCII.
     Will fail badly with UnicodeError on invalid input.
     This should be used where enocding and "safe" ambiguity should be avoided.
     Where strings are encoded but also in other ways are known to be ASCII, and
     where a unicode string is wanted without caring about encoding. For example
     to hex, base64, urlencoding, or are known to be identifiers.
     >>> ascii_str('a')
     'a'
     >>> ascii_str(u'a')
     Traceback (most recent call last):
     AssertionError: u'a'
     >>> ascii_str('å')
     Traceback (most recent call last):
     UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 0: ordinal not in range(128)
     >>> ascii_str(u'å')
     Traceback (most recent call last):
     AssertionError: u'\xc3\xa5'
     """
     assert isinstance(s, bytes), repr(s)
     # Note: we use "encode", even though we really *should* use "decode". But
     # we are in py2 and don't want py2, and encode is doing what we need for the
     # ascii subset.
     return s.encode('ascii')
 # Regex taken from http://www.regular-expressions.info/email.html
 email_re = re.compile(
     r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@"""
     r"""(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?""",
     re.IGNORECASE)
 def author_email(author):
     """
     Returns email address of given author string.
     If author contains <> brackets, only look inside that.
     If any RFC valid email address is found, return that.
     Else, return empty string.
     """
     if not author:
         return ''
     l = author.find('<') + 1
     if l != 0:
         r = author.find('>', l)
         if r != -1:
             author = author[l:r]
     m = email_re.search(author)
     if m is None:
         return ''
     return safe_str(m.group(0))
 def author_name(author):
     """
     get name of author, or else username.
     It'll try to find an email in the author string and just cut it off
     to get the username
     """
     if not author:
         return ''
     if '@' not in author:
         return author
     return author.replace(author_email(author), '').replace('<', '') \
         .replace('>', '').strip()

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, safe_unicode, 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, basestring) and value.isdigit():
+        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_unicode,
         '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:
         """
         self._app_settings_value = safe_unicode(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': safe_unicode(platform.platform()),
             'kallithea_version': kallithea.__version__,
             'git_version': safe_unicode(check_git_version()),
             'git_path': kallithea.CONFIG.get('git_path')
+        }
         return info

kallithea/model/scm.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.scm
 ~~~~~~~~~~~~~~~~~~~
 Scm model 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 9, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 import logging
 import os
 import posixpath
 import re
 import sys
 import traceback
 import pkg_resources
 from tg.i18n import ugettext as _
 import kallithea
 from kallithea import BACKENDS
 from kallithea.lib.auth import HasPermissionAny, HasRepoGroupPermissionLevel, HasRepoPermissionLevel, HasUserGroupPermissionLevel
 from kallithea.lib.exceptions import IMCCommitError, NonRelativePathError
 from kallithea.lib.hooks import process_pushed_raw_ids
 from kallithea.lib.utils import action_logger, get_filesystem_repos, make_ui
 from kallithea.lib.utils2 import safe_bytes, safe_str, safe_unicode, set_hook_environment
 from kallithea.lib.vcs import get_backend
 from kallithea.lib.vcs.backends.base import EmptyChangeset
 from kallithea.lib.vcs.exceptions import RepositoryError
 from kallithea.lib.vcs.nodes import FileNode
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 from kallithea.model.db import PullRequest, RepoGroup, Repository, Session, Ui, User, UserFollowing, UserLog
 log = logging.getLogger(__name__)
 class UserTemp(object):
     def __init__(self, user_id):
         self.user_id = user_id
     def __repr__(self):
         return "<%s('id:%s')>" % (self.__class__.__name__, self.user_id)
 class RepoTemp(object):
     def __init__(self, repo_id):
         self.repo_id = repo_id
     def __repr__(self):
         return "<%s('id:%s')>" % (self.__class__.__name__, self.repo_id)
 class _PermCheckIterator(object):
     def __init__(self, obj_list, obj_attr, perm_set, perm_checker, extra_kwargs=None):
         """
         Creates iterator from given list of objects, additionally
         checking permission for them from perm_set var
         :param obj_list: list of db objects
         :param obj_attr: attribute of object to pass into perm_checker
         :param perm_set: list of permissions to check
         :param perm_checker: callable to check permissions against
         """
         self.obj_list = obj_list
         self.obj_attr = obj_attr
         self.perm_set = perm_set
         self.perm_checker = perm_checker
         self.extra_kwargs = extra_kwargs or {}
     def __len__(self):
         return len(self.obj_list)
     def __repr__(self):
         return '<%s (%s)>' % (self.__class__.__name__, self.__len__())
     def __iter__(self):
         for db_obj in self.obj_list:
             # check permission at this level
             name = getattr(db_obj, self.obj_attr, None)
             if not self.perm_checker(*self.perm_set)(
                     name, self.__class__.__name__, **self.extra_kwargs):
                 continue
             yield db_obj
 class RepoList(_PermCheckIterator):
     def __init__(self, db_repo_list, perm_level, extra_kwargs=None):
         super(RepoList, self).__init__(obj_list=db_repo_list,
                     obj_attr='repo_name', perm_set=[perm_level],
                     perm_checker=HasRepoPermissionLevel,
                     extra_kwargs=extra_kwargs)
 class RepoGroupList(_PermCheckIterator):
     def __init__(self, db_repo_group_list, perm_level, extra_kwargs=None):
         super(RepoGroupList, self).__init__(obj_list=db_repo_group_list,
                     obj_attr='group_name', perm_set=[perm_level],
                     perm_checker=HasRepoGroupPermissionLevel,
                     extra_kwargs=extra_kwargs)
 class UserGroupList(_PermCheckIterator):
     def __init__(self, db_user_group_list, perm_level, extra_kwargs=None):
         super(UserGroupList, self).__init__(obj_list=db_user_group_list,
                     obj_attr='users_group_name', perm_set=[perm_level],
                     perm_checker=HasUserGroupPermissionLevel,
                     extra_kwargs=extra_kwargs)
 class ScmModel(object):
     """
     Generic Scm Model
     """
     def __get_repo(self, instance):
         cls = Repository
         if isinstance(instance, cls):
             return instance
         elif isinstance(instance, int):
             return cls.get(instance)
-        elif isinstance(instance, basestring):
+        elif isinstance(instance, str):
             if instance.isdigit():
                 return cls.get(int(instance))
             return cls.get_by_repo_name(instance)
         elif instance is not None:
             raise Exception('given object must be int, basestr or Instance'
                             ' of %s got %s' % (type(cls), type(instance)))
     @LazyProperty
     def repos_path(self):
         """
         Gets the repositories root path from database
         """
         q = Ui.query().filter(Ui.ui_key == '/').one()
         return q.ui_value
     def repo_scan(self, repos_path=None):
         """
         Listing of repositories in given path. This path should not be a
         repository itself. Return a dictionary of repository objects mapping to
         vcs instances.
         :param repos_path: path to directory containing repositories
         """
         if repos_path is None:
             repos_path = self.repos_path
         log.info('scanning for repositories in %s', repos_path)
         baseui = make_ui()
         repos = {}
         for name, path in get_filesystem_repos(repos_path):
             # name need to be decomposed and put back together using the /
             # since this is internal storage separator for kallithea
             name = Repository.normalize_repo_name(name)
             try:
                 if name in repos:
                     raise RepositoryError('Duplicate repository name %s '
                                           'found in %s' % (name, path))
                 else:
                     klass = get_backend(path[0])
                     if path[0] == 'hg' and path[0] in BACKENDS:
                         repos[name] = klass(safe_str(path[1]), baseui=baseui)
                     if path[0] == 'git' and path[0] in BACKENDS:
                         repos[name] = klass(path[1])
             except OSError:
                 continue
         log.debug('found %s paths with repositories', len(repos))
         return repos
     def get_repos(self, repos):
         """Return the repos the user has access to"""
         return RepoList(repos, perm_level='read')
     def get_repo_groups(self, groups=None):
         """Return the repo groups the user has access to
         If no groups are specified, use top level groups.
         """
         if groups is None:
             groups = RepoGroup.query() \
                 .filter(RepoGroup.parent_group_id == None).all()
         return RepoGroupList(groups, perm_level='read')
     def mark_for_invalidation(self, repo_name):
         """
         Mark caches of this repo invalid in the database.
         :param repo_name: the repo for which caches should be marked invalid
         """
         log.debug("Marking %s as invalidated and update cache", repo_name)
         repo = Repository.get_by_repo_name(repo_name)
         if repo is not None:
             repo.set_invalidate()
             repo.update_changeset_cache()
     def toggle_following_repo(self, follow_repo_id, user_id):
         f = UserFollowing.query() \
             .filter(UserFollowing.follows_repository_id == follow_repo_id) \
             .filter(UserFollowing.user_id == user_id).scalar()
         if f is not None:
             try:
                 Session().delete(f)
                 action_logger(UserTemp(user_id),
                               'stopped_following_repo',
                               RepoTemp(follow_repo_id))
                 return
             except Exception:
                 log.error(traceback.format_exc())
                 raise
         try:
             f = UserFollowing()
             f.user_id = user_id
             f.follows_repository_id = follow_repo_id
             Session().add(f)
             action_logger(UserTemp(user_id),
                           'started_following_repo',
                           RepoTemp(follow_repo_id))
         except Exception:
             log.error(traceback.format_exc())
             raise
     def toggle_following_user(self, follow_user_id, user_id):
         f = UserFollowing.query() \
             .filter(UserFollowing.follows_user_id == follow_user_id) \
             .filter(UserFollowing.user_id == user_id).scalar()
         if f is not None:
             try:
                 Session().delete(f)
                 return
             except Exception:
                 log.error(traceback.format_exc())
                 raise
         try:
             f = UserFollowing()
             f.user_id = user_id
             f.follows_user_id = follow_user_id
             Session().add(f)
         except Exception:
             log.error(traceback.format_exc())
             raise
     def is_following_repo(self, repo_name, user_id):
         r = Repository.query() \
             .filter(Repository.repo_name == repo_name).scalar()
         f = UserFollowing.query() \
             .filter(UserFollowing.follows_repository == r) \
             .filter(UserFollowing.user_id == user_id).scalar()
         return f is not None
     def is_following_user(self, username, user_id):
         u = User.get_by_username(username)
         f = UserFollowing.query() \
             .filter(UserFollowing.follows_user == u) \
             .filter(UserFollowing.user_id == user_id).scalar()
         return f is not None
     def get_followers(self, repo):
         repo = Repository.guess_instance(repo)
         return UserFollowing.query() \
                 .filter(UserFollowing.follows_repository == repo).count()
     def get_forks(self, repo):
         repo = Repository.guess_instance(repo)
         return Repository.query() \
                 .filter(Repository.fork == repo).count()
     def get_pull_requests(self, repo):
         repo = Repository.guess_instance(repo)
         return PullRequest.query() \
                 .filter(PullRequest.other_repo == repo) \
                 .filter(PullRequest.status != PullRequest.STATUS_CLOSED).count()
     def mark_as_fork(self, repo, fork, user):
         repo = self.__get_repo(repo)
         fork = self.__get_repo(fork)
         if fork and repo.repo_id == fork.repo_id:
             raise Exception("Cannot set repository as fork of itself")
         if fork and repo.repo_type != fork.repo_type:
             raise RepositoryError("Cannot set repository as fork of repository with other type")
         repo.fork = fork
         return repo
     def _handle_push(self, repo, username, ip_addr, action, repo_name, revisions):
         """
         Handle that the repository has changed.
         Adds an action log entry with the new revisions, and the head revision
         cache and in-memory caches are invalidated/updated.
         :param username: username who pushes
         :param action: push/push_local/push_remote
         :param repo_name: name of repo
         :param revisions: list of revisions that we pushed
         """
         set_hook_environment(username, ip_addr, repo_name, repo_alias=repo.alias, action=action)
         process_pushed_raw_ids(revisions) # also calls mark_for_invalidation
     def _get_IMC_module(self, scm_type):
         """
         Returns InMemoryCommit class based on scm_type
         :param scm_type:
         """
         if scm_type == 'hg':
             from kallithea.lib.vcs.backends.hg import MercurialInMemoryChangeset
             return MercurialInMemoryChangeset
         if scm_type == 'git':
             from kallithea.lib.vcs.backends.git import GitInMemoryChangeset
             return GitInMemoryChangeset
         raise Exception('Invalid scm_type, must be one of hg,git got %s'
                         % (scm_type,))
     def pull_changes(self, repo, username, ip_addr, clone_uri=None):
         """
         Pull from "clone URL" or fork origin.
         """
         dbrepo = self.__get_repo(repo)
         if clone_uri is None:
             clone_uri = dbrepo.clone_uri or dbrepo.fork and dbrepo.fork.repo_full_path
         if not clone_uri:
             raise Exception("This repository doesn't have a clone uri")
         repo = dbrepo.scm_instance
         repo_name = dbrepo.repo_name
         try:
             if repo.alias == 'git':
                 repo.fetch(clone_uri)
                 # git doesn't really have something like post-fetch action
                 # we fake that now.
                 # TODO: extract fetched revisions ... somehow ...
                 self._handle_push(repo,
                                   username=username,
                                   ip_addr=ip_addr,
                                   action='push_remote',
                                   repo_name=repo_name,
                                   revisions=[])
             else:
                 set_hook_environment(username, ip_addr, dbrepo.repo_name,
                                            repo.alias, action='push_remote')
                 repo.pull(clone_uri)
         except Exception:
             log.error(traceback.format_exc())
             raise
     def commit_change(self, repo, repo_name, cs, user, ip_addr, author, message,
                       content, f_path):
         """
         Commit a change to a single file
         :param repo: a db_repo.scm_instance
         """
         user = User.guess_instance(user)
         IMC = self._get_IMC_module(repo.alias)
         # decoding here will force that we have proper encoded values
         # in any other case this will throw exceptions and deny commit
         content = safe_str(content)
         path = safe_str(f_path)
         # message and author needs to be unicode
         # proper backend should then translate that into required type
         message = safe_unicode(message)
         author = safe_unicode(author)
         imc = IMC(repo)
         imc.change(FileNode(path, content, mode=cs.get_file_mode(f_path)))
         try:
             tip = imc.commit(message=message, author=author,
                              parents=[cs], branch=cs.branch)
         except Exception as e:
             log.error(traceback.format_exc())
             # clear caches - we also want a fresh object if commit fails
             self.mark_for_invalidation(repo_name)
             raise IMCCommitError(str(e))
         self._handle_push(repo,
                           username=user.username,
                           ip_addr=ip_addr,
                           action='push_local',
                           repo_name=repo_name,
                           revisions=[tip.raw_id])
         return tip
     def _sanitize_path(self, f_path):
         if f_path.startswith('/') or f_path.startswith('.') or '../' in f_path:
             raise NonRelativePathError('%s is not an relative path' % f_path)
         if f_path:
             f_path = posixpath.normpath(f_path)
         return f_path
     def get_nodes(self, repo_name, revision, root_path='/', flat=True):
         """
         Recursively walk root dir and return a set of all paths found.
         :param repo_name: name of repository
         :param revision: revision for which to list nodes
         :param root_path: root path to list
         :param flat: return as a list, if False returns a dict with description
         """
         _files = list()
         _dirs = list()
         try:
             _repo = self.__get_repo(repo_name)
             changeset = _repo.scm_instance.get_changeset(revision)
             root_path = root_path.lstrip('/')
             for topnode, dirs, files in changeset.walk(root_path):
                 for f in files:
                     _files.append(f.path if flat else {"name": f.path,
                                                        "type": "file"})
                 for d in dirs:
                     _dirs.append(d.path if flat else {"name": d.path,
                                                       "type": "dir"})
         except RepositoryError:
             log.debug(traceback.format_exc())
             raise
         return _dirs, _files
     def create_nodes(self, user, ip_addr, repo, message, nodes, parent_cs=None,
                      author=None, trigger_push_hook=True):
         """
         Commits specified nodes to repo.
         :param user: Kallithea User object or user_id, the committer
         :param repo: Kallithea Repository object
         :param message: commit message
         :param nodes: mapping {filename:{'content':content},...}
         :param parent_cs: parent changeset, can be empty than it's initial commit
         :param author: author of commit, cna be different that committer only for git
         :param trigger_push_hook: trigger push hooks
         :returns: new committed changeset
         """
         user = User.guess_instance(user)
         scm_instance = repo.scm_instance_no_cache()
         processed_nodes = []
         for f_path in nodes:
             content = nodes[f_path]['content']
             f_path = self._sanitize_path(f_path)
             f_path = safe_str(f_path)
             # decoding here will force that we have proper encoded values
             # in any other case this will throw exceptions and deny commit
-            if isinstance(content, (basestring,)):
+            if isinstance(content, (str,)):
                 content = safe_str(content)
             else:
                 content = content.read()
             processed_nodes.append((f_path, content))
         message = safe_unicode(message)
         committer = user.full_contact
         author = safe_unicode(author) if author else committer
         IMC = self._get_IMC_module(scm_instance.alias)
         imc = IMC(scm_instance)
         if not parent_cs:
             parent_cs = EmptyChangeset(alias=scm_instance.alias)
         if isinstance(parent_cs, EmptyChangeset):
             # EmptyChangeset means we we're editing empty repository
             parents = None
         else:
             parents = [parent_cs]
         # add multiple nodes
         for path, content in processed_nodes:
             imc.add(FileNode(path, content=content))
         tip = imc.commit(message=message,
                          author=author,
                          parents=parents,
                          branch=parent_cs.branch)
         if trigger_push_hook:
             self._handle_push(scm_instance,
                               username=user.username,
                               ip_addr=ip_addr,
                               action='push_local',
                               repo_name=repo.repo_name,
                               revisions=[tip.raw_id])
         else:
             self.mark_for_invalidation(repo.repo_name)
         return tip
     def update_nodes(self, user, ip_addr, repo, message, nodes, parent_cs=None,
                      author=None, trigger_push_hook=True):
         """
         Commits specified nodes to repo. Again.
         """
         user = User.guess_instance(user)
         scm_instance = repo.scm_instance_no_cache()
         message = safe_unicode(message)
         committer = user.full_contact
         author = safe_unicode(author) if author else committer
         imc_class = self._get_IMC_module(scm_instance.alias)
         imc = imc_class(scm_instance)
         if not parent_cs:
             parent_cs = EmptyChangeset(alias=scm_instance.alias)
         if isinstance(parent_cs, EmptyChangeset):
             # EmptyChangeset means we we're editing empty repository
             parents = None
         else:
             parents = [parent_cs]
         # add multiple nodes
         for _filename, data in nodes.items():
             # new filename, can be renamed from the old one
             filename = self._sanitize_path(data['filename'])
             old_filename = self._sanitize_path(_filename)
             content = data['content']
             filenode = FileNode(old_filename, content=content)
             op = data['op']
             if op == 'add':
                 imc.add(filenode)
             elif op == 'del':
                 imc.remove(filenode)
             elif op == 'mod':
                 if filename != old_filename:
                     # TODO: handle renames, needs vcs lib changes
                     imc.remove(filenode)
                     imc.add(FileNode(filename, content=content))
                 else:
                     imc.change(filenode)
         # commit changes
         tip = imc.commit(message=message,
                          author=author,
                          parents=parents,
                          branch=parent_cs.branch)
         if trigger_push_hook:
             self._handle_push(scm_instance,
                               username=user.username,
                               ip_addr=ip_addr,
                               action='push_local',
                               repo_name=repo.repo_name,
                               revisions=[tip.raw_id])
         else:
             self.mark_for_invalidation(repo.repo_name)
     def delete_nodes(self, user, ip_addr, repo, message, nodes, parent_cs=None,
                      author=None, trigger_push_hook=True):
         """
         Deletes specified nodes from repo.
         :param user: Kallithea User object or user_id, the committer
         :param repo: Kallithea Repository object
         :param message: commit message
         :param nodes: mapping {filename:{'content':content},...}
         :param parent_cs: parent changeset, can be empty than it's initial commit
         :param author: author of commit, cna be different that committer only for git
         :param trigger_push_hook: trigger push hooks
         :returns: new committed changeset after deletion
         """
         user = User.guess_instance(user)
         scm_instance = repo.scm_instance_no_cache()
         processed_nodes = []
         for f_path in nodes:
             f_path = self._sanitize_path(f_path)
             # content can be empty but for compatibility it allows same dicts
             # structure as add_nodes
             content = nodes[f_path].get('content')
             processed_nodes.append((f_path, content))
         message = safe_unicode(message)
         committer = user.full_contact
         author = safe_unicode(author) if author else committer
         IMC = self._get_IMC_module(scm_instance.alias)
         imc = IMC(scm_instance)
         if not parent_cs:
             parent_cs = EmptyChangeset(alias=scm_instance.alias)
         if isinstance(parent_cs, EmptyChangeset):
             # EmptyChangeset means we we're editing empty repository
             parents = None
         else:
             parents = [parent_cs]
         # add multiple nodes
         for path, content in processed_nodes:
             imc.remove(FileNode(path, content=content))
         tip = imc.commit(message=message,
                          author=author,
                          parents=parents,
                          branch=parent_cs.branch)
         if trigger_push_hook:
             self._handle_push(scm_instance,
                               username=user.username,
                               ip_addr=ip_addr,
                               action='push_local',
                               repo_name=repo.repo_name,
                               revisions=[tip.raw_id])
         else:
             self.mark_for_invalidation(repo.repo_name)
         return tip
     def get_unread_journal(self):
         return UserLog.query().count()
     def get_repo_landing_revs(self, repo=None):
         """
         Generates select option with tags branches and bookmarks (for hg only)
         grouped by type
         :param repo:
         """
         hist_l = []
         choices = []
         repo = self.__get_repo(repo)
         hist_l.append(('rev:tip', _('latest tip')))
         choices.append('rev:tip')
         if repo is None:
             return choices, hist_l
         repo = repo.scm_instance
         branches_group = ([(u'branch:%s' % k, k) for k, v in
                            repo.branches.items()], _("Branches"))
         hist_l.append(branches_group)
         choices.extend([x[0] for x in branches_group[0]])
         if repo.alias == 'hg':
             bookmarks_group = ([(u'book:%s' % k, k) for k, v in
                                 repo.bookmarks.items()], _("Bookmarks"))

kallithea/model/user_group.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.user_group
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 user group model 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: Oct 1, 2011
 :author: nvinot, marcink
 """
 import logging
 import traceback
 from kallithea.lib.exceptions import RepoGroupAssignmentError, UserGroupsAssignedException
 from kallithea.model.db import (
     Permission, Session, User, UserGroup, UserGroupMember, UserGroupRepoToPerm, UserGroupToPerm, UserGroupUserGroupToPerm, UserUserGroupToPerm)
 log = logging.getLogger(__name__)
 class UserGroupModel(object):
     def _create_default_perms(self, user_group):
         # create default permission
         default_perm = 'usergroup.read'
         def_user = User.get_default_user()
         for p in def_user.user_perms:
             if p.permission.permission_name.startswith('usergroup.'):
                 default_perm = p.permission.permission_name
                 break
         user_group_to_perm = UserUserGroupToPerm()
         user_group_to_perm.permission = Permission.get_by_key(default_perm)
         user_group_to_perm.user_group = user_group
         user_group_to_perm.user_id = def_user.user_id
         Session().add(user_group_to_perm)
         return user_group_to_perm
     def _update_permissions(self, user_group, perms_new=None,
                             perms_updates=None):
         from kallithea.lib.auth import HasUserGroupPermissionLevel
         if not perms_new:
             perms_new = []
         if not perms_updates:
             perms_updates = []
         # update permissions
         for member, perm, member_type in perms_updates:
             if member_type == 'user':
                 # this updates existing one
                 self.grant_user_permission(
                     user_group=user_group, user=member, perm=perm
+                )
             else:
                 # check if we have permissions to alter this usergroup's access
                 if HasUserGroupPermissionLevel('read')(member):
                     self.grant_user_group_permission(
                         target_user_group=user_group, user_group=member, perm=perm
+                    )
         # set new permissions
         for member, perm, member_type in perms_new:
             if member_type == 'user':
                 self.grant_user_permission(
                     user_group=user_group, user=member, perm=perm
+                )
             else:
                 # check if we have permissions to alter this usergroup's access
                 if HasUserGroupPermissionLevel('read')(member):
                     self.grant_user_group_permission(
                         target_user_group=user_group, user_group=member, perm=perm
+                    )
     def get(self, user_group_id):
         return UserGroup.get(user_group_id)
     def get_group(self, user_group):
         return UserGroup.guess_instance(user_group)
     def get_by_name(self, name, cache=False, case_insensitive=False):
         return UserGroup.get_by_group_name(name, cache=cache, case_insensitive=case_insensitive)
     def create(self, name, description, owner, active=True, group_data=None):
         try:
             new_user_group = UserGroup()
             new_user_group.owner = User.guess_instance(owner)
             new_user_group.users_group_name = name
             new_user_group.user_group_description = description
             new_user_group.users_group_active = active
             if group_data:
                 new_user_group.group_data = group_data
             Session().add(new_user_group)
             self._create_default_perms(new_user_group)
             self.grant_user_permission(user_group=new_user_group,
                                        user=owner, perm='usergroup.admin')
             return new_user_group
         except Exception:
             log.error(traceback.format_exc())
             raise
     def update(self, user_group, form_data):
         try:
             user_group = UserGroup.guess_instance(user_group)
             for k, v in form_data.items():
                 if k == 'users_group_members':
                     members_list = []
                     if v:
-                        v = [v] if isinstance(v, basestring) else v
+                        v = [v] if isinstance(v, str) else v
                         for u_id in set(v):
                             member = UserGroupMember(user_group.users_group_id, u_id)
                             members_list.append(member)
                             Session().add(member)
                     user_group.members = members_list
                 setattr(user_group, k, v)
             # Flush to make db assign users_group_member_id to newly
             # created UserGroupMembers.
             Session().flush()
         except Exception:
             log.error(traceback.format_exc())
             raise
     def delete(self, user_group, force=False):
         """
         Deletes user group, unless force flag is used
         raises exception if there are members in that group, else deletes
         group and users
         :param user_group:
         :param force:
         """
         user_group = UserGroup.guess_instance(user_group)
         try:
             # check if this group is not assigned to repo
             assigned_groups = UserGroupRepoToPerm.query() \
                 .filter(UserGroupRepoToPerm.users_group == user_group).all()
             assigned_groups = [x.repository.repo_name for x in assigned_groups]
             if assigned_groups and not force:
                 raise UserGroupsAssignedException(
                     'User Group assigned to %s' % ", ".join(assigned_groups))
             Session().delete(user_group)
         except Exception:
             log.error(traceback.format_exc())
             raise
     def add_user_to_group(self, user_group, user):
         """Return True if user already is in the group - else return the new UserGroupMember"""
         user_group = UserGroup.guess_instance(user_group)
         user = User.guess_instance(user)
         for m in user_group.members:
             u = m.user
             if u.user_id == user.user_id:
                 # user already in the group, skip
                 return True
         try:
             user_group_member = UserGroupMember()
             user_group_member.user = user
             user_group_member.users_group = user_group
             user_group.members.append(user_group_member)
             user.group_member.append(user_group_member)
             Session().add(user_group_member)
             return user_group_member
         except Exception:
             log.error(traceback.format_exc())
             raise
     def remove_user_from_group(self, user_group, user):
         user_group = UserGroup.guess_instance(user_group)
         user = User.guess_instance(user)
         user_group_member = None
         for m in user_group.members:
             if m.user_id == user.user_id:
                 # Found this user's membership row
                 user_group_member = m
                 break
         if user_group_member:
             try:
                 Session().delete(user_group_member)
                 return True
             except Exception:
                 log.error(traceback.format_exc())
                 raise
         else:
             # User isn't in that group
             return False
     def has_perm(self, user_group, perm):
         user_group = UserGroup.guess_instance(user_group)
         perm = Permission.guess_instance(perm)
         return UserGroupToPerm.query() \
             .filter(UserGroupToPerm.users_group == user_group) \
             .filter(UserGroupToPerm.permission == perm).scalar() is not None
     def grant_perm(self, user_group, perm):
         user_group = UserGroup.guess_instance(user_group)
         perm = Permission.guess_instance(perm)
         # if this permission is already granted skip it
         _perm = UserGroupToPerm.query() \
             .filter(UserGroupToPerm.users_group == user_group) \
             .filter(UserGroupToPerm.permission == perm) \
             .scalar()
         if _perm:
             return
         new = UserGroupToPerm()
         new.users_group = user_group
         new.permission = perm
         Session().add(new)
         return new
     def revoke_perm(self, user_group, perm):
         user_group = UserGroup.guess_instance(user_group)
         perm = Permission.guess_instance(perm)
         obj = UserGroupToPerm.query() \
             .filter(UserGroupToPerm.users_group == user_group) \
             .filter(UserGroupToPerm.permission == perm).scalar()
         if obj is not None:
             Session().delete(obj)
     def grant_user_permission(self, user_group, user, perm):
         """
         Grant permission for user on given user group, or update
         existing one if found
         :param user_group: Instance of UserGroup, users_group_id,
             or users_group_name
         :param user: Instance of User, user_id or username
         :param perm: Instance of Permission, or permission_name
         """
         user_group = UserGroup.guess_instance(user_group)
         user = User.guess_instance(user)
         permission = Permission.guess_instance(perm)
         # check if we have that permission already
         obj = UserUserGroupToPerm.query() \
             .filter(UserUserGroupToPerm.user == user) \
             .filter(UserUserGroupToPerm.user_group == user_group) \
             .scalar()
         if obj is None:
             # create new !
             obj = UserUserGroupToPerm()
             Session().add(obj)
         obj.user_group = user_group
         obj.user = user
         obj.permission = permission
         log.debug('Granted perm %s to %s on %s', perm, user, user_group)
         return obj
     def revoke_user_permission(self, user_group, user):
         """
         Revoke permission for user on given repository group
         :param user_group: Instance of RepoGroup, repositories_group_id,
             or repositories_group name
         :param user: Instance of User, user_id or username
         """
         user_group = UserGroup.guess_instance(user_group)
         user = User.guess_instance(user)
         obj = UserUserGroupToPerm.query() \
             .filter(UserUserGroupToPerm.user == user) \
             .filter(UserUserGroupToPerm.user_group == user_group) \
             .scalar()
         if obj is not None:
             Session().delete(obj)
             log.debug('Revoked perm on %s on %s', user_group, user)
     def grant_user_group_permission(self, target_user_group, user_group, perm):
         """
         Grant user group permission for given target_user_group
         :param target_user_group:
         :param user_group:
         :param perm:
         """
         target_user_group = UserGroup.guess_instance(target_user_group)
         user_group = UserGroup.guess_instance(user_group)
         permission = Permission.guess_instance(perm)
         # forbid assigning same user group to itself
         if target_user_group == user_group:
             raise RepoGroupAssignmentError('target repo:%s cannot be '
                                            'assigned to itself' % target_user_group)
         # check if we have that permission already
         obj = UserGroupUserGroupToPerm.query() \
             .filter(UserGroupUserGroupToPerm.target_user_group == target_user_group) \
             .filter(UserGroupUserGroupToPerm.user_group == user_group) \
             .scalar()

kallithea/tests/scripts/manual_test_crawler.py

➞

Show inline comments

 #!/usr/bin/env python3
 # -*- 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.tests.scripts.manual_test_crawler
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Test for crawling a project for memory usage
 This should be runned just as regular script together
 with a watch script that will show memory usage.
 watch -n1 ./kallithea/tests/mem_watch
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Apr 21, 2010
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 from __future__ import print_function
 import cookielib
 import os
 import sys
 import tempfile
 import time
 import urllib
 import urllib2
 from os.path import dirname
 from kallithea.lib import vcs
 from kallithea.lib.compat import OrderedSet
 from kallithea.lib.vcs.exceptions import RepositoryError
 __here__ = os.path.abspath(__file__)
 __root__ = dirname(dirname(dirname(__here__)))
 sys.path.append(__root__)
 PASES = 3
 HOST = 'http://127.0.0.1'
 PORT = 5000
 BASE_URI = '%s:%s/' % (HOST, PORT)
 if len(sys.argv) == 2:
     BASE_URI = sys.argv[1]
 if not BASE_URI.endswith('/'):
     BASE_URI += '/'
 print('Crawling @ %s' % BASE_URI)
 BASE_URI += '%s'
 PROJECT_PATH = os.path.join('/', 'home', 'username', 'repos')
 PROJECTS = [
     # 'linux-magx-pbranch',
     'CPython',
     'kallithea',
+]
 cj = cookielib.FileCookieJar(os.path.join(tempfile.gettempdir(), 'rc_test_cookie.txt'))
 o = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
 o.addheaders = [
     ('User-agent', 'kallithea-crawler'),
     ('Accept-Language', 'en - us, en;q = 0.5')
+]
 urllib2.install_opener(o)
 def _get_repo(proj):
-    if isinstance(proj, basestring):
+    if isinstance(proj, str):
         repo = vcs.get_repo(os.path.join(PROJECT_PATH, proj))
         proj = proj
     else:
         repo = proj
         proj = repo.name
     return repo, proj
 def test_changelog_walk(proj, pages=100):
     repo, proj = _get_repo(proj)
     total_time = 0
     for i in range(1, pages):
         page = '/'.join((proj, 'changelog',))
         full_uri = (BASE_URI % page) + '?' + urllib.urlencode({'page': i})
         s = time.time()
         f = o.open(full_uri)
         assert f.url == full_uri, 'URL:%s does not match %s' % (f.url, full_uri)
         size = len(f.read())
         e = time.time() - s
         total_time += e
         print('visited %s size:%s req:%s ms' % (full_uri, size, e))
     print('total_time', total_time)
     print('average on req', total_time / float(pages))
 def test_changeset_walk(proj, limit=None):
     repo, proj = _get_repo(proj)
     print('processing', os.path.join(PROJECT_PATH, proj))
     total_time = 0
     cnt = 0
     for i in repo:
         cnt += 1
         raw_cs = '/'.join((proj, 'changeset', i.raw_id))
         if limit and limit == cnt:
             break
         full_uri = (BASE_URI % raw_cs)
         print('%s visiting %s/%s' % (cnt, full_uri, i))
         s = time.time()
         f = o.open(full_uri)
         size = len(f.read())
         e = time.time() - s
         total_time += e
         print('%s visited %s/%s size:%s req:%s ms' % (cnt, full_uri, i, size, e))
     print('total_time', total_time)
     print('average on req', total_time / float(cnt))
 def test_files_walk(proj, limit=100):
     repo, proj = _get_repo(proj)
     print('processing', os.path.join(PROJECT_PATH, proj))
     total_time = 0
     paths_ = OrderedSet([''])
     try:
         tip = repo.get_changeset('tip')
         for topnode, dirs, files in tip.walk('/'):
             for dir in dirs:
                 paths_.add(dir.path)
                 for f in dir:
                     paths_.add(f.path)
             for f in files:
                 paths_.add(f.path)
     except RepositoryError as e:
         pass
     cnt = 0
     for f in paths_:
         cnt += 1
         if limit and limit == cnt:
             break
         file_path = '/'.join((proj, 'files', 'tip', f))
         full_uri = (BASE_URI % file_path)
         print('%s visiting %s' % (cnt, full_uri))
         s = time.time()
         f = o.open(full_uri)
         size = len(f.read())
         e = time.time() - s
         total_time += e
         print('%s visited OK size:%s req:%s ms' % (cnt, size, e))
     print('total_time', total_time)
     print('average on req', total_time / float(cnt))
 if __name__ == '__main__':
     for path in PROJECTS:
         repo = vcs.get_repo(os.path.join(PROJECT_PATH, path))
         for i in range(PASES):
             print('PASS %s/%s' % (i, PASES))
             test_changelog_walk(repo, pages=80)
             test_changeset_walk(repo, limit=100)
             test_files_walk(repo, limit=100)

0 comments (0 inline, 0 general)