fix optimisation with super session that may lead to integrity loss
at some point I've decided to stop ensuring ?1 cardinality was respected
when adding a new relation using a super session, to avoid the cost of
the delete query. That was yet discussable because it introduced unexpected
difference between execute and unsafe_execute, which is imo not worth it.
Also, now that rql() in migration script default to unsafe_execute, we
definitly don't want that implicit behaviour change (which already cause
bug when for instance adding another default workflow for an entity type:
without that fix we end up with *two* default workflows while the schema
tells we can have only one.
IMO we should go to the direction that super session skip all security
check, but nothing else, unless explicitly asked.
"""provide class to do Apache rewrite rules'job inside cubicweb (though functionnalities
are much more limited for the moment)
:organization: Logilab
:copyright: 2007-2010 LOGILAB S.A. (Paris, FRANCE), license is LGPL v2.
:contact: http://www.logilab.fr/ -- mailto:contact@logilab.fr
:license: GNU Lesser General Public License, v2.1 - http://www.gnu.org/licenses
"""
__docformat__ = "restructuredtext en"
from re import compile
from cubicweb.web import Redirect
from cubicweb.web.component import Component
class RewriteCond(object):
def __init__(self, condition, match='host', rules=(), action='rewrite'):
self.condition = compile(condition)
assert match in ('host', 'path'), match
self.match_part = match
self.rules = []
for rule, replace in rules:
rulergx = compile(rule)
self.rules.append( (rulergx, replace) )
assert action in ('rewrite', 'redirect', 'stop'), action
self.process = getattr(self, 'action_%s' % action)
def match(self, **kwargs):
self._match = self.condition.match(kwargs[self.match_part])
return not self._match is None
def action_rewrite(self, path):
for rgx, replace in self.rules:
if not rgx.match(path) is None:
matchdict = self._match.groupdict() or None
if not matchdict is None:
replace = replace % matchdict
return rgx.sub(replace, path)
return path
def action_redirect(self, path):
url = self.action_rewrite(path)
raise Redirect(url)
def action_stop(self, path):
return path
class ApacheURLRewrite(Component):
"""inherit from this class with actual rules to activate apache style rewriting
rules should have the form :
[('condition pattern 1', [('rule1 pattern', 'replace expression'),
('rule2 pattern', 'replace expression')],
('condition pattern 2', [('rule1 pattern', 'replace expression'),
('rule2 pattern', 'replace expression')]
]
for instance the equivalent of the following apache rules:
RewriteCond %{HTTP_HOST} ^logilab\.fr
RewriteRule ^/(.*) http://www.logilab.fr/$1 [L,R=301]
RewriteCond %{HTTP_HOST} ^www\.logilab\.fr
RewriteRule ^/(.*) http://localhost:8080/$1 [L,P]
RewriteCond %{HTTP_HOST} ^(.+)\.logilab\.fr
RewriteRule ^/(data/.*) http://localhost:8080/$1 [L,P]
RewriteRule ^/(json.*) http://localhost:8080/$1 [L,P]
RewriteRule ^/(.*) http://localhost:8080/m_%1/$1 [L,P]
could be written (considering that no "host rewritting" is necessary):
class MyAppRules(ApacheURLRewrite):
rules = [
RewriteCond('logilab\.fr', match='host',
rules=[('/(.*)', r'http://www.logilab.fr/\1')],
action='redirect'),
RewriteCond('(www)\.logilab\.fr', match='host', action='stop'),
RewriteCond('/(data|json)/', match='path', action='stop'),
RewriteCond('(?P<cat>.*)\.logilab\.fr', match='host',
rules=[('/(.*)', r'/m_%(cat)s/\1')]),
]
"""
__abstract__ = True
__regid__ = 'urlrewriter'
rules = []
def get_rules(self, req):
return self.rules
def rewrite(self, host, path, req):
for cond in self.get_rules(req):
if cond.match(host=host, path=path):
return cond.process(path)
return path