[repo] optimize massive insertion/deletion by using the new set_operation function
Idea is that on massive insertion, cost of handling the list of operation
become non negligeable, so we should minimize the number of operations in
that list.
The set_operation function ease usage of operation associated to data in
session.transaction_data, and we only add the operation when data set isn't
initialized yet, else we simply add data to the set. The operation then
simply process accumulated data.
"""the Bookmark entity type for internal links
:organization: Logilab
:copyright: 2001-2010 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"
_ = unicode
from yams.buildobjs import EntityType, RelationType, SubjectRelation, String
from cubicweb.schema import RRQLExpression
class Bookmark(EntityType):
"""bookmarks are used to have user's specific internal links"""
__permissions__ = {
'read': ('managers', 'users', 'guests',),
'add': ('managers', 'users',),
'delete': ('managers', 'owners',),
'update': ('managers', 'owners',),
}
title = String(required=True, maxsize=128, internationalizable=True)
path = String(maxsize=512, required=True,
description=_("relative url of the bookmarked page"))
bookmarked_by = SubjectRelation('CWUser',
description=_("users using this bookmark"))
class bookmarked_by(RelationType):
__permissions__ = {'read': ('managers', 'users', 'guests',),
# test user in users group to avoid granting permission to anonymous user
'add': ('managers', RRQLExpression('O identity U, U in_group G, G name "users"')),
'delete': ('managers', RRQLExpression('O identity U, U in_group G, G name "users"')),
}