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.
"""inline help system, using ReST file in products `wdoc` directory
:organization: Logilab
:copyright: 2008-2010 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 itertools import chain
from os.path import join
from bisect import bisect_right
from datetime import date
from logilab.common.changelog import ChangeLog
from logilab.common.date import strptime, todate
from logilab.mtconverter import CHARSET_DECL_RGX
from cubicweb.selectors import match_form_params, yes
from cubicweb.view import StartupView
from cubicweb.uilib import rest_publish
from cubicweb.web import NotFound, action
_ = unicode
# table of content management #################################################
try:
from xml.etree.ElementTree import parse
except ImportError:
from elementtree.ElementTree import parse
def build_toc_index(node, index):
try:
nodeidx = node.attrib['resource']
assert not nodeidx in index, nodeidx
index[nodeidx] = node
except KeyError:
pass
for child in node:
build_toc_index(child, index)
child.parent = node
def get_insertion_point(section, index):
if section.attrib.get('insertafter'):
snode = index[section.attrib['insertafter']]
node = snode.parent
idx = node.getchildren().index(snode) + 1
elif section.attrib.get('insertbefore'):
snode = index[section.attrib['insertbefore']]
node = snode.parent
idx = node.getchildren().index(snode)
else:
node = index[section.attrib['appendto']]
idx = None
return node, idx
def build_toc(config):
alltocfiles = reversed(tuple(config.locate_all_files('toc.xml')))
maintoc = parse(alltocfiles.next()).getroot()
maintoc.parent = None
index = {}
build_toc_index(maintoc, index)
# insert component documentation into the tree according to their toc.xml
# file
for fpath in alltocfiles:
toc = parse(fpath).getroot()
for section in toc:
node, idx = get_insertion_point(section, index)
if idx is None:
node.append(section)
else:
node.insert(idx, section)
section.parent = node
build_toc_index(section, index)
return index
def title_for_lang(node, lang):
for title in node.findall('title'):
if title.attrib['{http://www.w3.org/XML/1998/namespace}lang'] == lang:
return unicode(title.text)
def subsections(node):
return [child for child in node if child.tag == 'section']
# help views ##################################################################
class InlineHelpView(StartupView):
__select__ = match_form_params('fid')
__regid__ = 'wdoc'
title = _('site documentation')
def call(self):
fid = self._cw.form['fid']
vreg = self._cw.vreg
for lang in chain((self._cw.lang, vreg.property_value('ui.language')),
vreg.config.available_languages()):
rid = '%s_%s.rst' % (fid, lang)
resourcedir = vreg.config.locate_doc_file(rid)
if resourcedir:
break
else:
raise NotFound
self.tocindex = build_toc(vreg.config)
try:
node = self.tocindex[fid]
except KeyError:
node = None
else:
self.navigation_links(node)
self.w(u'<div class="hr"></div>')
self.w(u'<h1>%s</h1>' % (title_for_lang(node, self._cw.lang)))
data = open(join(resourcedir, rid)).read()
self.w(rest_publish(self, data))
if node is not None:
self.subsections_links(node)
self.w(u'<div class="hr"></div>')
self.navigation_links(node)
def navigation_links(self, node):
req = self._cw
parent = node.parent
if parent is None:
return
brothers = subsections(parent)
self.w(u'<div class="docnav">\n')
previousidx = brothers.index(node) - 1
if previousidx >= 0:
self.navsection(brothers[previousidx], 'prev')
self.navsection(parent, 'up')
nextidx = brothers.index(node) + 1
if nextidx < len(brothers):
self.navsection(brothers[nextidx], 'next')
self.w(u'</div>\n')
navinfo = {'prev': ('', 'data/previous.png', _('i18nprevnext_previous')),
'next': ('', 'data/next.png', _('i18nprevnext_next')),
'up': ('', 'data/up.png', _('i18nprevnext_up'))}
def navsection(self, node, navtype):
htmlclass, imgpath, msgid = self.navinfo[navtype]
self.w(u'<span class="%s">' % htmlclass)
self.w(u'%s : ' % self._cw._(msgid))
self.w(u'<a href="%s">%s</a>' % (
self._cw.build_url('doc/'+node.attrib['resource']),
title_for_lang(node, self._cw.lang)))
self.w(u'</span>\n')
def subsections_links(self, node, first=True):
sub = subsections(node)
if not sub:
return
if first:
self.w(u'<div class="hr"></div>')
self.w(u'<ul class="docsum">')
for child in sub:
self.w(u'<li><a href="%s">%s</a>' % (
self._cw.build_url('doc/'+child.attrib['resource']),
title_for_lang(child, self._cw.lang)))
self.subsections_links(child, False)
self.w(u'</li>')
self.w(u'</ul>\n')
class InlineHelpImageView(StartupView):
__regid__ = 'wdocimages'
__select__ = match_form_params('fid')
binary = True
templatable = False
content_type = 'image/png'
def call(self):
fid = self._cw.form['fid']
for lang in chain((self._cw.lang, self._cw.vreg.property_value('ui.language')),
self._cw.vreg.config.available_languages()):
rid = join('images', '%s_%s.png' % (fid, lang))
resourcedir = self._cw.vreg.config.locate_doc_file(rid)
if resourcedir:
break
else:
raise NotFound
self.w(open(join(resourcedir, rid)).read())
class ChangeLogView(StartupView):
__regid__ = 'changelog'
title = _('What\'s new?')
maxentries = 25
def call(self):
rid = 'ChangeLog_%s' % (self._cw.lang)
allentries = []
title = self._cw._(self.title)
restdata = ['.. -*- coding: utf-8 -*-', '', title, '='*len(title), '']
w = restdata.append
today = date.today()
for fpath in self._cw.vreg.config.locate_all_files(rid):
cl = ChangeLog(fpath)
encoding = 'utf-8'
# additional content may be found in title
for line in (cl.title + cl.additional_content).splitlines():
m = CHARSET_DECL_RGX.search(line)
if m is not None:
encoding = m.group(1)
continue
elif line.startswith('.. '):
w(unicode(line, encoding))
for entry in cl.entries:
if entry.date:
edate = todate(strptime(entry.date, '%Y-%m-%d'))
else:
edate = today
messages = []
for msglines, submsgs in entry.messages:
msgstr = unicode(' '.join(l.strip() for l in msglines), encoding)
msgstr += u'\n\n'
for submsglines in submsgs:
msgstr += ' - ' + unicode(' '.join(l.strip() for l in submsglines), encoding)
msgstr += u'\n'
messages.append(msgstr)
entry = (edate, messages)
allentries.insert(bisect_right(allentries, entry), entry)
latestdate = None
i = 0
for edate, messages in reversed(allentries):
if latestdate != edate:
fdate = self._cw.format_date(edate)
w(u'\n%s' % fdate)
w('~' * len(fdate))
latestdate = edate
for msg in messages:
w(u'* %s' % msg)
i += 1
if i > self.maxentries:
break
w('') # blank line
self.w(rest_publish(self, '\n'.join(restdata)))
class HelpAction(action.Action):
__regid__ = 'help'
__select__ = yes()
category = 'footer'
order = 0
title = _('Help')
def url(self):
return self._cw.build_url('doc/main')
class ChangeLogAction(action.Action):
__regid__ = 'changelog'
__select__ = yes()
category = 'footer'
order = 1
title = ChangeLogView.title
def url(self):
return self._cw.build_url('changelog')
class AboutAction(action.Action):
__regid__ = 'about'
__select__ = yes()
category = 'footer'
order = 2
title = _('about this site')
def url(self):
return self._cw.build_url('doc/about')