##############################################################################
#
# OpenERP, Open Source Business Applications
-# Copyright (C) 2012-2013 OpenERP S.A. (<http://openerp.com>).
+# Copyright (C) 2012-TODAY OpenERP S.A. (<http://openerp.com>).
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as
import socket
import threading
import time
+from email.utils import getaddresses
import openerp
from openerp.loglevels import ustr
tags_to_remove = ['html', 'body', 'font']
# allow new semantic HTML5 tags
-allowed_tags = clean.defs.tags | frozenset('article section header footer hgroup nav aside figure'.split())
-safe_attrs = clean.defs.safe_attrs | frozenset(['style'])
+allowed_tags = clean.defs.tags | frozenset('article section header footer hgroup nav aside figure main'.split() + [etree.Comment])
+safe_attrs = clean.defs.safe_attrs | frozenset(
+ ['style',
+ 'data-oe-model', 'data-oe-id', 'data-oe-field', 'data-oe-type', 'data-oe-expression', 'data-oe-translate', 'data-oe-nodeid',
+ 'data-publish', 'data-id', 'data-res_id', 'data-member_id', 'data-view-id'
+ ])
-def html_sanitize(src, silent=True):
+def html_sanitize(src, silent=True, strict=False):
if not src:
return src
src = ustr(src, errors='replace')
# html encode email tags
part = re.compile(r"(<(([^a<>]|a[^<>\s])[^<>]*)@[^<>]+>)", re.IGNORECASE | re.DOTALL)
src = part.sub(lambda m: cgi.escape(m.group(1)), src)
+ # html encode mako tags <% ... %> to decode them later and keep them alive, otherwise they are stripped by the cleaner
+ src = src.replace('<%', cgi.escape('<%'))
+ src = src.replace('%>', cgi.escape('%>'))
kwargs = {
'page_structure': True,
'forms': True, # remove form tags
'remove_unknown_tags': False,
'allow_tags': allowed_tags,
+ 'comments': False,
+ 'processing_instructions': False
}
if etree.LXML_VERSION >= (2, 3, 1):
# kill_tags attribute has been added in version 2.3.1
else:
kwargs['remove_tags'] = tags_to_kill + tags_to_remove
- if etree.LXML_VERSION >= (3, 1, 0):
- kwargs.update({
- 'safe_attrs_only': True,
- 'safe_attrs': safe_attrs,
- })
+ if strict:
+ if etree.LXML_VERSION >= (3, 1, 0):
+ # lxml < 3.1.0 does not allow to specify safe_attrs. We keep all attributes in order to keep "style"
+ kwargs.update({
+ 'safe_attrs_only': True,
+ 'safe_attrs': safe_attrs,
+ })
else:
- # lxml < 3.1.0 does not allow to specify safe_attrs. We keep all attributes in order to keep "style"
- kwargs['safe_attrs_only'] = False
+ kwargs['safe_attrs_only'] = False # keep oe-data attributes + style
+ kwargs['frames'] = False, # do not remove frames (embbed video in CMS blogs)
try:
# some corner cases make the parser crash (such as <SCRIPT/XSS SRC=\"http://ha.ckers.org/xss.js\"></SCRIPT> in test_mail)
cleaner = clean.Cleaner(**kwargs)
cleaned = cleaner.clean_html(src)
+ # MAKO compatibility: $, { and } inside quotes are escaped, preventing correct mako execution
+ cleaned = cleaned.replace('%24', '$')
+ cleaned = cleaned.replace('%7B', '{')
+ cleaned = cleaned.replace('%7D', '}')
+ cleaned = cleaned.replace('%20', ' ')
+ cleaned = cleaned.replace('%5B', '[')
+ cleaned = cleaned.replace('%5D', ']')
+ cleaned = cleaned.replace('<%', '<%')
+ cleaned = cleaned.replace('%>', '%>')
except etree.ParserError, e:
- if 'empty' in str(e):
- return ""
+ if 'empty' in str(e):
+ return ""
if not silent:
raise
logger.warning('ParserError obtained when sanitizing %r', src, exc_info=True)
raise
logger.warning('unknown error obtained when sanitizing %r', src, exc_info=True)
cleaned = '<p>Unknown error when sanitizing</p>'
+
+ # this is ugly, but lxml/etree tostring want to put everything in a 'div' that breaks the editor -> remove that
+ if cleaned.startswith('<div>') and cleaned.endswith('</div>'):
+ cleaned = cleaned[5:-6]
+
return cleaned
overlength_section_count = 0
cur_char_nbr = 0
for node in root.iter():
+ # comments do not need processing
+ # note: bug in node.get(value, default) for HtmlComments, default never returned
+ if node.tag == etree.Comment:
+ continue
# do not take into account multiple spaces that are displayed as max 1 space in html
node_text = ' '.join((node.text and node.text.strip(' \t\r\n') or '').split())
node.set('in_quote', '1')
node.set('tail_remove', '1')
if node.tag == 'blockquote' or node.get('text_quote') or node.get('text_signature'):
+ # here no quote_begin because we want to be able to remove some quoted
+ # text without removing all the remaining context
+ node.set('in_quote', '1')
+ if node.getparent() is not None and node.getparent().get('in_quote'):
+ # inside a block of removed text but not in quote_begin (see above)
node.set('in_quote', '1')
# shorten:
html = re.sub('<br\s*/?>', '\n', html)
html = re.sub('<.*?>', ' ', html)
html = html.replace(' ' * 2, ' ')
+ html = html.replace('>', '>')
+ html = html.replace('<', '<')
+ html = html.replace('&', '&')
# strip all lines
html = '\n'.join([x.strip() for x in html.splitlines()])
elif plaintext:
content = '\n%s\n' % plaintext2html(content, container_tag)
else:
- content = re.sub(r'(?i)(</?html.*>|</?body.*>|<!\W*DOCTYPE.*>)', '', content)
+ content = re.sub(r'(?i)(</?(?:html|body|head|!\s*DOCTYPE)[^>]*>)', '', content)
content = u'\n%s\n' % ustr(content)
# Force all tags to lowercase
html = re.sub(r'(</?)\W*(\w+)([ >])',
# Updated in 7.0 to match the model name as well
# Typical form of references is <timestamp-openerp-record_id-model_name@domain>
# group(1) = the record ID ; group(2) = the model (if any) ; group(3) = the domain
-reference_re = re.compile("<.*-open(?:object|erp)-(\\d+)(?:-([\w.]+))?.*@(.*)>", re.UNICODE)
+reference_re = re.compile("<.*-open(?:object|erp)-(\\d+)(?:-([\w.]+))?[^>]*@([^>]*)>", re.UNICODE)
-# Bounce regex
-# Typical form of bounce is bounce-128-crm.lead-34@domain
-# group(1) = the mail ID; group(2) = the model (if any); group(3) = the record ID
-bounce_re = re.compile("[\w]+-(\d+)-?([\w.]+)?-?(\d+)?", re.UNICODE)
def generate_tracking_message_id(res_id):
"""Returns a string that can be used in the Message-ID RFC822 header field
if not cr:
db_name = getattr(threading.currentThread(), 'dbname', None)
if db_name:
- local_cr = cr = openerp.registry(db_name).db.cursor()
+ local_cr = cr = openerp.registry(db_name).cursor()
else:
raise Exception("No database cursor found, please pass one explicitly")
""" Return a list of the email addresses found in ``text`` """
if not text:
return []
- return re.findall(r'([^ ,<@]+@[^> ,]+)', text)
+ return [addr[1] for addr in getaddresses([text])
+ # getaddresses() returns '' when email parsing fails, and
+ # sometimes returns emails without at least '@'. The '@'
+ # is strictly required in RFC2822's `addr-spec`.
+ if addr[1]
+ if '@' in addr[1]]