1 # -*- coding: utf-8 -*-
2 ##############################################################################
4 # OpenERP, Open Source Business Applications
5 # Copyright (C) 2012-2013 OpenERP S.A. (<http://openerp.com>).
7 # This program is free software: you can redistribute it and/or modify
8 # it under the terms of the GNU Affero General Public License as
9 # published by the Free Software Foundation, either version 3 of the
10 # License, or (at your option) any later version.
12 # This program is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 # GNU Affero General Public License for more details.
17 # You should have received a copy of the GNU Affero General Public License
18 # along with this program. If not, see <http://www.gnu.org/licenses/>.
20 ##############################################################################
22 from lxml import etree
26 import lxml.html.clean as clean
27 import openerp.pooler as pooler
33 from email.utils import getaddresses
35 from openerp.loglevels import ustr
37 _logger = logging.getLogger(__name__)
40 #----------------------------------------------------------
42 #----------------------------------------------------------
44 tags_to_kill = ["script", "head", "meta", "title", "link", "style", "frame", "iframe", "base", "object", "embed"]
45 tags_to_remove = ['html', 'body', 'font']
48 def html_sanitize(src):
51 src = ustr(src, errors='replace')
53 # html encode email tags
54 part = re.compile(r"(<(([^a<>]|a[^<>\s])[^<>]*)@[^<>]+>)", re.IGNORECASE | re.DOTALL)
55 src = part.sub(lambda m: cgi.escape(m.group(1)), src)
57 # some corner cases make the parser crash (such as <SCRIPT/XSS SRC=\"http://ha.ckers.org/xss.js\"></SCRIPT> in test_mail)
59 cleaner = clean.Cleaner(page_structure=True, style=False, safe_attrs_only=False, forms=False, kill_tags=tags_to_kill, remove_tags=tags_to_remove)
60 cleaned = cleaner.clean_html(src)
62 # lxml.clean version < 2.3.1 does not have a kill_tags attribute
64 cleaner = clean.Cleaner(page_structure=True, style=False, safe_attrs_only=False, forms=False, remove_tags=tags_to_kill + tags_to_remove)
65 cleaned = cleaner.clean_html(src)
67 if isinstance(e, etree.ParserError) and 'empty' in str(e):
69 _logger.warning('html_sanitize failed to parse %s' % (src))
70 cleaned = '<p>Impossible to parse</p>'
72 # MAKO compatibility: $, { and } inside quotes are escaped, preventing correct mako execution
73 cleaned = cleaned.replace('%24', '$')
74 cleaned = cleaned.replace('%7B', '{')
75 cleaned = cleaned.replace('%7D', '}')
76 cleaned = cleaned.replace('%20', ' ')
77 cleaned = cleaned.replace('%5B', '[')
78 cleaned = cleaned.replace('%5D', ']')
83 #----------------------------------------------------------
85 #----------------------------------------------------------
87 def html_email_clean(html):
88 """ html_email_clean: clean the html to display in the web client.
89 - strip email quotes (remove blockquote nodes)
90 - strip signatures (remove --\n{\n)Blahblah), by replacing <br> by
91 \n to avoid ignoring signatures converted into html
93 :param string html: sanitized html; tags like html or head should not
94 be present in the html string. This method therefore takes as input
95 html code coming from a sanitized source, like fields.html.
97 def _replace_matching_regex(regex, source, replace=''):
100 for item in re.finditer(regex, source):
101 dest += source[idx:item.start()] + replace
106 if not html or not isinstance(html, basestring):
111 # 0. remove encoding attribute inside tags
112 doctype = re.compile(r'(<[^>]*\s)(encoding=(["\'][^"\']*?["\']|[^\s\n\r>]+)(\s[^>]*|/)?>)', re.IGNORECASE | re.DOTALL)
113 html = doctype.sub(r"", html)
115 # 1. <br[ /]> -> \n, because otherwise the tree is obfuscated
116 br_tags = re.compile(r'([<]\s*[bB][rR]\s*\/?[>])')
117 html = _replace_matching_regex(br_tags, html, '__BR_TAG__')
119 # 2. form a tree, handle (currently ?) pure-text by enclosing them in a pre
120 root = lxml.html.fromstring(html)
121 if not len(root) and root.text is None and root.tail is None:
122 html = '<div>%s</div>' % html
123 root = lxml.html.fromstring(html)
125 # 2.5 remove quoted text in nodes
126 quote_tags = re.compile(r'(\n(>)+[^\n\r]*)')
127 for node in root.getiterator():
130 node.text = _replace_matching_regex(quote_tags, node.text)
132 # 3. remove blockquotes
133 quotes = [el for el in root.getiterator(tag='blockquote')]
135 # copy the node tail into parent text
137 parent = node.getparent()
138 parent.text = parent.text or '' + node.tail
140 node.getparent().remove(node)
142 # 4. strip signatures
143 signature = re.compile(r'([-]{2}[\s]?[\r\n]{1,2}[^\z]+)')
144 for elem in root.getiterator():
146 match = re.search(signature, elem.text)
148 elem.text = elem.text[:match.start()] + elem.text[match.end():]
150 match = re.search(signature, elem.tail)
152 elem.tail = elem.tail[:match.start()] + elem.tail[match.end():]
154 # 5. \n back to <br/>
155 html = etree.tostring(root, pretty_print=True)
156 html = html.replace('__BR_TAG__', '<br />')
159 # - ClEditor seems to love using <div><br /><div> -> replace with <br />
160 br_div_tags = re.compile(r'(<div>\s*<br\s*\/>\s*<\/div>)')
161 html = _replace_matching_regex(br_div_tags, html, '<br />')
166 #----------------------------------------------------------
167 # HTML/Text management
168 #----------------------------------------------------------
170 def html2plaintext(html, body_id=None, encoding='utf-8'):
171 """ From an HTML text, convert the HTML to plain text.
172 If @param body_id is provided then this is the tag where the
173 body (not necessarily <body>) starts.
175 ## (c) Fry-IT, www.fry-it.com, 2007
176 ## <peter@fry-it.com>
177 ## download here: http://www.peterbe.com/plog/html2plaintext
180 tree = etree.fromstring(html, parser=etree.HTMLParser())
182 if body_id is not None:
183 source = tree.xpath('//*[@id=%s]' % (body_id,))
185 source = tree.xpath('//body')
191 for link in tree.findall('.//a'):
192 url = link.get('href')
196 link.text = '%s [%s]' % (link.text, i)
197 url_index.append(url)
199 html = ustr(etree.tostring(tree, encoding=encoding))
200 # \r char is converted into , must remove it
201 html = html.replace(' ', '')
203 html = html.replace('<strong>', '*').replace('</strong>', '*')
204 html = html.replace('<b>', '*').replace('</b>', '*')
205 html = html.replace('<h3>', '*').replace('</h3>', '*')
206 html = html.replace('<h2>', '**').replace('</h2>', '**')
207 html = html.replace('<h1>', '**').replace('</h1>', '**')
208 html = html.replace('<em>', '/').replace('</em>', '/')
209 html = html.replace('<tr>', '\n')
210 html = html.replace('</p>', '\n')
211 html = re.sub('<br\s*/?>', '\n', html)
212 html = re.sub('<.*?>', ' ', html)
213 html = html.replace(' ' * 2, ' ')
214 html = html.replace('>', '>')
215 html = html.replace('<', '<')
218 html = ''.join([x.strip() for x in html.splitlines(True)])
220 for i, url in enumerate(url_index):
223 html += ustr('[%s] %s\n') % (i + 1, url)
227 def plaintext2html(text, container_tag=False):
228 """ Convert plaintext into html. Content of the text is escaped to manage
229 html entities, using cgi.escape().
230 - all \n,\r are replaced by <br />
231 - enclose content into <p>
232 - 2 or more consecutive <br /> are considered as paragraph breaks
234 :param string container_tag: container of the html; by default the
235 content is embedded into a <div>
237 text = cgi.escape(ustr(text))
239 # 1. replace \n and \r
240 text = text.replace('\n', '<br/>')
241 text = text.replace('\r', '<br/>')
243 # 2-3: form paragraphs
246 br_tags = re.compile(r'(([<]\s*[bB][rR]\s*\/?[>]\s*){2,})')
247 for item in re.finditer(br_tags, text):
248 final += text[idx:item.start()] + '</p><p>'
250 final += text[idx:] + '</p>'
254 final = '<%s>%s</%s>' % (container_tag, final, container_tag)
257 def append_content_to_html(html, content, plaintext=True, preserve=False, container_tag=False):
258 """ Append extra content at the end of an HTML snippet, trying
259 to locate the end of the HTML document (</body>, </html>, or
260 EOF), and converting the provided content in html unless ``plaintext``
262 Content conversion can be done in two ways:
263 - wrapping it into a pre (preserve=True)
264 - use plaintext2html (preserve=False, using container_tag to wrap the
266 A side-effect of this method is to coerce all HTML tags to
267 lowercase in ``html``, and strip enclosing <html> or <body> tags in
268 content if ``plaintext`` is False.
270 :param str html: html tagsoup (doesn't have to be XHTML)
271 :param str content: extra content to append
272 :param bool plaintext: whether content is plaintext and should
273 be wrapped in a <pre/> tag.
274 :param bool preserve: if content is plaintext, wrap it into a <pre>
275 instead of converting it into html
278 if plaintext and preserve:
279 content = u'\n<pre>%s</pre>\n' % ustr(content)
281 content = '\n%s\n' % plaintext2html(content, container_tag)
283 content = re.sub(r'(?i)(</?html.*>|</?body.*>|<!\W*DOCTYPE.*>)', '', content)
284 content = u'\n%s\n' % ustr(content)
285 # Force all tags to lowercase
286 html = re.sub(r'(</?)\W*(\w+)([ >])',
287 lambda m: '%s%s%s' % (m.group(1), m.group(2).lower(), m.group(3)), html)
288 insert_location = html.find('</body>')
289 if insert_location == -1:
290 insert_location = html.find('</html>')
291 if insert_location == -1:
292 return '%s%s' % (html, content)
293 return '%s%s%s' % (html[:insert_location], content, html[insert_location:])
295 #----------------------------------------------------------
297 #----------------------------------------------------------
299 email_re = re.compile(r"""
300 ([a-zA-Z][\w\.-]*[a-zA-Z0-9] # username part
302 [a-zA-Z0-9][\w\.-]* # domain must start with a letter ... Ged> why do we include a 0-9 then?
307 res_re = re.compile(r"\[([0-9]+)\]", re.UNICODE)
308 command_re = re.compile("^Set-([a-z]+) *: *(.+)$", re.I + re.UNICODE)
310 # Updated in 7.0 to match the model name as well
311 # Typical form of references is <timestamp-openerp-record_id-model_name@domain>
312 # group(1) = the record ID ; group(2) = the model (if any) ; group(3) = the domain
313 reference_re = re.compile("<.*-open(?:object|erp)-(\\d+)(?:-([\w.]+))?.*@(.*)>", re.UNICODE)
315 def generate_tracking_message_id(res_id):
316 """Returns a string that can be used in the Message-ID RFC822 header field
318 Used to track the replies related to a given object thanks to the "In-Reply-To"
319 or "References" fields that Mail User Agents will set.
322 rnd = random.SystemRandom().random()
323 except NotImplementedError:
324 rnd = random.random()
325 rndstr = ("%.15f" % rnd)[2:]
326 return "<%.15f.%s-openerp-%s@%s>" % (time.time(), rndstr, res_id, socket.gethostname())
328 def email_send(email_from, email_to, subject, body, email_cc=None, email_bcc=None, reply_to=False,
329 attachments=None, message_id=None, references=None, openobject_id=False, debug=False, subtype='plain', headers=None,
330 smtp_server=None, smtp_port=None, ssl=False, smtp_user=None, smtp_password=None, cr=None, uid=None):
331 """Low-level function for sending an email (deprecated).
333 :deprecate: since OpenERP 6.1, please use ir.mail_server.send_email() instead.
334 :param email_from: A string used to fill the `From` header, if falsy,
335 config['email_from'] is used instead. Also used for
336 the `Reply-To` header if `reply_to` is not provided
337 :param email_to: a sequence of addresses to send the mail to.
340 # If not cr, get cr from current thread database
343 db_name = getattr(threading.currentThread(), 'dbname', None)
345 local_cr = cr = pooler.get_db(db_name).cursor()
347 raise Exception("No database cursor found, please pass one explicitly")
351 mail_server_pool = pooler.get_pool(cr.dbname).get('ir.mail_server')
353 # Pack Message into MIME Object
354 email_msg = mail_server_pool.build_email(email_from, email_to, subject, body, email_cc, email_bcc, reply_to,
355 attachments, message_id, references, openobject_id, subtype, headers=headers)
357 res = mail_server_pool.send_email(cr, uid or 1, email_msg, mail_server_id=None,
358 smtp_server=smtp_server, smtp_port=smtp_port, smtp_user=smtp_user, smtp_password=smtp_password,
359 smtp_encryption=('ssl' if ssl else None), smtp_debug=debug)
361 _logger.exception("tools.email_send failed to deliver email")
368 def email_split(text):
369 """ Return a list of the email addresses found in ``text`` """
372 return [addr[1] for addr in getaddresses([text])
373 # getaddresses() returns '' when email parsing fails, and
374 # sometimes returns emails without at least '@'. The '@'
375 # is strictly required in RFC2822's `addr-spec`.