Produits Entr'ouvert » w.c.s. » Au quotidien

omit = wcs/ctl/Bouncers/*.py wcs/qommon/vendor/*.py

omit = wcs/ctl/Bouncers/*.py wcs/qommon/vendor/*.py

Email bounce detection code (wcs/ctl/Bounces/*) from Mailman:

 # http://www.gnu.org/software/mailman/

 #

 # This program is free software; you can redistribute it and/or

 # modify it under the terms of the GNU General Public License

 # as published by the Free Software Foundation; either version 2

 # of the License, or (at your option) any later version.

omit = wcs/ctl/Bouncers/*.py wcs/qommon/vendor/*.py

omit = wcs/ctl/Bouncers/*.py wcs/qommon/vendor/*.py

from wcs.qommon.bounces import Bounce

def test_bounces(pub):

    create_superuser(pub)

    app = login(get_app(pub))

    resp = app.get('/backoffice/')

    assert not 'bounces' in resp.body

    resp = app.get('/backoffice/settings/emails/options')

    resp.form['from'] = 'noreply@localhost'

    resp.form['bounce_handler'].checked = True

    resp = resp.form.submit('submit')

    pub.reload_cfg()

    assert pub.cfg['emails']['bounce_handler']

    bounce = Bounce()

    bounce.arrival_time = time.time()

    bounce.bounce_message = 'foobar'

    bounce.addrs = ['foo@localhost']

    bounce.email_type = 'change-password-request'

    bounce.original_rcpts = ['foo@localhost', 'bar@localhost']

    msg = MIMEText('hello world')

    msg['Subject'] = 'hello world'

    bounce.original_message = msg.as_string()

    bounce.store()

    resp = app.get('/backoffice/')

    assert 'bounces' in resp.body

    resp = resp.click('Bounces')

    resp = resp.click(href='%s/' % bounce.id, index=0)

    assert 'hello world' in resp.body

    resp = app.get('/backoffice/bounces/')

    resp = resp.click(href='%s/delete' % bounce.id, index=0)

    resp = resp.form.submit('cancel').follow()

    assert Bounce.count() == 1

    resp = resp.click(href='%s/delete' % bounce.id, index=0)

    resp = resp.form.submit('delete')

    assert Bounce.count() == 0

import email

from wcs.ctl.process_bounce import CmdProcessBounce

def test_normal_email():

    msg = email.message_from_string('test')

    msg['From'] = 'bar@example.net'

    msg['To'] = 'foo@example.net'

    addrs = CmdProcessBounce.get_bounce_addrs(msg)

    assert addrs is None

def test_bounce_email():

    msg = email.message_from_string('test')

    msg['From'] = 'bar@example.net'

    msg['To'] = 'foo@example.net'

    # this is how exim adds failed recipients in its outgoing bounce emails

    msg['x-failed-recipients'] = 'baz@example.net'

    addrs = CmdProcessBounce.get_bounce_addrs(msg)

    assert addrs == ['baz@example.net']

from wcs.ctl.process_bounce import CmdProcessBounce

def test_get_bounce_addrs():

    msg = MIMEText('Hello world')

    assert CmdProcessBounce.get_bounce_addrs(msg) is None

    msg = MIMEMultipart(_subtype='mixed')

    msg.attach(MIMEText('Hello world'))

    msg.attach(MIMEText('<p>Hello world</p>', _subtype='html'))

    assert CmdProcessBounce.get_bounce_addrs(msg) is None

    msg = MIMEText('Hello world')

    msg['x-failed-recipients'] = 'foobar@localhost'

    assert CmdProcessBounce.get_bounce_addrs(msg) == ['foobar@localhost']

    msg = MIMEText('''failed addresses follow:

                      foobar@localhost

                      message text follows:''')

    assert CmdProcessBounce.get_bounce_addrs(msg) == ['foobar@localhost']

# w.c.s. - web application for online forms

# Copyright (C) 2005-2010  Entr'ouvert

#

# This program is free software; you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation; either version 2 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, see <http://www.gnu.org/licenses/>.

import time

import email.Parser

from django.utils.encoding import force_text

from quixote import get_response, redirect

from quixote.directory import Directory

from quixote.html import htmltext, TemplateIO

from ..qommon import _

from ..qommon import errors

from ..qommon import misc

from ..qommon.bounces import Bounce

from ..qommon.backoffice.menu import html_top

from ..qommon.admin.menu import command_icon

from ..qommon.form import *

from ..qommon.misc import get_cfg

def get_email_type_label(type):

    from .settings import EmailsDirectory

    return EmailsDirectory.emails_dict.get(type, {}).get('description')

class BouncePage(Directory):

    _q_exports = ['', 'delete']

    def __init__(self, component):

        self.bounce = Bounce.get(component)

        get_response().breadcrumb.append((component + '/', _('bounce')))

    def _q_index(self):

        html_top('bounces', title = _('Bounce'))

        r = TemplateIO(html=True)

        r += htmltext('<div class="form">')

        if self.bounce.email_type:

            r += htmltext('<div class="title">%s</div>') % _('Email Type')

            r += htmltext('<div class="StringWidget content">')

            r += _(get_email_type_label(self.bounce.email_type))

            r += htmltext('</div>')

        r += htmltext('<div class="title">%s</div>') % _('Arrival Time')

        r += htmltext('<div class="StringWidget content">')

        r += misc.localstrftime(time.localtime(self.bounce.arrival_time))

        r += htmltext('</div>')

        if self.bounce.addrs:

            r += htmltext('<div class="title">%s</div>') % _('Failed Addresses')

            r += htmltext('<div class="StringWidget content">')

            r += ', '.join(self.bounce.addrs)

            r += htmltext('</div>')

        if self.bounce.original_rcpts:

            r += htmltext('<div class="title">%s</div>') % _('Original Recipients')

            r += htmltext('<div class="StringWidget content">')

            r += ', '.join(self.bounce.original_rcpts)

            r += htmltext('</div>')

        r += htmltext('<div class="title">%s</div>') % _('Bounce Message')

        r += htmltext('<div class="StringWidget content">')

        r += htmltext('<pre style="max-height: 20em;">')

        r += self.bounce.bounce_message

        r += htmltext('</pre>')

        r += htmltext('</div>')

        if self.bounce.original_message:

            parser = email.Parser.Parser()

            msg = parser.parsestr(self.bounce.original_message)

            if msg.is_multipart():

                for m in msg.get_payload():

                    if m.get_content_type() == 'text/plain':

                        break

                else:

                    m = None

            elif msg.get_content_type() == 'text/plain':

                m = msg

            else:

                m = None

            r += htmltext('<div class="title">%s</div>') % _('Original Message')

            r += htmltext('<div class="StringWidget content">')

            r += _('Subject: ')

            subject, charset = email.Header.decode_header(msg['Subject'])[0]

            if charset:

                encoding = get_publisher().site_charset

                r += force_text(subject, charset).encode(encoding)

            else:

                r += subject

            r += htmltext('</div>')

            if m:

                r += htmltext('<div class="StringWidget content">')

                r += htmltext('<pre>')

                r += m.get_payload()

                r += htmltext('</pre>')

                r += htmltext('</div>')

        r += htmltext('</div>') # form

        return r.getvalue()

    def delete(self):

        form = Form(enctype='multipart/form-data')

        form.widgets.append(HtmlWidget('<p>%s</p>' % _(

                        'You are about to irrevocably delete this bounce.')))

        form.add_submit('delete', _('Delete'))

        form.add_submit('cancel', _('Cancel'))

        if form.get_widget('cancel').parse():

            return redirect('..')

        if not form.is_submitted() or form.has_errors():

            get_response().breadcrumb.append(('delete', _('Delete')))

            html_top('bounces', title = _('Delete Bounce'))

            r = TemplateIO(html=True)

            r += htmltext('<h2>%s</h2>') % _('Deleting Bounce')

            r += form.render()

            return r.getvalue()

        else:

            self.bounce.remove_self()

            return redirect('..')

class BouncesDirectory(Directory):

    _q_exports = ['']

    def _q_traverse(self, path):

        get_response().breadcrumb.append( ('bounces/', _('Bounces')) )

        return Directory._q_traverse(self, path)

    def is_visible(self, *args):

        return bool(get_cfg('emails', {}).get('bounce_handler') is True)

    def _q_index(self):

        html_top('bounces', title = _('Bounces'))

        bounces = Bounce.select(ignore_errors=True)

        bounces.sort(lambda x,y: cmp(x.arrival_time, y.arrival_time))

        r = TemplateIO(html=True)

        r += htmltext('<ul class="biglist">')

        for bounce in bounces:

            r += htmltext('<li>')

            r += htmltext('<strong class="label">')

            r += misc.localstrftime(time.localtime(bounce.arrival_time))

            if bounce.email_type:

                r += ' - '

                r += _(get_email_type_label(bounce.email_type))

            r += htmltext('</strong>')

            r += htmltext('<p class="details">')

            if bounce.addrs:

                r += ', '.join(bounce.addrs)

            r += htmltext('</p>')

            r += htmltext('<p class="commands">')

            r += command_icon('%s/' % bounce.id, 'view')

            r += command_icon('%s/delete' % bounce.id, 'remove', popup = True)

            r += htmltext('</p></li>')

        r += htmltext('</ul>')

        return r.getvalue()

    def _q_lookup(self, component):

        try:

            return BouncePage(component)

        except KeyError:

            raise errors.TraversalError()

                ('bounces', N_('Bounces')),

import wcs.admin.bounces

    bounces = wcs.admin.bounces.BouncesDirectory()

        ('bounces/', N_('Bounces'), {'check_display_function': bounces.is_visible}),

# Copyright (C) 1998-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Contains all the common functionality for msg bounce scanning API.

This module can also be used as the basis for a bounce detection testing

framework.  When run as a script, it expects two arguments, the listname and

the filename containing the bounce message.

"""

import sys

# If a bounce detector returns Stop, that means to just discard the message.

# An example is warning messages for temporary delivery problems.  These

# shouldn't trigger a bounce notification, but we also don't want to send them

# on to the list administrator.

class _Stop:

    pass

Stop = _Stop()

BOUNCE_PIPELINE = [

    'DSN',

    'Qmail',

    'Postfix',

    'Yahoo',

    'Caiwireless',

    'Exchange',

    'Exim',

    'Netscape',

    'Compuserve',

    'Microsoft',

    'GroupWise',

    'SMTP32',

    'SimpleMatch',

    'SimpleWarning',

    'Yale',

    'LLNL',

    ]

# msg must be a mimetools.Message

def ScanMessages(mlist, msg):

    for module in BOUNCE_PIPELINE:

        modname = 'Mailman.Bouncers.' + module

        __import__(modname)

        addrs = sys.modules[modname].process(msg)

        if addrs:

            # Return addrs even if it is Stop. BounceRunner needs this info.

            return addrs

    return []

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Parse mystery style generated by MTA at caiwireless.net."""

import re

import email

from cStringIO import StringIO

tcre = re.compile(r'the following recipients did not receive this message:',

                  re.IGNORECASE)

acre = re.compile(r'<(?P<addr>[^>]*)>')

def process(msg):

    if msg.get_content_type() <> 'multipart/mixed':

        return None

    # simple state machine

    #     0 == nothing seen

    #     1 == tag line seen

    state = 0

    # This format thinks it's a MIME, but it really isn't

    for line in email.Iterators.body_line_iterator(msg):

        line = line.strip()

        if state == 0 and tcre.match(line):

            state = 1

        elif state == 1 and line:

            mo = acre.match(line)

            if not mo:

                return None

            return [mo.group('addr')]

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Compuserve has its own weird format for bounces."""

import re

import email

dcre = re.compile(r'your message could not be delivered', re.IGNORECASE)

acre = re.compile(r'Invalid receiver address: (?P<addr>.*)')

def process(msg):

    # simple state machine

    #    0 = nothing seen yet

    #    1 = intro line seen

    state = 0

    addrs = []

    for line in email.Iterators.body_line_iterator(msg):

        if state == 0:

            mo = dcre.search(line)

            if mo:

                state = 1

        elif state == 1:

            mo = dcre.search(line)

            if mo:

                break

            mo = acre.search(line)

            if mo:

                addrs.append(mo.group('addr'))

    return addrs

# Copyright (C) 1998-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Parse RFC 3464 (i.e. DSN) bounce formats.

RFC 3464 obsoletes 1894 which was the old DSN standard.  This module has not

been audited for differences between the two.

"""

from email.Iterators import typed_subpart_iterator

from email.Utils import parseaddr

from cStringIO import StringIO

from BouncerAPI import Stop

try:

    True, False

except NameError:

    True = 1

    False = 0

def check(msg):

    # Iterate over each message/delivery-status subpart

    addrs = []

    for part in typed_subpart_iterator(msg, 'message', 'delivery-status'):

        if not part.is_multipart():

            # Huh?

            continue

        # Each message/delivery-status contains a list of Message objects

        # which are the header blocks.  Iterate over those too.

        for msgblock in part.get_payload():

            # We try to dig out the Original-Recipient (which is optional) and

            # Final-Recipient (which is mandatory, but may not exactly match

            # an address on our list).  Some MTA's also use X-Actual-Recipient

            # as a synonym for Original-Recipient, but some apparently use

            # that for other purposes :(

            #

            # Also grok out Action so we can do something with that too.

            action = msgblock.get('action', '').lower()

            # Some MTAs have been observed that put comments on the action.

            if action.startswith('delayed'):

                return Stop

            if not action.startswith('fail'):

                # Some non-permanent failure, so ignore this block

                continue

            params = []

            foundp = False

            for header in ('original-recipient', 'final-recipient'):

                for k, v in msgblock.get_params([], header):

                    if k.lower() == 'rfc822':

                        foundp = True

                    else:

                        params.append(k)

                if foundp:

                    # Note that params should already be unquoted.

                    addrs.extend(params)

                    break

                else:

                    # MAS: This is a kludge, but SMTP-GATEWAY01.intra.home.dk

                    # has a final-recipient with an angle-addr and no

                    # address-type parameter at all. Non-compliant, but ...

                    for param in params:

                        if param.startswith('<') and param.endswith('>'):

                            addrs.append(param[1:-1])

    # Uniquify

    rtnaddrs = {}

    for a in addrs:

        if a is not None:

            realname, a = parseaddr(a)

            rtnaddrs[a] = True

    return rtnaddrs.keys()

def process(msg):

    # A DSN has been seen wrapped with a "legal disclaimer" by an outgoing MTA

    # in a multipart/mixed outer part.

    if msg.is_multipart() and msg.get_content_subtype() == 'mixed':

        msg = msg.get_payload()[0]

    # The report-type parameter should be "delivery-status", but it seems that

    # some DSN generating MTAs don't include this on the Content-Type: header,

    # so let's relax the test a bit.

    if not msg.is_multipart() or msg.get_content_subtype() <> 'report':

        return None

    return check(msg)

# Copyright (C) 2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Recognizes (some) Microsoft Exchange formats."""

import re

import email.Iterators

scre = re.compile('did not reach the following recipient')

ecre = re.compile('MSEXCH:')

a1cre = re.compile('SMTP=(?P<addr>[^;]+); on ')

a2cre = re.compile('(?P<addr>[^ ]+) on ')

def process(msg):

    addrs = {}

    it = email.Iterators.body_line_iterator(msg)

    # Find the start line

    for line in it:

        if scre.search(line):

            break

    else:

        return []

    # Search each line until we hit the end line

    for line in it:

        if ecre.search(line):

            break

        mo = a1cre.search(line)

        if not mo:

            mo = a2cre.search(line)

        if mo:

            addrs[mo.group('addr')] = 1

    return addrs.keys()

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Parse bounce messages generated by Exim.

Exim adds an X-Failed-Recipients: header to bounce messages containing

an `addresslist' of failed addresses.

"""

from email.Utils import getaddresses

def process(msg):

    all = msg.get_all('x-failed-recipients', [])

    return [a for n, a in getaddresses(all)]

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""This appears to be the format for Novell GroupWise and NTMail

X-Mailer: Novell GroupWise Internet Agent 5.5.3.1

X-Mailer: NTMail v4.30.0012

X-Mailer: Internet Mail Service (5.5.2653.19)

"""

import re

from email.Message import Message

from cStringIO import StringIO

acre = re.compile(r'<(?P<addr>[^>]*)>')

def find_textplain(msg):

    if msg.get_content_type() == 'text/plain':

        return msg

    if msg.is_multipart:

        for part in msg.get_payload():

            if not isinstance(part, Message):

                continue

            ret = find_textplain(part)

            if ret:

                return ret

    return None

def process(msg):

    if msg.get_content_type() <> 'multipart/mixed' or not msg['x-mailer']:

        return None

    addrs = {}

    # find the first text/plain part in the message

    textplain = find_textplain(msg)

    if not textplain:

        return None

    body = StringIO(textplain.get_payload())

    while 1:

        line = body.readline()

        if not line:

            break

        mo = acre.search(line)

        if mo:

            addrs[mo.group('addr')] = 1

        elif '@' in line:

            i = line.find(' ')

            if i == 0:

                continue

            if i < 0:

                addrs[line] = 1

            else:

                addrs[line[:i]] = 1

    return addrs.keys()

# Copyright (C) 2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""LLNL's custom Sendmail bounce message."""

import re

import email

acre = re.compile(r',\s*(?P<addr>\S+@[^,]+),', re.IGNORECASE)

def process(msg):

    for line in email.Iterators.body_line_iterator(msg):

        mo = acre.search(line)

        if mo:

            return [mo.group('addr')]

    return []

# Copyright (C) 1998-2003 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Microsoft's `SMTPSVC' nears I kin tell."""

import re

from cStringIO import StringIO

from types import ListType

scre = re.compile(r'transcript of session follows', re.IGNORECASE)

def process(msg):

    if msg.get_content_type() <> 'multipart/mixed':

        return None

    # Find the first subpart, which has no MIME type

    try:

        subpart = msg.get_payload(0)

    except IndexError:

        # The message *looked* like a multipart but wasn't

        return None

    data = subpart.get_payload()

    if isinstance(data, ListType):

        # The message is a multi-multipart, so not a matching bounce

        return None

    body = StringIO(data)

    state = 0

    addrs = []

    while 1:

        line = body.readline()

        if not line:

            break

        if state == 0:

            if scre.search(line):

                state = 1

        if state == 1:

            if '@' in line:

                addrs.append(line)

    return addrs

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Netscape Messaging Server bounce formats.

I've seen at least one NMS server version 3.6 (envy.gmp.usyd.edu.au) bounce

messages of this format.  Bounces come in DSN MIME format, but don't include

any -Recipient: headers.  Gotta just parse the text :(

NMS 4.1 (dfw-smtpin1.email.verio.net) seems even worse, but we'll try to

decipher the format here too.

"""

import re

from cStringIO import StringIO

pcre = re.compile(

    r'This Message was undeliverable due to the following reason:',

    re.IGNORECASE)

acre = re.compile(

    r'(?P<reply>please reply to)?.*<(?P<addr>[^>]*)>',

    re.IGNORECASE)

def flatten(msg, leaves):

    # give us all the leaf (non-multipart) subparts

    if msg.is_multipart():

        for part in msg.get_payload():

            flatten(part, leaves)

    else:

        leaves.append(msg)

def process(msg):

    # Sigh.  Some show NMS 3.6's show

    #     multipart/report; report-type=delivery-status

    # and some show

    #     multipart/mixed;

    if not msg.is_multipart():

        return None

    # We're looking for a text/plain subpart occuring before a

    # message/delivery-status subpart.

    plainmsg = None

    leaves = []

    flatten(msg, leaves)

    for i, subpart in zip(range(len(leaves)-1), leaves):

        if subpart.get_content_type() == 'text/plain':

            plainmsg = subpart

            break

    if not plainmsg:

        return None

    # Total guesswork, based on captured examples...

    body = StringIO(plainmsg.get_payload())

    addrs = []

    while 1:

        line = body.readline()

        if not line:

            break

        mo = pcre.search(line)

        if mo:

            # We found a bounce section, but I have no idea what the official

            # format inside here is.  :(  We'll just search for <addr>

            # strings.

            while 1:

                line = body.readline()

                if not line:

                    break

                mo = acre.search(line)

                if mo and not mo.group('reply'):

                    addrs.append(mo.group('addr'))

    return addrs

# Copyright (C) 1998-2003 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Parse bounce messages generated by Postfix.

This also matches something called `Keftamail' which looks just like Postfix

bounces with the word Postfix scratched out and the word `Keftamail' written

in in crayon.

It also matches something claiming to be `The BNS Postfix program', and

`SMTP_Gateway'.  Everybody's gotta be different, huh?

"""

import re

from cStringIO import StringIO

def flatten(msg, leaves):

    # give us all the leaf (non-multipart) subparts

    if msg.is_multipart():

        for part in msg.get_payload():

            flatten(part, leaves)

    else:

        leaves.append(msg)

# are these heuristics correct or guaranteed?

pcre = re.compile(r'[ \t]*the\s*(bns)?\s*(postfix|keftamail|smtp_gateway)',

                  re.IGNORECASE)

rcre = re.compile(r'failure reason:$', re.IGNORECASE)

acre = re.compile(r'<(?P<addr>[^>]*)>:')

def findaddr(msg):

    addrs = []

    body = StringIO(msg.get_payload())

    # simple state machine

    #     0 == nothing found

    #     1 == salutation found

    state = 0

    while 1:

        line = body.readline()

        if not line:

            break

        # preserve leading whitespace

        line = line.rstrip()

        # yes use match to match at beginning of string

        if state == 0 and (pcre.match(line) or rcre.match(line)):

            state = 1

        elif state == 1 and line:

            mo = acre.search(line)

            if mo:

                addrs.append(mo.group('addr'))

            # probably a continuation line

    return addrs

def process(msg):

    if msg.get_content_type() not in ('multipart/mixed', 'multipart/report'):

        return None

    # We're looking for the plain/text subpart with a Content-Description: of

    # `notification'.

    leaves = []

    flatten(msg, leaves)

    for subpart in leaves:

        if subpart.get_content_type() == 'text/plain' and \

           subpart.get('content-description', '').lower() == 'notification':

            # then...

            return findaddr(subpart)

    return None

# Copyright (C) 1998-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Parse bounce messages generated by qmail.

Qmail actually has a standard, called QSBMF (qmail-send bounce message

format), as described in

    http://cr.yp.to/proto/qsbmf.txt

This module should be conformant.

"""

import re

import email.Iterators

# Other (non-standard?) intros have been observed in the wild.

introtags = [

    'Hi. This is the',

    "We're sorry. There's a problem",

    'Check your send e-mail address.'

    ]

acre = re.compile(r'<(?P<addr>[^>]*)>:')

def process(msg):

    addrs = []

    # simple state machine

    #    0 = nothing seen yet

    #    1 = intro paragraph seen

    #    2 = recip paragraphs seen

    state = 0

    for line in email.Iterators.body_line_iterator(msg):

        line = line.strip()

        if state == 0:

            for introtag in introtags:

                if line.startswith(introtag):

                    state = 1

                    break

        elif state == 1 and not line:

            # Looking for the end of the intro paragraph

            state = 2

        elif state == 2:

            if line.startswith('-'):

                # We're looking at the break paragraph, so we're done

                break

            # At this point we know we must be looking at a recipient

            # paragraph

            mo = acre.match(line)

            if mo:

                addrs.append(mo.group('addr'))

            # Otherwise, it must be a continuation line, so just ignore it

        # Not looking at anything in particular

    return addrs

# Copyright (C) 1998-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Something which claims

X-Mailer: <SMTP32 vXXXXXX>

What the heck is this thing?  Here's a recent host:

% telnet 207.51.255.218 smtp

Trying 207.51.255.218...

Connected to 207.51.255.218.

Escape character is '^]'.

220 X1 NT-ESMTP Server 208.24.118.205 (IMail 6.00 45595-15)

"""

import re

import email

ecre = re.compile('original message follows', re.IGNORECASE)

acre = re.compile(r'''

    (                                             # several different prefixes

    user\ mailbox[^:]*:                           # have been spotted in the

    |delivery\ failed[^:]*:                       # wild...

    |unknown\ user[^:]*:

    |undeliverable\ +to

    )

    \s*                                           # space separator

    (?P<addr>.*)                                  # and finally, the address

    ''', re.IGNORECASE | re.VERBOSE)

def process(msg):

    mailer = msg.get('x-mailer', '')

    if not mailer.startswith('<SMTP32 v'):

        return

    addrs = {}

    for line in email.Iterators.body_line_iterator(msg):

        if ecre.search(line):

            break

        mo = acre.search(line)

        if mo:

            addrs[mo.group('addr')] = 1

    return addrs.keys()

# Copyright (C) 1998-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Recognizes simple heuristically delimited bounces."""

import re

import email.Iterators

def _c(pattern):

    return re.compile(pattern, re.IGNORECASE)

# This is a list of tuples of the form

#

#     (start cre, end cre, address cre)

#

# where `cre' means compiled regular expression, start is the line just before

# the bouncing address block, end is the line just after the bouncing address

# block, and address cre is the regexp that will recognize the addresses.  It

# must have a group called `addr' which will contain exactly and only the

# address that bounced.

PATTERNS = [

    # sdm.de

    (_c('here is your list of failed recipients'),

     _c('here is your returned mail'),

     _c(r'<(?P<addr>[^>]*)>')),

    # sz-sb.de, corridor.com, nfg.nl

    (_c('the following addresses had'),

     _c('transcript of session follows'),

     _c(r'<(?P<fulladdr>[^>]*)>|\(expanded from: <?(?P<addr>[^>)]*)>?\)')),

    # robanal.demon.co.uk

    (_c('this message was created automatically by mail delivery software'),

     _c('original message follows'),

     _c('rcpt to:\s*<(?P<addr>[^>]*)>')),

    # s1.com (InterScan E-Mail VirusWall NT ???)

    (_c('message from interscan e-mail viruswall nt'),

     _c('end of message'),

     _c('rcpt to:\s*<(?P<addr>[^>]*)>')),

    # Smail

    (_c('failed addresses follow:'),

     _c('message text follows:'),

     _c(r'\s*(?P<addr>\S+@\S+)')),

    # newmail.ru

    (_c('This is the machine generated message from mail service.'),

     _c('--- Below the next line is a copy of the message.'),

     _c('<(?P<addr>[^>]*)>')),

    # turbosport.com runs something called `MDaemon 3.5.2' ???

    (_c('The following addresses did NOT receive a copy of your message:'),

     _c('--- Session Transcript ---'),

     _c('[>]\s*(?P<addr>.*)$')),

    # usa.net

    (_c('Intended recipient:\s*(?P<addr>.*)$'),

     _c('--------RETURNED MAIL FOLLOWS--------'),

     _c('Intended recipient:\s*(?P<addr>.*)$')),

    # hotpop.com

    (_c('Undeliverable Address:\s*(?P<addr>.*)$'),

     _c('Original message attached'),

     _c('Undeliverable Address:\s*(?P<addr>.*)$')),

    # Another demon.co.uk format

    (_c('This message was created automatically by mail delivery'),

     _c('^---- START OF RETURNED MESSAGE ----'),

     _c("addressed to '(?P<addr>[^']*)'")),

    # Prodigy.net full mailbox

    (_c("User's mailbox is full:"),

     _c('Unable to deliver mail.'),

     _c("User's mailbox is full:\s*<(?P<addr>[^>]*)>")),

    # Microsoft SMTPSVC

    (_c('The email below could not be delivered to the following user:'),

     _c('Old message:'),

     _c('<(?P<addr>[^>]*)>')),

    # Yahoo on behalf of other domains like sbcglobal.net

    (_c('Unable to deliver message to the following address\(es\)\.'),

     _c('--- Original message follows\.'),

     _c('<(?P<addr>[^>]*)>:')),

    # kundenserver.de

    (_c('A message that you sent could not be delivered'),

     _c('^---'),

     _c('<(?P<addr>[^>]*)>')),

    # another kundenserver.de

    (_c('A message that you sent could not be delivered'),

     _c('^---'),

     _c('^(?P<addr>[^\s@]+@[^\s@:]+):')),

    # thehartford.com

    (_c('Delivery to the following recipients failed'),

     _c("Bogus - there actually isn't anything"),

     _c('^\s*(?P<addr>[^\s@]+@[^\s@]+)\s*$')),

    # and another thehartfod.com/hartfordlife.com

    (_c('^Your message\s*$'),

     _c('^because:'),

     _c('^\s*(?P<addr>[^\s@]+@[^\s@]+)\s*$')),

    # kviv.be (NTMail)

    (_c('^Unable to deliver message to'),

     _c(r'\*+\s+End of message\s+\*+'),

     _c('<(?P<addr>[^>]*)>')),

    # earthlink.net supported domains

    (_c('^Sorry, unable to deliver your message to'),

     _c('^A copy of the original message'),

     _c('\s*(?P<addr>[^\s@]+@[^\s@]+)\s+')),

    # ademe.fr

    (_c('^A message could not be delivered to:'),

     _c('^Subject:'),

     _c('^\s*(?P<addr>[^\s@]+@[^\s@]+)\s*$')),

    # andrew.ac.jp

    (_c('^Invalid final delivery userid:'),

     _c('^Original message follows.'),

     _c('\s*(?P<addr>[^\s@]+@[^\s@]+)\s*$')),

    # E500_SMTP_Mail_Service@lerctr.org

    (_c('------ Failed Recipients ------'),

     _c('-------- Returned Mail --------'),

     _c('<(?P<addr>[^>]*)>')),

    # cynergycom.net

    (_c('A message that you sent could not be delivered'),

     _c('^---'),

     _c('(?P<addr>[^\s@]+@[^\s@)]+)')),

    # Next one goes here...

    ]

def process(msg, patterns=None):

    if patterns is None:

        patterns = PATTERNS

    # simple state machine

    #     0 = nothing seen yet

    #     1 = intro seen

    addrs = {}

    # MAS: This is a mess. The outer loop used to be over the message

    # so we only looped through the message once.  Looping through the

    # message for each set of patterns is obviously way more work, but

    # if we don't do it, problems arise because scre from the wrong

    # pattern set matches first and then acre doesn't match.  The

    # alternative is to split things into separate modules, but then

    # we process the message multiple times anyway.

    for scre, ecre, acre in patterns:

        state = 0

        for line in email.Iterators.body_line_iterator(msg):

            if state == 0:

                if scre.search(line):

                    state = 1

            if state == 1:

                mo = acre.search(line)

                if mo:

                    addr = mo.group('addr')

                    if addr:

                        addrs[mo.group('addr')] = 1

                elif ecre.search(line):

                    break

        if addrs:

            break

    return addrs.keys()

# Copyright (C) 2001-2006 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,

# USA.

"""Recognizes simple heuristically delimited warnings."""

from BouncerAPI import Stop

from SimpleMatch import _c

from SimpleMatch import process as _process

# This is a list of tuples of the form

#

#     (start cre, end cre, address cre)

#

# where `cre' means compiled regular expression, start is the line just before

# the bouncing address block, end is the line just after the bouncing address

# block, and address cre is the regexp that will recognize the addresses.  It

# must have a group called `addr' which will contain exactly and only the

# address that bounced.

patterns = [

    # pop3.pta.lia.net

    (_c('The address to which the message has not yet been delivered is'),

     _c('No action is required on your part'),

     _c(r'\s*(?P<addr>\S+@\S+)\s*')),

    # Next one goes here...

    ]

def process(msg):

    if _process(msg, patterns):

        # It's a recognized warning so stop now

        return Stop

    else:

        return []

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Yahoo! has its own weird format for bounces."""

import re

import email

from email.Utils import parseaddr

tcre = re.compile(r'message\s+from\s+yahoo\.\S+', re.IGNORECASE)

acre = re.compile(r'<(?P<addr>[^>]*)>:')

ecre = re.compile(r'--- Original message follows')

def process(msg):

    # Yahoo! bounces seem to have a known subject value and something called

    # an x-uidl: header, the value of which seems unimportant.

    sender = parseaddr(msg.get('from', '').lower())[1] or ''

    if not sender.startswith('mailer-daemon@yahoo'):

        return None

    addrs = []

    # simple state machine

    #     0 == nothing seen

    #     1 == tag line seen

    state = 0

    for line in email.Iterators.body_line_iterator(msg):

        line = line.strip()

        if state == 0 and tcre.match(line):

            state = 1

        elif state == 1:

            mo = acre.match(line)

            if mo:

                addrs.append(mo.group('addr'))

                continue

            mo = ecre.match(line)

            if mo:

                # we're at the end of the error response

                break

    return addrs

# Copyright (C) 2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

"""Yale's mail server is pretty dumb.

Its reports include the end user's name, but not the full domain.  I think we

can usually guess it right anyway.  This is completely based on examination of

the corpse, and is subject to failure whenever Yale even slightly changes

their MTA. :(

"""

import re

from cStringIO import StringIO

from email.Utils import getaddresses

scre = re.compile(r'Message not delivered to the following', re.IGNORECASE)

ecre = re.compile(r'Error Detail', re.IGNORECASE)

acre = re.compile(r'\s+(?P<addr>\S+)\s+')

def process(msg):

    if msg.is_multipart():

        return None

    try:

        whofrom = getaddresses([msg.get('from', '')])[0][1]

        if not whofrom:

            return None

        username, domain = whofrom.split('@', 1)

    except (IndexError, ValueError):

        return None

    if username.lower() <> 'mailer-daemon':

        return None

    parts = domain.split('.')

    parts.reverse()

    for part1, part2 in zip(parts, ('edu', 'yale')):

        if part1 <> part2:

            return None

    # Okay, we've established that the bounce came from the mailer-daemon at

    # yale.edu.  Let's look for a name, and then guess the relevant domains.

    names = {}

    body = StringIO(msg.get_payload())

    state = 0

    # simple state machine

    #     0 == init

    #     1 == intro found

    while 1:

        line = body.readline()

        if not line:

            break

        if state == 0 and scre.search(line):

            state = 1

        elif state == 1 and ecre.search(line):

            break

        elif state == 1:

            mo = acre.search(line)

            if mo:

                names[mo.group('addr')] = 1

    # Now we have a bunch of names, these are either @yale.edu or

    # @cs.yale.edu.  Add them both.

    addrs = []

    for name in names.keys():

        addrs.append(name + '@yale.edu')

        addrs.append(name + '@cs.yale.edu')

    return addrs

# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software Foundation, Inc.

#

# This program is free software; you can redistribute it and/or

# modify it under the terms of the GNU General Public License

# as published by the Free Software Foundation; either version 2

# of the License, or (at your option) any later version.

# 

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

# 

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software 

# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

# w.c.s. - web application for online forms

# Copyright (C) 2005-2010  Entr'ouvert

#

# This program is free software; you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation; either version 2 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, see <http://www.gnu.org/licenses/>.

import sys

import time

import os

import email.Parser

from Bouncers import BouncerAPI

from ..qommon.ctl import Command

COMMA_SPACE = ', '

class CmdProcessBounce(Command):

    name = 'process_bounce'

    def execute(self, base_options, sub_options, args):

        from ..qommon.tokens import Token

        from ..qommon.bounces import Bounce

        from .. import publisher

        try:

            publisher.WcsPublisher.configure(self.config)

            pub = publisher.WcsPublisher.create_publisher(

                    register_tld_names=False)

        except: