First iteration of email alias generation for wg- and drafts ( issue ietf-tools#713 ).

 - Legacy-Id: 5902

Author: fingon

ietf/bin/aliasutil.py

@@ -0,0 +1,123 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# -*- Python -*-
+#
+# $Id: aliasutil.py $
+#
+# Author: Markus Stenberg <mstenber@cisco.com>
+#
+"""
+
+Mailing list alias dumping utilities
+
+"""
+
+ad_replacement = {
+    # 2009 Ad replacements
+    'chris.newman@sun.com': 'alexey.melnikov@isode.com',
+    'townsley@cisco.com': 'rdroms@cisco.com',
+    'jon.peterson@neustar.biz': 'rjsparks@nostrum.com',
+    'dward@cisco.com': 'adrian@olddog.co.uk',
+
+    # 2010 AD replacements -- activate after spring ietf
+    'lisa@osafoundation.org': 'stpeter@stpeter.im',
+    'fluffy@cisco.com': 'gonzalo.camarillo@ericsson.com',
+    'rcallon@juniper.net': 'stbryant@cisco.com',
+    'pasi.eronen@nokia.com': 'turners@ieca.com',
+    'magnus.westerlund@ericsson.com': 'ietfdbh@comcast.net',
+
+    # 2011 AD replacements -- activate after spring ietf
+    'alexey.melnikov@isode.com': 'presnick@qualcomm.com',
+    'tim.polk@nist.gov': 'stephen.farrell@cs.tcd.ie',
+    'lars.eggert@nokia.com': 'wes@mti-systems.com',
+
+    #2013 AD replacements
+    'housley@vigilsec.com': 'jari.arkko@piuha.net',
+    'rdroms.ietf@gmail.com': 'ted.lemon@nominum.com',
+    'rbonica@juniper.net': 'jjaeggli@zynga.com',
+    'rjsparks@nostrum.com': 'rbarnes@bbn.com',
+    }
+
+email_replacement = {
+    'barryleiba@computer.org': 'barryleiba@gmail.com',
+    'greg.daley@eng.monash.edu.au': 'gdaley@netstarnetworks.com',
+    'radia.perlman@sun.com': 'radia@alum.mit.edu',
+    'lisa@osafoundation.org': 'lisa.dusseault@gmail.com',
+    'lisa.dusseault@messagingarchitects.com': 'lisa.dusseault@gmail.com',
+    'scott.lawrence@nortel.com': 'scottlawrenc@avaya.com',
+    'charliep@computer.org': 'charliep@computer.org, charles.perkins@earthlink.net',
+    'yaronf@checkpoint.com': 'yaronf.ietf@gmail.com',
+    'mary.barnes@nortel.com': 'mary.ietf.barnes@gmail.com',
+    'scottlawrenc@avaya.com': 'xmlscott@gmail.com',
+    'henk@ripe.net': 'henk@uijterwaal.nl',
+    'jonne.soininen@nsn.com': 'jonne.soininen@renesasmobile.com',
+    'tom.taylor@rogers.com': 'tom.taylor.stds@gmail.com',
+    'rahul@juniper.net': 'raggarwa_1@yahoo.com',
+    'dward@juniper.net': 'dward@cisco.com',
+    'alan.ford@roke.co.uk': 'alanford@cisco.com',
+    'rod.walsh@nokia.com': 'roderick.walsh@tut.fi',
+    'bob.hinden@nokia.com': 'bob.hinden@gmail.com',
+    'martin.thomson@commscope.com': 'martin.thomson@gmail.com',
+    'rjs@estacado.net': 'rjsparks@nostrum.com',
+    'rbarnes@bbn.com': 'rlb@ipv.sx',
+}
+
+
+def rewrite_email_address(email, is_ad):
+    """ Prettify the email address (and if it's empty, skip it by
+    returning None). """
+    if not email:
+        return
+    email = email.strip()
+    if not email:
+        return
+    if email[0]=='<' and email[-1] == '>':
+        email = email[1:-1]
+    # If it doesn't look like email, skip
+    if '@' not in email and '?' not in email:
+        return
+    if is_ad:
+        email = ad_replacement.get(email, email)
+    email = email_replacement.get(email, email)
+    return email
+
+def rewrite_address_list(l):
+    """ This utility function makes sure there is exactly one instance
+    of an address within the result list, and preserves order
+    (although it may not be relevant to start with) """
+    h = {}
+    for address in l:
+        #address = address.strip()
+        if h.has_key(address): continue
+        h[address] = True
+        yield address
+
+def dump_sublist(alias, f, wg, is_adlist=False):
+    if f:
+        l = f(wg)
+    else:
+        l = wg
+    if not l:
+        return
+    # Nones in the list should be skipped
+    l = filter(None, l)
+
+    # Make sure emails are sane and eliminate the Nones again for
+    # non-sane ones
+    l = [rewrite_email_address(e, is_adlist) for e in l]
+    l = filter(None, l)
+
+    # And we'll eliminate the duplicates too but preserve order
+    l = list(rewrite_address_list(l))
+    if not l:
+        return
+    try:
+        print '%s: %s' % (alias, ', '.join(l))
+    except UnicodeEncodeError:
+        # If there's unicode in email address, something is badly
+        # wrong and we just silently punt
+        # XXX - is there better approach?
+        print '# Error encoding', alias, repr(l)
+        return
+    return l
+

ietf/bin/generate-draft-aliases

@@ -0,0 +1,141 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# -*- Python -*-
+#
+# $Id: generate-draft-aliases $
+#
+# Author: Markus Stenberg <markus.stenberg@iki.fi>
+#
+"""
+
+This code dumps Django model InternetDraft's contents as postfix email
+aliases
+
+<no suffix> (same as -authors)
+.authors (list of authors)
+.chairs (WG chairs)
+.notify (notify emails(?))
+.ad (sponsoring AD)
+.all (all of the above)
+
+TODO:
+
+- results somewhat inconsistent with the results from the old tool;
+  should examine why (ask me for diff tool if interested in fixing it)
+
+"""
+
+DRAFT_EMAIL_SUFFIX='@tools.ietf.org'
+
+# boilerplate (from various other ietf/bin scripts)
+import os, sys
+basedir = os.path.abspath(os.path.join(os.path.dirname(__file__), "../.."))
+sys.path = [ basedir ] + sys.path
+
+from ietf import settings
+from django.core import management
+management.setup_environ(settings)
+
+from ietf.doc.models import Document
+from ietf.group.utils import get_group_chairs_emails
+from aliasutil import *
+import time
+
+def get_draft_ad_emails(draft):
+    " Get AD email for the given draft, if any. "
+    ad = draft.ad
+    #return [ad and ad.user and ad.user.email]
+    return [ad and ad.email_address()]
+
+def get_draft_authors_emails(draft):
+    " Get list of authors for the given draft."
+
+    # This feels 'correct'; however, it creates fairly large delta
+    return [email.email_address() for email in draft.authors.all()]
+
+    # This gives fairly small delta compared to current state,
+    # however, it seems to be wrong (doesn't check for emails being
+    # active etc).
+    #return [email.address for email in draft.authors.all()]
+
+def get_draft_notify_emails(draft):
+    " Get list of email addresses to notify for the given draft."
+    n = draft.notify
+    if not n:
+        return
+    l = []
+    draft_email = draft.name + DRAFT_EMAIL_SUFFIX
+    for e in n.split(','):
+        # If the draft name itself is listed as notify list element, we
+        # expand it (to make results better verifiable with the old ones)
+        e = e.strip()
+        if e == draft_email:
+            l.extend(get_draft_authors_emails(draft))
+        else:
+            l.append(e)
+    # Alternative: if we don't want to do expansion, just this would be
+    # perhaps better (MTA can do expansion too):
+    # l = n.split(',')
+    return l
+
+if __name__ == '__main__':
+    import datetime
+    import time
+
+    # Year ago?
+    show_since = datetime.datetime.now() - datetime.timedelta(365)
+    # 10 years ago?
+    #show_since = datetime.datetime.now() - datetime.timedelta(10 * 365)
+
+    modname = 'ietf.generate_draft_aliases'
+    date = time.strftime("%Y-%m-%d_%H:%M:%S")
+    print '# Generated by python -m %s at %s' % (modname, date)
+
+    drafts = Document.objects.all()
+
+    # Drafts with active status
+    active_drafts = drafts.filter(states__slug='active')
+
+    # Drafts that expired within year
+    inactive_recent_drafts = drafts.exclude(states__slug='active').filter(expires__gte=show_since)
+
+    interesting_drafts = active_drafts | inactive_recent_drafts
+
+    for draft in interesting_drafts.distinct().iterator():
+        # Omit RFCs, we care only about drafts
+        if draft.docalias_set.filter(name__startswith='rfc'):
+            continue
+        name = draft.name
+        done = []
+        all = []
+        def handle_sublist(name, f, o, is_ad=False):
+            r = dump_sublist(name, f, o, is_ad)
+            if r:
+                done.append(name)
+                all.extend(r)
+            return r
+        #.authors (/and no suffix) = authors
+
+        # First, do no suffix case
+        # If no authors, don't generate list either
+        r = dump_sublist(name, get_draft_authors_emails, draft)
+        if not r:
+            continue
+        handle_sublist('%s%s' % (name, '.authors'), get_draft_authors_emails, draft)
+        wg = draft.group
+        if wg:
+            # .chairs = WG chairs
+            handle_sublist('%s%s' % (name, '.chairs'), get_group_chairs_emails, wg)
+
+        # .ad = sponsoring AD
+        handle_sublist('%s%s' % (name, '.ad'), get_draft_ad_emails, draft, True)
+
+        # .notify = notify email list from the Document
+        handle_sublist('%s%s' % (name, '.notify'), get_draft_notify_emails, draft)
+
+        # .all = everything on 'done' (recursive aliases)
+        #dump_sublist('%s%s' % (name, '.all'), None, done)
+        # .all = everything on 'all' (expanded aliases)
+        dump_sublist('%s%s' % (name, '.all'), None, all)
+
+

ietf/bin/generate-wg-aliases

@@ -0,0 +1,91 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# -*- Python -*-
+#
+# $Id: generate-wg-aliases $
+#
+# Author: Markus Stenberg <markus.stenberg@iki.fi>
+#
+"""
+
+This code dumps Django model IETFWG's contents as two sets of postfix
+mail lists: -ads, and -chairs
+
+"""
+
+# boilerplate (from various other ietf/bin scripts)
+import os, sys
+basedir = os.path.abspath(os.path.join(os.path.dirname(__file__), "../.."))
+sys.path = [ basedir ] + sys.path
+
+from ietf import settings
+from django.core import management
+management.setup_environ(settings)
+
+from ietf.group.models import Group
+from ietf.group.utils import get_group_ads_emails, get_group_chairs_emails, get_area_ads_emails, get_area_chairs_emails
+from ietf.person.models import Email
+from aliasutil import *
+
+# from secr/utils/group.py..
+ACTIVE_STATES=['active', 'bof', 'proposed']
+
+if __name__ == '__main__':
+    import datetime
+    import time
+
+    # Year ago?
+    #show_since = datetime.datetime.now() - datetime.timedelta(365)
+
+    # 2 years ago?
+    #show_since = datetime.datetime.now() - datetime.timedelta(2 * 365)
+
+    # 3 years ago?
+    #show_since = datetime.datetime.now() - datetime.timedelta(3 * 365)
+
+    # 5 years ago?
+    show_since = datetime.datetime.now() - datetime.timedelta(5 * 365)
+
+    modname = 'ietf.generate_wg_aliases'
+    date = time.strftime("%Y-%m-%d_%H:%M:%S")
+    print '# Generated by python -m %s at %s' % (modname, date)
+    wgs = Group.objects.filter(type='wg').all()
+
+    print '# WGs'
+    # - status = Active
+    active_wgs = wgs.filter(state__in=ACTIVE_STATES)
+
+    # - activity within last year? (use concluded_date)
+    inactive_recent_wgs = wgs.exclude(state__in=ACTIVE_STATES).filter(time__gte=show_since)
+    interesting_wgs = active_wgs | inactive_recent_wgs
+
+    for wg in interesting_wgs.distinct().iterator():
+        name = wg.acronym
+        dump_sublist('%s%s' % (name, '-ads'), get_group_ads_emails, wg, True)
+        dump_sublist('%s%s' % (name, '-chairs'), get_group_chairs_emails, wg)
+
+    print '# RGs'
+    # - status = Active
+    rgs = Group.objects.filter(type='rg').all()
+    active_rgs = rgs.filter(state__in=ACTIVE_STATES)
+
+    # - activity within last year? (use concluded_date)
+    inactive_recent_rgs = rgs.exclude(state__in=ACTIVE_STATES).filter(time__gte=show_since)
+    interesting_rgs = active_rgs | inactive_recent_rgs
+
+    for rg in interesting_rgs.distinct().iterator():
+        name = rg.acronym
+        #dump_sublist('%s%s' % (name, '-ads'), get_group_ads_emails, rg, True)
+        dump_sublist('%s%s' % (name, '-chairs'), get_group_chairs_emails, rg)
+
+    # Additionally, for areaz, we should list -ads and -chairs
+    # (for every chair in active groups within the area).
+    print '# Areas'
+    areas = Group.objects.filter(type='area').all()
+    active_areas = areas.filter(state__in=ACTIVE_STATES)
+    for area in active_areas:
+        name = area.acronym
+        dump_sublist('%s%s' % (name, '-ads'), get_area_ads_emails, area, True)
+        dump_sublist('%s%s' % (name, '-chairs'), get_area_chairs_emails, area)
+
+