#!/usr/bin/env python
#####################################################################
# $Id: moz2ldap.py,v 1.3 2004/01/03 14:28:25 stefanor Exp $
#
# Stefano Rivera's Mozilla Address Book ldif to ldap ldif converter
# Version 1.0
# Copyright (C) 2003 Stefano Rivera <scripts "at" rivera.za.net>
#
# 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; version 2 of the
# License.
#
# 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., 59 Temple Place, Suite 330, Boston,
# MA 02111-1307 USA
#####################################################################

import base64, re, sys
fieldRE = re.compile("^([^:]+):(:)? (.*)$")
specialRE = re.compile("[^[:print:]]", re.M) # Matches text that needs base64 encoding
dnRE = re.compile("^cn=(.*?)(, ?mail=(.*))?$")
weirdDescRE = re.compile("^(.*?)<(.*?):(.*?)>(.*)\r?\n?$", re.M) # Sometimes they contain CRLF, so strip too...

baseDN = "ou=Addresses, dc=owlsbarn, dc=rivera, dc=za, dc=net"
myPeopleDn = "ou=People, " + baseDN
myGroupsDn = "ou=Groups, " + baseDN

remap = {
  "xmozillanickname": "mozillaNickname",
  "custom1": "mozillaCustom1",
  "custom2": "mozillaCustom2",
  "custom3": "mozillaCustom3",
  "custom4": "mozillaCustom4",
  "custom1": "mozillaCustom1",
  "homeurl": "mozillaHomeUrl",
  "mozilla_AimScreenName": "nsAIMid",
  "workurl": "mozillaWorkUrl",
  "xmozillausehtmlmail": "mozillaUseHtmlMail",
  "street": "streetAddress",
  "st": "stateOrProvinceName",
  "l": "localityName",
}

#####################################################################
def usage():
  print("""Stefano Rivera's Mozilla to LDAP converter:
  Usage: moz2ldap.py [OPTION] [FILE]... 
  Options:
    -h, --help   Show this message.

  If no files are given, it will read standard input.
""")

#####################################################################
def fieldSort(a, b):
  "Get those dns first, closesly followed by objectclasses"
  if a[0] == "dn":
    return -100
  elif b[0] == "dn":
    return 100
  elif a[0] == "objectclass":
    return -50
  elif b[0] == "objectclass":
    return 50
  else:
    return cmp(a[0], b[0])

#####################################################################
def processFile(inputFile):
  "Process the given file"
  record = {}
  nextLine = inputFile.readline()
  while True:
    # Lines can break, so we need to read one line ahead...
    line = nextLine
    nextLine = inputFile.readline()
    if line == "": # EOF
      if len(record) != 0:
        processRecord(record)
      break
    elif line == "\r\n" or line == "\n" or line == "\r": # End of record
      if len(record) != 0:
        processRecord(record)
        record = {}
    else:
      line = line.strip()
      while nextLine.startswith(" ") or nextLine.startswith("\t"): # Continued line
        nextLine.strip()
        line += nextLine
        nextLine = inputFile.readline()
      f = fieldRE.match(line)
      if f == None:
        sys.stderr.write("Bad field '%s', skipping...\n" % line)
        continue
      name = f.group(1)
      if f.group(2) != None: #i.e. base64 encoded
        value = base64.decodestring(f.group(3))
      else:
        value = f.group(3)
      if name not in record:
        record[name] = []
      record[name].append(value)

#####################################################################
def processRecord(record):
  "Check the record, fix it, and output it"
  if "dn" not in record:
    sys.stderr.write("Error, no dn!, skipping...\n")
    return
  if record['dn'][0].endswith(baseDN): # Already processed by me
    pass
  else:
    d = dnRE.match(record['dn'][0])
    if d != None:
      cn = d.group(1)
      if d.group(2) != None: # i.e. Person
        dn = myPeopleDn
      else:
        dn = myGroupsDn
      if cn == " " or cn == "":
        sys.stderr.write("Missing cn in dn '%s', skipping...\n" % record['dn'][0])
        return
      # Drop unneccesary quotation marks
      if "," not in cn and cn[0] == '"' and cn[-1] == '"':
        cn = cn[1:-1]
      # Handle commas gracefully
      cnField = cn
      if "," in cn:
        if cn[0] == '"' and cn[-1] == '"':
          cnField = cn[1:-1]
        else:
          cn = '"' + cn + '"'
      record['dn'][0] = "cn=%s, %s" % (cn, dn)
      record['cn'][0] = cnField
    else:
      sys.stderr.write("Bad dn: '%s', skipping...\n" % record['dn'][0])
      return
  
  if "modifytimestamp" in record:
    del record['modifytimestamp']
  
  if "mozillaAbPersonObsolete" in record['objectclass']:
    record['objectclass'].remove("mozillaAbPersonObsolete")
    record['objectclass'].append("mozillaOrgPerson")
    for old, new in remap.items():
      if old in record:
        tmp = record[old]
        del record[old]
        record[new] = tmp
    if "mozillaUseHtmlMail" in record:
      # OpenLDAP likes capital boolean values
      record['mozillaUseHtmlMail'][0] = record['mozillaUseHtmlMail'][0].upper()

  if "groupOfNames" in record['objectclass']:
    # Reformat member dns
    newMembers = []
    for member in record['member']:
      m = dnRE.match(member)
      if m != None:
        if m.group(2) != None: # i.e. Person
          dn = myPeopleDn
        else:
          dn = myGroupsDn
        newMembers.append("cn=%s, %s" % (m.group(1), dn))
      else:
        sys.stderr.write("Bad member: '%s', skipping...\n" % member)
    record['member'] = newMembers
    # Try to preserve nickname
    if "xmozillanickname" in record:
      if "mozillaOrgPerson" not in record['objectclass']:
        record['objectclass'].append("mozillaOrgPerson")
      tmp = record['xmozillanickname']
      del record['xmozillanickname']
      record['mozillaNickname'] = tmp

  # Mozilla can be naughty and forget to obey MUSTs. :-(
  # I have already taken care of cn in the dn section
  if "person" in record['objectclass']:
    if not "sn" in record:
      record['sn'] = record['cn'][0].split()[-1:]

  # These cropped up in an addr book of mine. Either they come from Eudora imports or old Mozillas...
  # Handle gracefully
  if "description" in record:
    while True:
      w = weirdDescRE.match(record["description"][0])
      if w == None:
        if record['description'][0].strip() == "":
          del record['description']
        break # No such thing as do ... while in Python ;-)
      if w.group(2) == "otheremail":
        record['mozillaSecondEmail'] = [w.group(3)]
      record['description'][0] = w.group(1) + w.group(4)
    
  # Output record
  sr = record.items()
  sr.sort(fieldSort)
  for key, values in sr:
    for value in values:
      if value.startswith(" ") or specialRE.search(value) != None:
        value = base64.encodestring(value)
        sys.stdout.write("%s::" % key)
        for line in value.splitlines(True):
          sys.stdout.write(" %s" % line)
      else:
        sys.stdout.write("%s: %s\n" %(key, value))
  sys.stdout.write("\n")

#####################################################################
def main():
  if len(sys.argv) > 1:
    if "-h" in sys.argv or "--help" in sys.argv:
      usage()
      sys.exit(0)
    for inputFileName in sys.argv[1:]:
      inputFile = open(inputFileName, "r")
      processFile(inputFile)
      inputFile.close()
  else:
    processFile(sys.stdin)

if __name__ == '__main__':
  main()