Using python to parse email for action server

I found this python code on the forums as an answer to something that relates to my problem. I don't really understand python, so can somebody tell me why this isn't working?

(Some background information: I have a web form that get automatically emailed to openERP, which then automatically creates a lead. However, when a lead is created, info like phone and name do not get read from the email and sorted into their corresponding fields in the lead's form.)

# You can use the following variables:
#  - self: ORM model of the record on which the action is triggered
#  - object: browse_record of the record on which the action is triggered if there is one, otherwise None
#  - pool: ORM model pool (i.e. self.pool)
#  - time: Python time module
#  - cr: database cursor
#  - uid: current user id
#  - context: current context
# If you plan to return an action, assign: action = {...}

def parse_description(description):
  '''
   there is parse function
   It is example for parsing messages like this:

   Name: John
   Phone: +100500
  '''
  fields=['Name','Phone']
  _dict={}
  description=description.lower()
  for line in description.split('\n'):
    for field in fields:
        if field in line:
            split_line=line.split(':')
            if len(split_line)>1:
                pre_dict[field]=line.split(':')[1]
  return  dict

lead=self.browse(cr,uid,context['active_id'],context=context)
description=lead['description']
_dict=parse_description(description)
self.write(cr,uid,context['active_id'],{
                        'partner_name':_dict.get('name'),
                        'contact_name':_dict.get('name'),
                        'phone':_dict.get(u'phone'),
                        'mobile':_dict.get(u'phone')})

Update:

I got these traceback while I am fetching mail

2014-07-01 13:39:40,188 4992 INFO v8_demo openerp.addons.mail.mail_thread: Routing 
mail from Atul Jain <jain.atul43@gmail.com> to jain.atul10@hotmail.com with 
Message-Id <CAG=2G76_SRthL3ybGGyx2Lai5H=RMNxUOjRRR=+5-ODrcgtEZw@mail.gmail.com>:
fallback to model:crm.lead, thread_id:False, custom_values:None, uid:1
2014-07-01 13:39:40,445 4992 ERROR v8_demo openerp.addons.fetchmail.fetchmail: 
Failed to fetch mail from imap server Gmail.
Traceback (most recent call last):
File "/home/atul/openerp-8/openerp/addons/fetchmail/fetchmail.py", line 206, in 
fetch_mail
action_pool.run(cr, uid, [server.action_id.id], {'active_id': res_id, 'active_ids'
:[res_id], 'active_model': context.get("thread_model", server.object_id.model)})
File "/home/atul/openerp-8/openerp/addons/base/ir/ir_actions.py", line 967, in run
res = func(cr, uid, action, eval_context=eval_context, context=run_context)
File "/home/atul/openerp-8/openerp/addons/base/ir/ir_actions.py", line 805,  
in run_action_code_multi
eval(action.code.strip(), eval_context, mode="exec", nocopy=True)  # nocopy allows 
to return 'action'
File "/home/atul/openerp-8/openerp/tools/safe_eval.py", line 254, in safe_eval
return eval(c, globals_dict, locals_dict)
File "", line 14, in <module>
File "", line 4, in parse_description
ValueError: "'bool' object has no attribute 'lower'" while evaluating
u"def parse_description(description):
fields=['name','phone']
_dict={}
description=description.lower()
for line in description.split('\\n'):
for field in fields: 
if field in line:
split_line=line.split(':')
if len(split_line)>1:
_dict[field]=split_line[1]
return _dict
lead=self.browse(cr,uid,context['active_id'],context=context)\ndescription=lead['description']
_dict=parse_description(description)

 self.write(cr,uid,context['active_id'],{                'partner_name':_dict.get('name'),                             'contact_name':_dict.get('name'),        
'phone':_dict.get(u'phone'),
'mobile':_dict.get(u'phone')})"

Please help me in understanding the problem.

Solution

I've fixed the parse_description function:

def parse_description(description):
  '''
   there is parse function
   It is example for parsing messages like this:

   Name: John
   Phone: +100500
  '''
  fields=['name','phone']
  _dict={}
  description=description.lower()
  for line in description.split('\n'):
    for field in fields:
        if field in line:
            split_line=line.split(':')
            if len(split_line)>1:
                _dict[field]=split_line[1]
  return _dict

I changed the fields values to lower case because all operations on the description are on description.lower().
On the line pre_dict[field]=line.split(':')[1], you are splitting line to get your result. This has already been done: split_line=line.split(':') so you can just replace the pre_dict line with pre_dict[field]=split_line[1]
On that same line you are using a variable, pre_dict, which hasn't been referenced before. I think you mean to use _dict so the line should be _dict[field]=split_line[1]
The function returns dict which is a type, not a variable. You probably want it to return the dictionary which contains the field data, so it should return _dict instead; otherwise you'll always get the result <type 'dict'>

As for the remaining code, there's not enough context for me to understand what's happening or what's wrong. At least the parse_description function should be working now.