'''
Created on Aug 28, 2009

@author: jpd
'''
import os
import sys
import subprocess
import HTMLParser
import logging
from optparse import OptionParser
import urllib


def setupLogging(name, level = logging.DEBUG):
    logger = logging.getLogger(name)
    logger.setLevel(logging.DEBUG)
    # create console handler and set level to debug
    ch = logging.StreamHandler()
    ch.setLevel(level)
    # create formatter
    formatter = logging.Formatter("%(asctime)s: %(name)s : %(levelname)s : %(message)s")
    # add formatter to ch
    ch.setFormatter(formatter)
    # add ch to logger
    logger.addHandler(ch)
    return logger

class Company(dict)  :
    
    def __init__(self):
        pass
    
    
class CROParser(HTMLParser.HTMLParser):
    
    def __init__(self):
        if hasattr(HTMLParser.HTMLParser, '__init__'):
            HTMLParser.HTMLParser.__init__(self)

        self._log = logging.getLogger("cro")
        self._companyList = []
        self._tableCount = 0
        self._dataCount = 0
        self._companyCount = 0
        self._company = Company()
        
    def parse(self, s):
        #"Parse the given string 's'."
        self.feed(s)
        self.close()
        
    def handle_starttag(self, tag, attrs):
        self._log.debug("start tag : %s" % tag)
        if (tag == "table") and (self._companyCount > 0) :
             #self._log.debug(  "%s %s" % (tag, attrs))
             self._tableCount = self._tableCount + 1
             self._log.debug("Table Count : %i" % self._tableCount)
                 

    def handle_data(self, data):

        self._log.debug("Handle_data : '%s'" % data)
        
        if data.startswith('Found: '):
            self._companyCount = int(data.split(": ")[1])
            self._log.debug("Found %i companies" % self._companyCount)
            
        if (data != 'Home') and (self._tableCount == 2) and  (self._companyCount > 0) :
             
            self._dataCount = self._dataCount + 1
            self._log.debug("Parsing : %s, dataCount : %i" % (data, self._dataCount))
            if self._dataCount == 2 :
                self._company[ 'Name' ] = data
            if self._dataCount == 3 :
                self._company[ 'Number' ] = data
            if self._dataCount == 4 :
                self._company[ 'Address' ] = data
            if self._dataCount == 5 :
                self._company[ 'Type' ] = data
                self._dataCount = 0
                self._companyCount = self._companyCount - 1
                self._companyList.append(self._company.copy())
                self._tableCount = 0
                self._log.debug("parsed : %s " % self._companyList[-1])
            
    def handle_endtag(self, tag):
        #print "end  tag : %s" % tag
        if tag == "table":
            pass

    def companies(self):
        return self._companyList
    
if __name__ == '__main__':

    log = setupLogging("cro")
    parser = OptionParser(usage = "%prog <list of files>",
                           version = "%prog release: 0.1")
    
    parser.add_option("-c", "--company", dest = "company", default = "PutPlace",
                       help = "return information on company named <company>")
    
    parser.add_option("-d", "--debug", dest = "debug", action = "store_true", default = False,
                       help = "turn on debugging")
    
    (options, args) = parser.parse_args()
    
    if options.debug :
        logging.getLogger("cro").setLevel(logging.DEBUG)
    else:
        logging.getLogger("cro").setLevel(logging.INFO)
        
    request = { 'radOption'     : 'COMPANY',
                'name'          : options.company,
                'radSearch'     : 'CONTAINSWORDS',
                'address'       : '',
                'number'        : '',
                'alpha'         : '',
                'submit2'       : 'Submit',
                'radPrevOption' : 'EXISTING' }
    
    params = urllib.urlencode(request)
    
    #print "params = %s" % params
    f = urllib.urlopen("http://www.cro.ie/search/companyindexe.asp", params)
    

    parser = CROParser()

    parser.parse(f.read())
    
    for i in parser.companies():
        print i