telemach_parse.py

def telemachmob_parse(line):
    err = ''
    line = line.encode('utf-8')
    # Split the line by semicolons
    f = line.split(';')
    if f[8] != '1':
        return (None, err)
    rec = Record()
    rec.tel = f[0].strip()
    if not re.match("\d+", rec.tel):
        err = "tel=%s" % rec.tel

    # Extract relevant parts assuming the order as per the provided format
    rec.naziv1 = f[1].strip() if len(f) > 2 else ''  # company name
    rec.naziv2 = f[2].strip() if len(f) > 2 else ''  # company name
    mjesto_ptt_adresa = f[3].strip() if len(f) > 3 else ''  # combined address field
    rec.tip = f[4].strip() if len(f) > 4 else ''  # type
    #aktivan = f[8].strip() if len(f) > 8 else ''  # active status

    #fix za prazne zapise, samo tel je prisutan
    if rec.naziv1 == '' and rec.naziv2 == '':
        return (None, '') #ignore
    if rec.naziv1 == '':
        err = "%s nedostaje naziv1" % err

    rec.mjesto = ''
    rec.ptt = ''
    rec.adresa = ''
    rec.kbr = ''
    rec.kbp = ''

    # Split the combined address field by comma
    address_parts = mjesto_ptt_adresa.split(',')

    rec.mjesto = address_parts[0].strip() if len(address_parts) > 0 else ''
    rec.ptt = address_parts[1].strip() if len(address_parts) > 1 else ''
    adresa_kbrkpo = address_parts[2].strip() if len(address_parts) > 2 else ''

    # Process the adresa_kbrkpo to split into adresa, kbr, and kbp using the provided regex pattern
    m = re.match(r"(?P<adresa>.*) (?P<kbr>[0-9]+)( )?(?P<kbp>[a-zA-Z])?", adresa_kbrkpo)
    if m:
        rec.adresa = m.group('adresa').strip()
        rec.kbr = m.group('kbr').strip()
        rec.kbp = m.group('kbp').strip() if m.group('kbp') else ''
    else:
        rec.adresa = adresa_kbrkpo
        rec.kbr = ''
        rec.kbp = ''

    try:
        rec.tip = {'Poslovni': 'Poslovni', 'Privatni':'Privatni', 'Obrtnici':'Obrtnici', 'N/A':'N/A'}[f[4]]
    except:
        pass
        #err = "%s tip=%s" % (err, f[4])
    if rec.tip == 'N/A' or not rec.tip:
        #ako ne ime onda je poslovni
        if rec.naziv2 == '' or not rec.naziv2:
            rec.tip = 'Poslovni'
        else:
            #ako ima ime i prezime možda je firma ako je u prezimenu d.o.o. ili d.d. TODO
            #ako ima space u ime ili prezime onda err
            tmp = rec.naziv2.lower()
            tmp1 = rec.naziv1.lower()
            if tmp.find("d.d.") >= 0 or tmp.find("d.o.o.") >= 0 or tmp1.find("d.d.") >= 0 or tmp1.find("d.o.o.") >= 0:
                rec.tip = 'Poslovni'
            else:
                if tmp1.find(" ") < 0 and tmp.find(" ") < 0 and tmp.find("*") < 0:
                    #nema space u imeni i prezimenu
                    rec.tip = 'Privatni'
    if rec.tip == 'Privatni':
        if not rec.naziv2 or rec.naziv2 == '*':
            #privatni bez imena
            m = re.match(".* (d\.d\.|d\.o\.o|obrt).*", rec.naziv1, re.I | re.U)
            if m:
                rec.tip = 'Poslovni'
            else:
                rec.tip = 'N/A'
        elif rec.naziv2.lower() == 'd.o.o.':
            rec.tip = 'Poslovni'
    if rec.tip == 'N/A':
        if rec.naziv1 and rec.naziv2 and rec.naziv2 != '*':
            rec.tip='Privatni'
    if rec.tip not in tipovi:
        err = "%s tip=%s" % (err, rec.tip)
    if rec.tip == 'Privatni':
        if rec.naziv1:
            rec.naziv1 = rec.naziv1.decode('utf-8').capitalize().encode('utf-8')
            rec.naziv1 = re.sub("\(.*", "", rec.naziv1).strip()
        if rec.naziv2:
            rec.naziv2 = rec.naziv2.decode('utf-8').capitalize().encode('utf-8')
            rec.naziv2 = re.sub("\(.*", "", rec.naziv2).strip()
    else:
        if rec.naziv1:
            rec.naziv1 = rec.naziv1.replace("\\\"", "\"")
    rec.ppb = 0 #za sada nemamo flag za reverse search
    if rec.tip == 'Poslovni' or rec.tip == 'Obrtnici':
        rec.ppb = 1
    #pprint.pprint(rec.__dict__)
    if err == ' tip=N/A':
        logger.warn('ignore %s', line)
        return (None, '')
    return (rec, err)