Python script to split Google contacts into individual VCF files.

vcard-split.py

#!/usr/bin/python3
# original version taken from https://gist.github.com/johnjohndoe/dada51bb886d419334de6fe13ce6d7bb
# Howto:                    
#* using python 3 call script, specify file name and add fist and last name if asked for.

#split vcf files

import re
import os #for working dir
import quopri # for utf-8 decodin

print("Welcome!")
print("This script converts a vcf file containing multiple contacts into multiple files.") 
fname = input('Please specify the name of the input vcf file: ')

print("Processing file " + fname) 

working_dir = os.getcwd()
split_prefix = 'vcard_'
output_name_default = 'contacts-part-'
output_name = output_name_default
charset_string = 'CHARSET=UTF-8;ENCODING=QUOTED-PRINTABLE:'
vcards_per_file = 1

with open(fname,'r') as f:
    count = 0
    output_count = 1
    results = []
    for line in f:
        # Read line by line until we reach BEGIN:VCARD again
        if ("BEGIN:VCARD" in line):
            count += 1
        if (count <= vcards_per_file):
            ## uncomment in case last and first name needs to be switched
            # if line.startswith('N:'):
            #     # LAST;FIRST -> FIRST;LAST Names
            #     line = re.sub(r'N:([^;]*);([^;]+)(;+)', r'N:\2;\1\3', line)
            #In case of a character set, ask user to correct it
            if line.startswith('N;' + charset_string):
                print("*** Charset for field N detected.")
                list = line[42:].split(';')
                for item in list:
                    decoded_string = quopri.decodestring(item)
                    print (decoded_string.decode('utf-8'))
                print("Please enter last and firstname without umlaut.")
                correct_last_name = input('lastname: ')
                correct_first_name = input('firstname: ')
                line = 'N:' + correct_last_name + ';' + correct_first_name + ';;;\n'
            if line.startswith('FN;' + charset_string):
                print("Charset for field FN detected.")
                list = line[43:].split(';')
                for item in list:
                    decoded_string = quopri.decodestring(item)
                    print (decoded_string.decode('utf-8'))
                correct_last_name = input('lastname: ')
                correct_first_name = input('firstname: ')
                line = 'FN:' + correct_first_name + ' ' + correct_last_name
            #In case a field name is contained, use it for the file name
            if line.startswith('FN:'):
                output_name = line[3:] # cuts off 'FN:'
                #remove escapes in names
                escapes = ''.join([chr(char) for char in range(1, 32)])
                translator = str.maketrans('', '', escapes)
                output_name = output_name.translate(translator)
            results.append(line)
        else:
            # We have read over a BEGIN:VCARD for more than vcards_per_file times and need to close the old file and start a new one
            # Write file with values stored up to now
            with open(os.path.abspath(os.path.join(working_dir, split_prefix + output_name + '_' + str(output_count) + '.vcf')),'w') as oFile:
                for item in results:
                    oFile.write(item)

            #increment outputfile count
            output_count += 1

            #clear results list and append line that we have read just now
            del results[:]
            results.append(line)

            #set counter back to 1
            count = 1

            # Set filename back to default
            output_name = output_name_default

    #write the last set of results to a file
    with open(os.path.abspath(os.path.join(working_dir, split_prefix + output_name + '_' + str(output_count) + '.vcf')),'w') as oFile:
        for item in results:
            oFile.write(item)