Google Voice Takeout Parser

google_voice_texts.py

# Python script to parse google voice takeout messages into csv
# command line arguments

from bs4 import BeautifulSoup
import pandas as pd
import os
import argparse

# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--input", required=True,
                help="path to exported Google Voice directory (has 'Voice' folder inside it)")
ap.add_argument("-o", "--output", required=False,
                help="path to folder where CSV file will be saved. Google Voice directory is used if not supplied")
args = vars(ap.parse_args())
# print('input: {}'.format(args["input"]))

takeout_dir = args["input"]
exp_dir = args["output"] or takeout_dir
tko_calls_dir = os.path.join(takeout_dir, 'Voice', 'Calls')

message_df = pd.DataFrame(
    columns=['Date', 'Time', 'From', 'Content', 'Type', 'Date (UTC)'])

# Function to get all text log files


def get_html_files(calls_dir):
    file_list = []
    for file in os.listdir(calls_dir):
        if file.endswith(".html"):
            file_list.append(os.path.join(calls_dir, file))
        else:
            continue
    return file_list

# Function to process file



def parse_file(file_path):
    print(f"Processing file: {file_path}")
    basename, extension = os.path.splitext(os.path.basename(file_path))
    parts = basename.split(' - ')
    
    if parts[0] == 'Group Conversation':
        msg_type = 'Text'
        utc_date = parts[1]
        number = ''  # You may want to set a default value or handle this case differently
    elif len(parts) > 3:  # Number has a '-' in it
        number = ' - '.join(parts[:-2])  # Join all parts except the last two
        utc_date = parts[-1]
        msg_type = parts[-2]
    elif len(parts) == 2:  # No message type
        number = parts[0]
        utc_date = parts[1]
        msg_type = 'Unknown'  # or 'Text' if you want to assume it's a text message
    else:
        number, msg_type, utc_date = parts
    
    soup = BeautifulSoup(open(file_path), 'html.parser')
    
    # Rest of the function remains the same

    def parse_sms_file(soup):
        messages = soup.find_all('div', {'class': 'message'})
        message_ct = len(messages)
        for x in range(len(messages)):
            try:
                msg_1 = messages[x]
                iter_date = msg_1.find_all('abbr')[0].text.split(
                    ',')[0] + msg_1.find_all('abbr')[0].text.split(',')[1]
                iter_time = msg_1.find_all('abbr')[0].text.split(',')[
                    2].split('\n')[0][1:]
                # iter_number = msg_1.find_all('a')[0].text.replace('+', '')
                iter_message = msg_1.find_all('q')[0].text

                message_df.loc[len(message_df)] = [
                    iter_date,
                    iter_time,
                    number,
                    iter_message,
                    msg_type,
                    utc_date
                ]
            except Exception as e:
                print(e)
                continue

    def parse_voice_file(soup):
        messages = soup.find_all('div', {'class': 'haudio'})
        message_ct = len(messages)
        for x in range(len(messages)):
            try:
                msg_1 = messages[x]
                iter_date = msg_1.find_all('abbr')[0].text.split(
                    ',')[0] + msg_1.find_all('abbr')[0].text.split(',')[1]
                iter_time = msg_1.find_all('abbr')[0].text.split(',')[
                    2].split('\n')[0][1:]
                # iter_message = msg_1.find_all('span')[4].text
                if msg_type == 'Voicemail':
                    try:
                        iter_message = "Transcription: " + msg_1.find_all(
                            "span", class_="full-text")[0].text
                    except Exception as e:
                        iter_message = "NO TRANSCRIPTION. SEE AUDIO FILE"
                        print("Proabably no transcription:", file_path, e)
                else:
                    try:
                        iter_message = "Call Duration: " + msg_1.find_all('abbr')[1].text[1:][:8]
                    except:
                        iter_message = ''

                message_df.loc[len(message_df)] = [
                    iter_date,
                    iter_time,
                    number,
                    iter_message,
                    msg_type,
                    utc_date
                ]
            except Exception as e:
                print('File {} encountered: {}', file_path, e)
                continue

    if msg_type == 'Text':
        parse_sms_file(soup)
    else:
        parse_voice_file(soup)


# Function to save dataframe
def save_df(df):
    save_path = os.path.join(exp_dir, 'Message_Export.csv')
    df.to_csv(save_path, index=False)
    print('Message DF saved: {}'.format(save_path))

# Main function to process all files


def main():
    html_files = get_html_files(tko_calls_dir)
    for file in html_files:
        parse_file(file)
    save_df(message_df)


if __name__ == '__main__':
    main()