ConvertGoogleHistory.py

import json
import csv
import sys
from datetime import datetime
import os

def has_keys(dictionary, keys):
    return all(key in dictionary for key in keys)

def make_reader(in_json):

    # Open location history data
    json_data = json.loads(open(in_json).read())
    #Will read the following keys
    keys_to_check = ['timestamp', 'longitudeE7', 'latitudeE7', 'accuracy']

    # Get the easy fields
    for item in json_data['locations']:
        if has_keys(item, keys_to_check):
            timestamp = item['timestamp']
            if ('.' in timestamp):
                date = datetime.strptime(timestamp, '%Y-%m-%dT%H:%M:%S.%fZ').date()
            else:
                date = datetime.strptime(timestamp, '%Y-%m-%dT%H:%M:%SZ').date()
            tm = timestamp.split('T')[1].split('Z')[0]
            longitude = item['longitudeE7']/10000000.0
            latitude = item['latitudeE7']/10000000.0
            accuracy = item['accuracy']

            yield [date, tm, longitude, latitude, accuracy]


def getFullPath(inPath):
    if(not os.path.isabs(inPath)):
        # we need to set up the absolute path
        script_path = os.path.abspath(__file__)
        path, file = os.path.split(script_path)
        inPath = os.path.join(path, inPath)
    return inPath


# Read the Parameters
in_file = sys.argv[1]
out_file = sys.argv[2]

in_file = getFullPath(in_file)
out_file = getFullPath(out_file)

features = []
# add the Headers
features.append(['Date', 'Time', 'Longitude', 'Latitude', 'Accuracy'])
print("Reading {0}".format(in_file))

reader = make_reader(in_file)

for r in reader:
    features.append(r)

print('Read {0} Records'.format(len(features)-1))

# write this data
with open(out_file, 'w', newline='')as f:
    writer = csv.writer(f)
    writer.writerows(features)

Hello, I am new here and want to run this script on my location-history.json file, which I downloaded from Android phone, but it doesn't seem to work. Do you have basic instructions for running the python script? I have Windows 11 OS and Python 3.11.3. When I run in Python Idle, I get:
Traceback (most recent call last):
File "C:\Users\Rhythm Rider\AppData\Local\Programs\Python\Python311\ConvertGoogleHistory.py", line 43, in
in_file = sys.argv[1]
IndexError: list index out of range

Thanks

@GitMae99 Keep your Records.json file in the same folder as this script, and then you can run the following command: python ConvertGoogleHistory.py Records.json out.csv The script reads the first file (i.e. Records.json) and then generates the second file.

@GitMae99 Thanks for making this script. Sorry, I'm new to running python scripts I'm getting this error.

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.

Install the latest PowerShell for new features and improvements! https://aka.ms/PSWindows

PS D:\OneDrive - Rubix LLC\Software Data\Google History> python ConvertGoogleHistory.py Records.json out.csv
Reading D:\OneDrive - Rubix LLC\Software Data\Google History\Records.json
Traceback (most recent call last):
File "ConvertGoogleHistory.py", line 56, in
for r in reader:
File "ConvertGoogleHistory.py", line 18, in make_reader
for item in json_data['locations']:
TypeError: list indices must be integers, not str
PS D:\OneDrive - Rubix LLC\Software Data\Google History>

@devdattaT Thanks for the response.
When I go to google takeout and export from there, I only get three files, encryptedBackups.txt, settings.json and tombstones.csv. So no Records.json that I can find.
When I run the script with the export I did from my Android phone (which does give me a .json data file) I'm getting the error:
Reading C:\Temp\Records.json
Traceback (most recent call last):
File "C:\Temp\ConvertGoogleHistory.py", line 56, in
for r in reader:
File "C:\Temp\ConvertGoogleHistory.py", line 18, in make_reader
for item in json_data['locations']:
~~~~~~~~~^^^^^^^^^^^^^
KeyError: 'locations'

Maybe it doesn't work with the file I exported from the android through the settings??

@GitMae99 @devdattaT I'm running into the same problem, and it looks like the location-history.json file from Google is in a different format. There is no "locations" array in the json file. It has a "semanticSegments" array at the top. I assume this is because of Google's change to the timeline. I probably only have access to the Semantic Location History information, and not the Raw Location History data.

I made a minor adjustment using ChatGPT (not a coder at all, so please check) and it worked!

please replace path in/out with the relevant details.. the file that was released was kepler compatible for me! Maybe someone with more coding experience can adjust, improve it?

import json
import csv
from datetime import datetime
import os

def make_reader(in_json):
    # Open location history data
    with open(in_json, 'r') as file:
        json_data = json.load(file)
    
    for item in json_data:
        end_time = item.get('endTime', 'Unknown')
        start_time = item.get('startTime', 'Unknown')

        if '.' in end_time:
            end_date = datetime.strptime(end_time, '%Y-%m-%dT%H:%M:%S.%f%z').date()
        else:
            end_date = datetime.strptime(end_time, '%Y-%m-%dT%H:%M:%S%z').date()
        
        end_tm = end_time.split('T')[1].split('+')[0]

        if '.' in start_time:
            start_date = datetime.strptime(start_time, '%Y-%m-%dT%H:%M:%S.%f%z').date()
        else:
            start_date = datetime.strptime(start_time, '%Y-%m-%dT%H:%M:%S%z').date()
        
        start_tm = start_time.split('T')[1].split('+')[0]

        # Extract location details
        location = 'Unknown Location'
        lat, lon = None, None
        
        if 'visit' in item:
            placeLocation = item['visit'].get('topCandidate', {}).get('placeLocation', 'Unknown')
            if 'geo:' in placeLocation:
                location = placeLocation.split('geo:')[1]
                lat, lon = location.split(',')
        
        elif 'activity' in item:
            placeLocation = item['activity'].get('start', 'Unknown')
            if 'geo:' in placeLocation:
                location = placeLocation.split('geo:')[1]
                lat, lon = location.split(',')

        yield [end_date, end_tm, start_date, start_tm, lat, lon]

def getFullPath(inPath):
    if not os.path.isabs(inPath):
        # we need to set up the absolute path
        script_path = os.path.abspath(__file__)
        path, file = os.path.split(script_path)
        inPath = os.path.join(path, inPath)
    return inPath

# Hard-coded file paths
in_file = 'path/to/your/input.json'  # Replace with your input JSON file path
out_file = 'path/to/your/output.csv'  # Replace with your desired output CSV file path

in_file = getFullPath(in_file)
out_file = getFullPath(out_file)

features = []
# add the Headers
features.append(['End Date', 'End Time', 'Start Date', 'Start Time', 'Latitude', 'Longitude'])
print("Reading {0}".format(in_file))

reader = make_reader(in_file)

for r in reader:
    features.append(r)

print('Read {0} Records'.format(len(features) - 1))

# write this data
with open(out_file, 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerows(features)

@dinilj007 - I do a decent amount of python coding but for small things like this I use GPT also. Yours didn't work for my timeline export, so I fed GPT a sample of some of mine, followed by your script and the error message and it gave me a working script. Took less than 2 mins, which is why I didn't figure it out myself haha:

import json
import csv
from datetime import datetime
import os

def make_reader(in_json):
    # Open location history data
    with open(in_json, 'r') as file:
        json_data = json.load(file)

    # Access the "semanticSegments" list
    segments = json_data.get("semanticSegments", [])

    for segment in segments:
        start_time = segment.get('startTime', 'Unknown')
        end_time = segment.get('endTime', 'Unknown')

        # Process the start and end times
        if start_time != 'Unknown':
            if '.' in start_time:
                start_date = datetime.strptime(start_time, '%Y-%m-%dT%H:%M:%S.%f%z').date()
            else:
                start_date = datetime.strptime(start_time, '%Y-%m-%dT%H:%M:%S%z').date()

            start_tm = start_time.split('T')[1].split('-')[0]
        else:
            start_date = 'Unknown'
            start_tm = 'Unknown'

        if end_time != 'Unknown':
            if '.' in end_time:
                end_date = datetime.strptime(end_time, '%Y-%m-%dT%H:%M:%S.%f%z').date()
            else:
                end_date = datetime.strptime(end_time, '%Y-%m-%dT%H:%M:%S%z').date()

            end_tm = end_time.split('T')[1].split('-')[0]
        else:
            end_date = 'Unknown'
            end_tm = 'Unknown'

        # Extract location details from "visit" or "timelinePath"
        lat, lon = None, None

        if 'visit' in segment:
            placeLocation = segment['visit'].get('topCandidate', {}).get('placeLocation', {}).get('latLng', None)
            if placeLocation:
                lat, lon = [float(coord.strip('°')) for coord in placeLocation.split(", ")]

        elif 'timelinePath' in segment:
            # Extract the first point from timelinePath if it exists
            if len(segment['timelinePath']) > 0:
                point = segment['timelinePath'][0].get('point', None)
                if point:
                    lat, lon = [float(coord.strip('°')) for coord in point.split(", ")]

        # Yield the extracted data
        yield [end_date, end_tm, start_date, start_tm, lat, lon]

def getFullPath(inPath):
    if not os.path.isabs(inPath):
        # we need to set up the absolute path
        script_path = os.path.abspath(__file__)
        path, file = os.path.split(script_path)
        inPath = os.path.join(path, inPath)
    return inPath

# Hard-coded file paths
in_file = 'input.json'  # Replace with your input JSON file path
out_file = 'output.csv'  # Replace with your desired output CSV file path

in_file = getFullPath(in_file)
out_file = getFullPath(out_file)

features = []
# add the Headers
features.append(['End Date', 'End Time', 'Start Date', 'Start Time', 'Latitude', 'Longitude'])
print("Reading {0}".format(in_file))

reader = make_reader(in_file)

for r in reader:
    features.append(r)

print('Read {0} Records'.format(len(features) - 1))

# write this data
with open(out_file, 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerows(features)