Location History Parser
Paste a Google Takeout or Dawarich location-history JSON export and get a per-day summary of places visited and distance traveled — text, CSV, or JSON, in km or miles. Everything runs locally; nothing is uploaded.
About this tool
This parser turns a raw location-history JSON export into a readable per-day summary: how many places you visited each day and how far you traveled. Paste one JSON file, pick your output format and unit, and read the result — no account, no upload, no map to pan around. Everything runs locally in your browser as WebAssembly.
It auto-detects the common export shapes:
- Google Takeout classic Records — a
{ "locations": [ … ] }file withlatitudeE7/longitudeE7integers andtimestamp/timestampMs. - Semantic Location History — the monthly
{ "timelineObjects": [ … ] }files withplaceVisitandactivitySegmentobjects (named places and the distance each activity already records). - On-device Timeline — the newer
{ "semanticSegments": [ … ] }export withvisit/activity/timelinePathandlatLngstrings. - Dawarich / generic point arrays — a plain
[ { "latitude", "longitude", "timestamp" }, … ], and GeoJSONPointFeatureCollections with a time property.
How places and distance are computed. When an export names your visited
places (Semantic and on-device Timeline), those named visits are counted
directly, and each day's distance comes from the distance / distanceMeters
each activity segment carries — the same figure the export reports. For raw GPS
streams (classic Records, Dawarich, GeoJSON) there are no named places, so the
tool uses stop detection: a run of consecutive points that stays within the
stop radius (meters) and spans at least the minimum stay (minutes) counts
as one visited place. Distance for raw streams is the great-circle (haversine)
sum of consecutive points on a 6,371 km-radius sphere.
Worked example
Paste this Semantic Location History snippet, leave the output on Text and unit on km:
{"timelineObjects":[
{"placeVisit":{"location":{"latitudeE7":400000000,"longitudeE7":-740000000,"name":"Home"},"duration":{"startTimestamp":"2024-01-01T08:00:00Z"}}},
{"activitySegment":{"distance":5000,"duration":{"startTimestamp":"2024-01-01T09:00:00Z"}}},
{"placeVisit":{"location":{"latitudeE7":401000000,"longitudeE7":-740000000,"name":"Office"},"duration":{"startTimestamp":"2024-01-01T10:00:00Z"}}}
]}
You get:
2024-01-01 — 2 places, 5.00 km
Total over 1 day: 2 places, 5.00 km
Switch Output to CSV for date,places,distance_km rows you can open in a
spreadsheet, or JSON for structured days that include each day's place names.
FAQ
Which exports does it accept?
Google Takeout classic Records (locations[] with latitudeE7 /
longitudeE7), Semantic Location History (timelineObjects[]), the newer
on-device Timeline (semanticSegments[]), a GeoJSON Point
FeatureCollection with a time property, and a plain array of points
({latitude, longitude, timestamp}, e.g. a Dawarich export). The format is
detected automatically from the JSON structure.
Is my location data uploaded anywhere?
No. The parser is compiled to WebAssembly and runs entirely in your browser. The JSON you paste never leaves your device — there is no server, upload, or network request involved.
The days are off by one — how do I fix that?
Timestamps in the exports are UTC. Set the UTC offset (hours) to your local
zone so each record buckets into the right calendar day — for example -5 for US
Eastern, 1 for Central Europe, or 5.5 for India. A visit recorded at 02:00
UTC moves to the previous day once you apply a negative offset.
What do stop radius and minimum stay do?
They only apply to raw GPS streams (classic Records, Dawarich, GeoJSON) that have no named places. A cluster of consecutive points staying within stop radius meters of an anchor, spanning at least minimum stay minutes, is counted as one visited place. Larger radius or longer stay means fewer, more significant places. Exports that already name your places (Semantic, on-device Timeline) ignore both settings.
How is distance calculated, and does it match Google's number?
For Semantic and on-device Timeline exports, distance is the
distance / distanceMeters each activity segment already carries, so it
matches the export's own figure. For raw point streams there is no recorded
distance, so the tool sums the great-circle (haversine) distance between
consecutive points — a straight-line-between-fixes estimate that runs a little
short of real road distance.
Can it read a Takeout ZIP, KML, or draw a map?
No. This is a single-JSON, text-in / text-out tool. Unzip your Takeout archive first and paste one JSON file; convert KML to JSON separately; and use a mapping app for an interactive map or heatmap. Unnamed stops are reported per day rather than reverse-geocoded to street addresses (that would need an online geocoding service).
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool location-history-parser '{"timelineObjects":[{"placeVisit":{"location":{"latitudeE7":400000000,"longitudeE7":-740000000,"name":"Home"},"duration":{"startTimestamp":"2024-01-01T08:00:00Z"}}},{"activitySegment":{"distance":5000,"duration":{"startTimestamp":"2024-01-01T09:00:00Z"}}}]}'New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/location-history-parser/?input=%7B%22timelineObjects%22%3A%5B%7B%22placeVisit%22%3A%7B%22location%22%3A%7B%22latitudeE7%22%3A400000000%2C%22longitudeE7%22%3A-740000000%2C%22name%22%3A%22Home%22%7D%2C%22duration%22%3A%7B%22startTimestamp%22%3A%222024-01-01T08%3A00%3A00Z%22%7D%7D%7D%2C%7B%22activitySegment%22%3A%7B%22distance%22%3A5000%2C%22duration%22%3A%7B%22startTimestamp%22%3A%222024-01-01T09%3A00%3A00Z%22%7D%7D%7D%5D%7D&output=text&unit=km&utc_offset=0&place_radius=150&min_stay=10Machine-readable descriptor: tool.json — title + parameters JSON Schema for agents.