This is a small JavaScript tool to geotag your own photos, which in turn are derived from Google location history movement data. The tool can be called directly via the geotagger-js.html file. The wp-geotagger-js.php file can be included in a custom Wordpress theme and make the tool available there.
To read the EXIF data of the photos I use Exif-js.
More about this and example: entry in my blog (German)
Read in other languages: Deutsch
Download the directory assets and the file geotagger-js.html and place them exactly like this in a folder. If you now open the local file geotagger-js.html with the internet browser of your choice (e.g. file:///home/name/dir/geotagger-js.html), you should directly see this picture:
Load the directory assets and the file wp-geotagger-js.php into your theme directory. In the file functions.php geotagger-js is then included as follows:
require_once('geotagger-js/wp-geotagger-js.php');
From now on the tool can be included with the shortcode [geotagger_js master_class="geotagger-js"] on any page or post. The optional parameter master_class specifies the name of a CSS class in case the default class conflicts with the theme.
Visit Google Takeout to download your location history. You'll need the data in JSON format. Google doesn't offer anything else as of today (March 2022) anyway.
You can now load the JSON files from Google via drag & drop or via the file dialog. In the next step you can select one or more images.
If GPS data can be found in the Google dataset at least on the day the photo was taken, the tool will display the ones closest to the photo. With two buttons you can either copy the coordinates to the clipboard or use the appropriate 'exiftool' command with all parameters to store the GPS data in the EXIF record of the photo.
I decided to transfer the location history into the simple array 'timeline'. The elements of the array are the individual days to which GPS data can be assigned. A tag in the format timeline['YYYY-MM-DD'] is then again an array. Its elements are then an object from the actual data. So an element in the array could look like this:
timeline['2022-03-20'][3].theDate; // Instance of the Date object
timeline['2022-03-20'][3].theLat; // Latitude
timeline['2022-03-20'][3].theLon; // Longitude
timeline['2022-03-20'][3].theAction; // Source of the data point
The entry theAction is only an informative string and is currently not used any further. The array of objects is sorted by time.
To populate the array, the function addWaypoint(theDate, theLat, theLong, theAction) is called.
The format of the files is a bit sobering. After all, there is the definition of GeoJSON files, which form a standard. This format is not. But at least it is a standard JSON file.
Each element of the large array consists of several more or less exciting blocks. The interesting blocks are the following:
- activitySegment: Here an activity is described. Funny thing is that here you can see a ranking from Google per activity, how likely what kind of locomotion is.
- activitySegment.simplifiedRawPath: This is a subset of the activitySegment's, which itself contains waypoints with time information.
- placeVisit: This tells you from when to when you were at which place. Beside the for me exciting coordinates, there is again a ranking with probabilities, which concrete places it could have been.
In the function parseWalk(timeline) an object from the JSON array is passed in each case. Waypoints are extracted from this and then transferred into the custom data format.