Activity Database
A set of Python3 scripts to import and export activity / work / health data into a common format — a SQLite database.
All scripts assume that the database resides in the current directory and is named activity.db.
Available conversions:
- Import Apple health database and append (only new items) to
activity.db. - Import GPX and append to
activity.db. - View workouts and export as GPX.
In case that append is not desired, delete activity.db and a new one will be created instead.
The database holds the "important" information found in the Apple health database, including heart rate, step count, flights climbed, even if not during a workout. In addition, workouts, with route (and heart rate data), pause/resume events and weather are stored.
Why?
-
Consolidate all activity sources. If only I could import workouts with routes to the iPhone health database...
-
Upload Apple watch workouts recorded with the internal workout app, including route, to fitness cloud providers (Strava, etc.).
-
A very simple format to be used as in input to data analysis tools.
Workouts Script
List all activities:
./workout.py --list
The result is a CSV containing all workouts. It can be directed to a file and opened by, e.g., Excel:
./workout.py --list > workouts.csv
open workouts.csv # in MacOS
Activities are referred by their starting date, as appears in the CSV. So, if the CSV looks like:
Time,Type,Source,Duration [mins],Distance [m]
2017-09-02T07:43:57+0300,Running,Strava,35,5769
2017-09-19T08:28:30+0300,Running,Runkeeper,40,6649
2017-09-19T21:24:07+0300,Cycling,Koby’s Apple Watch,16,3611
2017-09-21T08:28:48+0300,Running,iSmoothRun,34,5867
Then the cycling recorded on the Apple watch, can be exported to GPX as follows:
./workout.py --gpx "2017-09-19T21:24:07+0300"
iPhone Apple Health Converter
iOS11 allows full export of Apple Heath database, including routes in workouts, although location coordinates are trancated to around 10 m resolution.
From the Health app, press the upper right "face" icon, and choose "Export Health Data".
Send the resulting export.zip to a computer (AirDrop is the fastest if using a Mac).
In the current directory export.zip, and previous activity.db (if present) should reside.
Execute:
./import_health.py
The script is super-slow, so be patient.
TODO: make it faster!
Database
The database is a simple SQLite file.
Standard backup and restore can be done using the sqlite3 client:
# Dump
sqlite3 activity.db .dump > activity_backup.sql
# Restore (activity.db does not have to exist)
sqlite3 activity.db < activity_backup.sql
The backup is highly compressible.
See db.py.
LICENSE: MIT