A collection of python scripts to support analyzing the FP sensor data structures. Eventually, the analysis may allow for a cloud-less tool for sensor data retrieval.
For the BLE related scripts to work, the latest version of bluepy is required, preferably installed with pip:
sudo -h pip install bluepy
Also, a fairly recent version of bluez (version 5.41 or above) is required.
Scan for available FP devices and read out live data or history data.
In lack of official documentation, the structure of the history data was derived by comparing downloaded raw data from a number of sensors. With some fields showing identical vales in all the files, not all of the data structure could be assigned. Also, there is not yet any sense to be made of the calibration data stored in the sensors.
Downloaded history data is saved to files named hist-FPID-SID.dat for further analysis, with
FPID being the short ID of the sensor (last two bytes in address) and SID being the session id.
The data comprises raw sensor data from which physical values can be calculated. See Sensordata.md for details.
Add matching cloud data to sensor history data for correlation analysis. Matching is done by timestamps. Supports old and new cloud API via credential profiles. Note that for old and new API, separate credentials are required.
For cloud API access, credentials must be supplied in a file credentials.json in form of a profile
(e.g. named "api1" or "api2"). A profile must specify a "method" with value "oldapi" or "newapi"
to choose the right API methods and the base URL of the API server.
See example credentials.json.tmpl.
One of the profile names specified in credentials.json must be passed to the script. For details, check
fp-clouddata.py -h
For any file with downloaded sensor data passed to the script, a new file named comp-PROFILE-FPID-SID.dat
will be created containing all session data for which cloud data could be retrieved.
Listen for available BTLE devices and try to read out their characteristics. Output is provided in mediawiki table format.
Listen for advertisments and output advertised fields