https://github.com/dustin/tesladb

Thing to grab data from the Tesla API and store it in sqlite and mqtt.
https://github.com/dustin/tesladb

Last synced: about 2 months ago
JSON representation

Thing to grab data from the Tesla API and store it in sqlite and mqtt.

• Host: GitHub
• URL: https://github.com/dustin/tesladb
• Owner: dustin
• License: bsd-3-clause
• Created: 2019-09-29T00:35:46.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2024-10-26T22:45:27.000Z (about 2 months ago)
• Last Synced: 2024-10-26T23:53:27.181Z (about 2 months ago)
• Language: Haskell
• Size: 197 KB
• Stars: 6
• Watchers: 8
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • Changelog: ChangeLog.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# tesladb

Tesladb isn't really a database.  Rather it is a way to grab data from

Tesla's user API and store it in both sqlite and MQTT.

In most common consumption, the other side of MQTT is sent to InfluxDB

or some other kind of time series database for later visualization in

Grafana or the like.

## Building

This is a [Haskell stack tools](https://docs.haskellstack.org/) based

project.  Given a working `stack`, one should be able to execute

`stack build` which will fetch all of the dependencies and build.

After building, you will have several executables.

teslauth: Authenticates to Tesla's API (with your credentials) and

stores them in a local database.  _For obvious reasons_ treat this

carefully.  While it doesn't store your actual Tesla account password,

it does have credentials allowing access to your Tesla account

(including things like reading your car state or honking you horn).

teslacatcher: Drains an MQTT topic to a sqlite database.  This can

also backfill missing data, effectively synchronizing state locally.

tesladb: Runs continuiously polling Tesla's API to write to a database

and MQTT, by default.

tesladbfix: Filters out fields that have bad timestamps to a table

named "rejects" in the database.

## Setup and Run

With Haskell `stack`, you can run the binaries from the source

checkout using `stack exec`.  Alternatively use `stack install` to

install to Stack's local bin directory.  Examples below are post install.

The first thing you'll need to do is auth.  This runs once initially,

then needs to be run once a day with the -r flag to refresh.

```

teslauth --email "[email protected]" \

--dbpath ~/var/tesla.db

```

Then, as mentioned, run approximately once per day (from cron or such

which checks for `0` exit status)  to refresh the auth token as follows:

```

teslauth -r --dbpath ~/var/tesla.db

```

After that, running tesladb itself is usually a matter of running the

main command with arguments.

```

tesladb --vname "mycar" \

--dbpath ~/var/tesladb.db \

--mqtt-uri mqtts://user:pass@host/#tesladb \

--mqtt-topic tesla/x/data \

--listen-topic 'tesla/x/in/#' \

--enable-commands -v

```

One problem you'll probably run into is that you don't know your

vehicle name to pass to `--vname`.  If you run the command without the

vehicle name, it will introspect the Tesla API to give you a list of

possible vehicles.

This will run every 600s or so as long as the auth is valid and the

API isn't returning errors.  The API _does_ tend to return errors

regularly. If that is the case, just back off and run it again using

whatever makes sense on your system to do such things (e.g. upstart,

smf).

## What Does This Do?

The resulting MQTT topics both receive telemetry and status as well as

take input messages for car control.  Need to open your frunk?  Don't

bother with the Tesla app, just craft an MQTT message to do it for

you.  Imagine the automation possibilities.