Periodically tweets warnings about locations with high CO₂ concentrations.
This script was written in Python 3.7 as that's what Microsoft Azure supports,
but can be run anywhere with a suitable Python installation,
pyenv is handy when using multiple
versions of Python on one machine.
Required modules are all available on PyPi and can be installed with:
pip install -r requirements.txt
The script can then simply be run with:
python SierraBot.py <url of summary> --daemon <tweet frequency>
E.g. @CLSierra2020 is run with
python SierraBot.py https://gp2020-sierra.azurewebsites.net/api/summary --daemon 15mins
For --daemon and other options taking time periods parsing is done with the
pytimeparse module, which
support a wide range of fairly relaxed formats.
Options to keep the bot running include running it in the background using the
--log-file option to send log output to a file or using
GNU Screen.
| Option | Usage | Default |
|---|---|---|
-h, --help |
show help | |
-warn <max. concentration>, --warning-threshold <max. concentration> |
CO₂ concentration (ppm) at which to post a warning | 1400 ppm |
-safe <safe concentration>, --safety-threshold <safe concentration> |
CO₂ concentration (ppm) at which to post a safety notice | 1000 ppm |
--averaging-period <time threshold> |
Time period for which to consider conditions | All time (as far as provided by the summary) |
--warning-period <warning period> |
How frequently to repeat warnings about rooms with detrimental conditions | <time threshold> if given, otherwise 90 minutes |
-keys <key file>, --key-file <key file> |
File containing Twitter API keys | keys.json |
-templates <template file>, --template-file <template file> |
File containing templates for tweets | templates.json |
-log [log file], --log-file [log file] |
File for log messages | SierraBot.py.log if option provided, otherwise log messages go to stdout/stderr |
--logging-level <logging level> |
Minimum level of log messages, one of: CRITICAL, ERROR, WARNING, INFO, DEBUG, NOTSET |
WARNING |
Running a Twitter bot requires access to Twitter's API, which you can request
here. You'll
need to create an app to get an API key and secret and an access token and secret.
When you first run SierraBot.py it will create a .json file in which to place
these. The script will log a DEBUG message with the account handle if you want
to check it has access.
(in rough order of implementation)
- Include timestamps in templates
- Only create template file if argument given but invalid/nonexistant
- More/richer tweet templates
- Provide more values to templates, e.g. thresholds, values of CO₂, temperature, etc.
- Make
--daemonswitch optional, preserving state over executions, so can be used with external schedulers e.g. cron - Cycle through locations over some period, rather than sending a barrage of tweets at once
- Some sort of combining metric to represent stuffiness/quality of work environment
- Occasional room recommendations
- Would want as a separate thread, complicates things when daemon-mode is optional