Marty Oehme
fed14b4f4b
Added basic continuous integration tests to run on any push: On main branch, the python program is built. On tagged commit, a gitea release is created. Fixed first detected build pipeline issues: Fixing mypy library stubs missing for some imported libraries. Fixed two small typing errors for Repetitions. The steps run on basic python containers, onto which the ci steps simply install poetry. This takes a little more processing time during pipeline running (~16s per step), but also gives a lot of flexibility in container usage. Added script which assists in creating an automatic release by extracting the current version and newest changes from the semantic changelog. This is then used in the gitea release preparation as title and content of the release message. The files built in the dist directory by poetry will be attached. |
||
|---|---|---|
| src/habitmove | ||
| tests | ||
| tools | ||
| .gitignore | ||
| .woodpecker.yml | ||
| CHANGELOG.md | ||
| noxfile.py | ||
| poetry.lock | ||
| pyproject.toml | ||
| README.md | ||
habitmove
Takes habit in one habit-tracking application and transforms them ready to use for another.
Currently can take an export of nomie habits in json format and convert it to be importable in Loop Habit Tracker. Plans for reverse migration are on the roadmap, and ultimately this tool ideally understands more and more habit formats to prevent app lock-in.
Confirmed working for nomie version 5.6.4 and Loop Habit Tracker version 2.0.2 and 2.0.3. Presumably works for other nomie 5.x versions and other Loop 2.x versions as well, but that is untested.
Installation
Installation can be accomplished through pip:
pip install habitmove
Requirements:
habitmove requires at least Python 3.7.
It has only been tested on GNU/Linux (amd64) though it should work on other platforms.
Usage
Run as a cli utility habitmove currently takes a single argument: the nomie database .json file to import habits from.
Invoked like: habitmove nomie-export.json.
The output as a Loop Habit Tracker database will be written to output.db in the present working directory.
Can also take an existing Loop Habit database (exported from the application),
and add the nomie exported habits and checkmarks to it.
Simply put the exported Loop database in the same directory and call it output.db,
it will not (should not™️) overwrite anything.
If there are any duplicated habits however,
it will add duplications of the existing repetitions into the database.
Development
To enable easy development on the app, install poetry and let it do all dependency management for you by doing:
poetry install
poetry run habitmove <nomie-json>
To see a set up more closely resembling the final cli environment, with its libraries loaded as environmental dependencies enter the poetry shell:
poetry shell
The package can eventually also be used as a library to load nomie data to work with in Python,
or to move data into Loop Habit Tracker.
Take a look at the Parser and Transformer interfaces respectively.
To run tests for the app, simply invoke pytest through poetry run pytest or from within the poetry shell.
To run larger scale test automation, make sure you habe nox installed and run poetry run or again through the shell.
You can exclude integration tests that take longer and inspect the complete database output of the program through the parameters -m "not e2e" for both pytest and nox (which also does it automatically).