search

yaml / yaml-test-suite

Comprehensive, language independent Test Suite for YAML
MIT License
172 stars 58 forks source link

Make test data available as JSON #46

Open flucknugget opened 4 years ago

flucknugget commented 4 years ago

I've been messing about in different languages and different YAML implementations. To make this easier I made a JSON file containing all named tests and their data.

This file will be outdated soon and only has the named tests, so how about adding something similar containing all tests?

Also finding the data branch was way too hard, please make the readme section explaining it bigger and or bolder ;)

perlpunk commented 4 years ago

I've been messing about in different languages and different YAML implementations. To make this easier I made a JSON file containing all named tests and their data.

If this is more handy than the data branch, we could generate a JSON file in a separate branch. Would be one JSON file per test also ok or do you need one single file?

What do others think?

This file will be outdated soon and only has the named tests, so how about adding something similar containing all tests?

What do you mean by "named tests"?

Also finding the data branch was way too hard, please make the readme section explaining it bigger and or bolder ;)

Ideally people should rather use the data release instead of the branch directly. However, I've been very busy with other things and haven't made a release for a while :-/

flucknugget commented 4 years ago

If this is more handy than the data branch, we could generate a JSON file in a separate branch. Would be one JSON file per test also ok or do you need one single file?

Why not both? But I would put it on the YAML Test Matrix website to make it easier to find. Like a full version of data-valid.js and data-invalid.js as JSON. Those files were almost what I was looking for, they just didn't have names and were JS instead of JSON.

What do you mean by "named tests"?

The name folder. Maybe that's all of them I don't know. The IDs have no meaning so I just went through that folder and ignored everything not linked from it.

Ideally people should rather use the data release instead of the branch directly. However, I've been very busy with other things and haven't made a release for a while :-/

I didn't even consider that, but that's on me. Maybe add a "Download" section to the readme explaining all the different ways to get the data. Because there's the files on the master branch, files on the data branch, releases, and now there's gonna be JSON files, too.

perlpunk commented 4 years ago

Sorry, I have not forgotten about this, there were just so many other urgent issues...

perlpunk commented 4 years ago

I think that it could be efficient to store one file per data point, containing all tests: in-yaml.js, in-json.js, test-event.js, tags.js, ....

We will have more test cases in the future, but also more types of data points.

When running tests, you usually run one type of test on all test cases.