search

poseidon-framework / poseidon-hs

A toolset to work with modular genotype databases in the Poseidon format
https://poseidon-framework.github.io/#/trident
MIT License
7 stars 2 forks source link

new option to output all janno columns in list #297

Closed stschiff closed 4 months ago

stschiff commented 5 months ago

This PR introduces an API-flag --fullJanno for retrieving extended individual info with all standard janno columns. A similar option exists for the Web API, see Changelog.

codecov[bot] commented 5 months ago

Codecov Report

Attention: Patch coverage is 76.92308% with 6 lines in your changes are missing coverage. Please review.

Project coverage is 68.20%. Comparing base (466870a) to head (fc9cbdc).

Files Patch % Lines
src/Poseidon/CLI/List.hs 66.66% 3 Missing and 3 partials :warning:
Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #297 +/- ## ========================================== + Coverage 68.15% 68.20% +0.05% ========================================== Files 26 26 Lines 3470 3482 +12 Branches 391 392 +1 ========================================== + Hits 2365 2375 +10 - Misses 714 715 +1 - Partials 391 392 +1 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

stschiff commented 5 months ago

Thanks for the review @nevrome, I have addressed the first comment by having introduced a new type. It's in ServerClient.hs, see my response above.

Regarding the Web-API, I don't quite understand why we would like to cater for the case of using a new client with an old server? That should never happen. Even in the unlikely case that users actually use trident serve locally, they would use a consistent server-client pair. And our server will always be up to date. So I don't see how the behaviour you discovered above would have any ramifications in production?

With respect to the additionalJannoColumns=ALL -> I get your point, but I think introducing an additional option is also not great, because it would lead to illegal semantics, like: &additionalJannoColumns=Country,Publication&fullJanno. Also, I am not even sure whether URLs even support boolean-like switch options like the one here. The usual way options are used is as key-value pair.

I think the rule that ALL retains a special meaning as a column name is a good option considering all alternatives. It is simple, easy to remember and avoids illegal option-combinations. What do you think?

That the CLI introduces a special option `--fullJanno´ is not inconsistent in my view. A CLI is structured very differently than a URL, as it allows more expressiveness and hence can make things more explicit.

nevrome commented 5 months ago

As stated in the meeting: You have convinced me :+1: