search

nih-cfde / update-content-registry

Code and workflows for adding content to the content registry.
https://app-staging.nih-cfde.org/
BSD 3-Clause "New" or "Revised" License
0 stars 3 forks source link

[MRG] generate URLs for lists of valid IDs dynamically #71

Closed ctb closed 2 years ago

ctb commented 2 years ago

So https://github.com/nih-cfde/update-content-registry/pull/68 does a really nice job of addressing #70 but the new scripts/retrieve-ids.sh file implements some behavior that Snakemake can do more nicely (in terms of integrating into file date checking, etc.)

This PR demonstrates how to shift that behavior back into Snakemake, and as a bonus (?) removes the "valid ID" files from github.

raynamharris commented 2 years ago

I like this a lot! I was reading about lamba and parms but wasn't quite sure how to use them.

The only thing is that that the rule all isn't quite working to incorporate the retrieve and retrieve_term_wc rules. I had to run make retrieve to get the list to successfully run make. This is why the action failed.

ctb commented 2 years ago

ahh, I see.

The simplest and most correct (but also most tedious) way to deal with this is to add explicit references to data/validate/TERM.csv to all the rules that use them - e.g. add data/validate/gene.csv to the inputs for rule gene_json_alias_widget. I'll do that when I have a moment, or you can ;)

ctb commented 2 years ago

fixed! ready for merge I think.

raynamharris commented 2 years ago

works great now. thanks! the snakemake output is very nice as well.