Documentation

[HadrienG]

Hi again!

I'm more or less done testing the software itself and overall I must say it works well! I'd like to suggest a few improvements to the documentation if you don't mind:

• [x] Although the command-line options are well explained, I think it'd be a good improvement to have a quickstart section with one or two real examples, i.e. "How to compare my samples together?" or "How to quickly compare my samples to RefSeq?"

• [x] You provide sketches for RefSeq (this is great!) but it is not so straight-forward how to query those: finch dist sketches/refseq_sketches_31_1000/*.sk fails with -bash: /Users/hadrien/.cargo/bin/finch: Argument list too long on my computer. Having a walkthrough on how to do that would improve usability. I used find but not sure there isn't a better way.

• [x] There are not guidelines for contributing. Adding a small statement about where to report issues, how to contribute to your software and eventually a Code of Conduct for people wising to contribute would be - in my opinion - great additions.

Best, Hadrien.

---

Review ref for tracking

[bovee]

Thanks! These are great suggestions.

I've gone ahead and added a small contribution section to the main README and selected a CODE_OF_CONDUCT (GitHub helpfully makes this very easy).

The first issue is fairly straight-forward, but for the second issue I think this may be a fundamental problem with how we've structured the databases. We should probably merge all of those into one file (because the bash problem is upstream of the code we have control over).

[HadrienG]

Thanks for your quick reply.

I do not think this is a fundamental problem and as you said what I stumbled upon is a bash problem. Of course it would be great to have a unified database or to allow finch to take a directory as input in mind for a future release.

For now I think potential users would be more than happy with a short tutorial in the README on how you use this databases.

I took the liberty of amending my issue with one more point. Nevermind. I couldn't get python to plot finch hist but the error was on my side.

Best, Hadrien

[bovee]

Hi Hadrien,

Sorry for the delay on closing this out (holidays!), but we added a "quick start" to our README and merged together the RefSeq sketches into single files (to fix the glob issue with Bash) in pull request #14, so I think this is complete now. Please let me know if there's anything I missed or if there's anything else you'd like with this.

Best, Roderick

[HadrienG]

Hi!

No worries about the delay. I do not think I have other issues to raise. I'll test again with some example data with 0.1.5 and if it's runs fine I feel comfortable recommending for publication.

Best, Hadrien