I found a couple of issues with the annota2seq mod...
# riboseq
j
I found a couple of issues with the annota2seq module while working up for deltate, would someone care to review for me? https://github.com/nf-core/modules/pull/9510
ohno2 1
šŸ‘€ 1
s
I can review. But we need to communicate responsibilities better, otherwise we run the risk of working on the same problems: • https://github.com/nf-core/modules/pull/9508 • https://github.com/nf-core/modules/pull/9511
j
sorry, I didn't even know anyone else was working on these pieces
Let's merge yours
s
Thanks!
Hang on!
Let's merge your 9510 and my 9511.
j
Yep, just spotted some missing pieces
Want me to do it?
(merge mine into yours I mean)
s
I have closed 9508, because your 9510 is more comprehensive. However, can you either merge 9511 into your 9510 (or into master directly)?
j
I'll merge 9511 and add you as a coauthor on 9510
s
Thanks!
Speaking of merging, I have also submitted a new module for plastid: https://github.com/nf-core/modules/pull/9482 Do you have time to review?
As well as this PR, which provides test data for the new module: https://github.com/nf-core/test-datasets/pull/1809
And FYI: I am working the issues listed here: https://github.com/nf-core/riboseq/issues/90. Next up (and the only missing item) is point 4
opt$exclude_samples_col
.
j
sure, I'll take a look
thankyou 1
Review submitted, in general there are a few nf-core niceties to conform to. I realise they're hard to get used to, give me a shout if you need a hand.
s
Thanks. The kind of feedback I needed!
Do you happen to know where to find a complete example of a module and pipeline that uses topics for the versions? Point 6 and 7 of the documentation appear redundant and I am unsure about where to put actual values and what the placeholders are. Of all the 762 modules, I found only one that makes use of topics and it is clearly wrong.
j
It's very new. I won't insist if you want to do the old-school thing, but fwiw: https://nf-co.re/blog/2025/version_topics I'm happy to handle the pipeline side
s
Thanks! Unfortunately, this wasn't really helpful, though. It looks wrong, too: On this page it reads:
- "{{ component }} --version":
type: *eval*
description: The expression to obtain the version of the tool
Which makes sense to me. On your linked page it reads:
- fastqc --version | sed "/FastQC v/!d; s/.*v//:
type: *string*
description: The command used to generate the version of the tool
Which doesn't make sense to me, because I would intuitively think that the command must be `eval`uated. Also, the
sed
command lacks a closing
"
, which I can hardly image is correct. And I am supposed to wrap all of this in quotes as in example 1 or not? I find both versions in different places. Without a working example, I can only guess what is correct, or fiddle around with the code for a couple of hours until I get it right.
j
Just do the old school way, we'll figure this out later
Or just copy what I do in ribodetector: https://github.com/nf-core/modules/blob/master/modules/nf-core/ribodetector/main.nf, up to you. The linter is insisting on the lack of closing quote in the meta.yml right now, suspect it's a bug, will chat to the tools folks
In fact I already did: https://github.com/nf-core/tools/issues/3941 (and there's a fix in progress)
@Sebastian Uhrig https://github.com/nf-core/modules/pull/9482 good to go now
šŸ™Œ 1