Hey pipeline maintainers, Drafting a new release f...
# pipeline-maintainers
f
Hey pipeline maintainers, Drafting a new release for #molkart. We sort of messed up a bit the Changelog after first release and added things there throughout continuous development. Can we reconcile this by updating CHANGELOG in this release PR to remove all of the noise and include only changes for releases (in this case 1.0.0 and the upcoming 1.1.0)? If not, what's the suggested way to go forward?
j
If it's not released you can go to town
f
We have one release 1.0.0 and are preparing 1.1.0
So it is released and the reason I ask here ๐Ÿ˜‰
j
I mean if the up coming release with the messed up chsnglog hasn't been merged to master
Unless I misunderstand the problem
f
Nahh unfortunately there is already an ugly Changelog currently in master: https://github.com/nf-core/molkart/blob/master/CHANGELOG.md
I want to replace that with a cleaner one like atacseq. Only info for actual releases: https://github.com/nf-core/atacseq/blob/2.1.2/CHANGELOG.md
Well actually we want to base it more on rnaseq: https://github.com/nf-core/rnaseq/blob/3.14.0/CHANGELOG.md
Must have fallen through the cracks during initial release OR we aren't strict enough on Changelog formatting across pipelines ๐Ÿคท . Sorry about that
j
Ahhh i see. IMO that's completely fine, although I wouldnt use atacseq as an example
It doesn't seem to follow the standard that comes from the template
Check mag
Or sarek
๐Ÿš€ 1
I like mag most personally as gives most credit
f
I feel like RNAseq gives even more credit with callouts to new contributors?
j
Not to what the contributed to though
f
But I guess I see your point of sticking to
Added
,
Changed
and
Fixed
party this 1
j
And mag also thanks the person who reported issue too
But yeah it's those headers that are the important ones
๐Ÿ‘ 1
But I'm biased towards mag for obvious reasons ๐Ÿคฃ
f
I will post the finished PR here once we removed the old stuff and adujsted formatting to adhere to the headers from the template
๐Ÿ‘ 1
j
Also sarek and RNASeq have lots of experimental stuff in then nowadays so I don't like to use them as examples so much
f
You add the release dates on dev afterward I am guessing?
j
Push just before merge to dev yes
Too master*
f
Ahh or that of course
@James Fellows Yates how does this look? https://github.com/nf-core/molkart/pull/103/files
j
The file itself won't load on my phone from the pr Can you send a direct link to the file or a screenshot
f
CleanShot 2025-02-26 at 15.45.50.png
That would be the new
CHANGELOG
for nf-core/molkart which we would keep consistent in formatting for releases going forward
j
Much nicer! Where you say you've updated all modules, does that include version bumps?
Of the tools themselves
f
I think I simply meant
nf-core modules update --all
j
That could theoretically include tool updates though?
f
I mean of course, that's the nature of nf-core module updates right ๐Ÿ˜…
I think I know what you are getting at and I don't like it ๐Ÿ˜„ shakefist
j
In which case we normally have a table of dependencies in the changelog to indicate tooling has updated
As it can effect results outputs if algorithms and what not have changed ;)
100 rainbow 1
So it's important to inform users
Sooooo....
Shall I say it or you ;)
f
CleanShot 2025-02-26 at 15.56.05.png
party this 1
j
(given you the โœ… though, I have to go to bed now)
f
All good, I got the gist, thanks for your time keanu thanks
salute 1
k
Thanks for the feedback James!