search

ARTbio / tools-artbio

Collection of galaxy tools developed by the artbio-platform at the IBPS (Institut de Biologie Paris-Seine)
MIT License
12 stars 18 forks source link

element.identifier does not seem to work with the Galaxy branch release_16.01 #81

Closed drosofff closed 8 years ago

drosofff commented 8 years ago

and as a result, merge output of metavisitor workflows 3-1, 3-2 and 3-3 are not really usable... Any suggestion ?

drosofff commented 8 years ago

Yes I see that https://github.com/galaxyproject/galaxy/pull/1693 is still not merged and in the dev branch, right ?

mvdbeek commented 8 years ago

The PR is in 16.04 and dev, normally this should work on 16.04. Also I suppose element.identifier is a typo, right? It should be element_identifier. But are you sure this is why it's not working? From the issue that the PR addresses:

This works fine in most cases, except when running a workflow on a dataset collection that has been moved to a new/different history.

I think there was another problem with repeat sections and element_identifiers: https://github.com/galaxyproject/galaxy/pull/2029

So we could just cherry-pick those changes onto 16.01, or we can move forward to release_16.04, which I think is a more subtle upgrade compared to 15.10 -> 16.01, and we're running this on plastispi for around ~2 weeks without problems now.

drosofff commented 8 years ago

Yes, it's a typo. I am just rerunning workflows that used to work when I prepared metavisitor manuscript. I remember there was an issue with data manager and 16.04 (?). How is this fixed ?

Christophe Antoniewski

ARTbio http://artbio.fr/ - Tel +33 1 44 27 70 05 Drosophila Genetics & Epigenetics http://drosophile.org/ - Tel +33 1 44 2 _7 34 39_Mobile +33 6 68 60 51 50 https://twitter.com/ARTbio_IBPS https://twitter.com/drosofff

2016-05-08 17:28 GMT+02:00 Marius van den Beek notifications@github.com:

The PR is in 16.04 and dev, normally this should work on 16.04. Also I suppose element.identifier is a typo, right? It sould be element_identifier. But are you sure this is why it's not working? From the issue that the PR addresses:

This works fine in most cases, except when running a workflow on a dataset collection that has been moved to a new/different history.

I think there was another problem with repeat sections and element_identifiers: galaxyproject/galaxy#2029 https://github.com/galaxyproject/galaxy/pull/2029

So we could just cherry-pick those changes onto 16.01, or we can move forward to release_16.04, which I think is a more subtle upgrade compared to 15.10 -> 16.01, and we're running this on plastispi for around ~2 weeks without problems.

— You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub https://github.com/ARTbio/tools-artbio/issues/81#issuecomment-217727034

mvdbeek commented 8 years ago

Yes, it's a typo. I am just rerunning workflows that used to work when I prepared metavisitor manuscript. I remember there was an issue with data manager and 16.04 (?). How is this fixed ?

So the datamanager issue was solved by upgrading to the latest commit in 16.01. And that has overwritten the cherry-picked changes in the 2 PRs that concern element_identifiers. This is the problem with cherry-picking stuff :/. We should probably create our own release branches and regularly PR changes from upstream galaxy into our branch.

drosofff commented 8 years ago

OK, I see. But what i am trying to finalize here is a ansible and docker release of metavisitor that work as announced in the doc. So, in short, is a Galaxy version 16.04 compatible with both data manager ok and element_identifier working with copied histories ?

mvdbeek commented 8 years ago

This works now, right? Feel free to reopen if not.