Request: integrate iRODS fetcher into master?

[AliceSLee]

I've been happily using bdbag to materialise URLs into datasets and I'm now hoping to do the same with iRODS paths. This functionality seems to have been developed in the iRODS-integration branch some years ago. I'm planning to merge this code into my forked bdbag master branch and test it out. Is there any interest in this community in reviving the iRODS integration (to end up on main branch)? Or any insight as to what the problems were when it was being worked on? Cheers, Alice

[mikedarcy]

Hi Alice,

This prototype integration was done quite a long time ago (6 years!) in a sort of mini hack-a-thon at RENCI. I do recall at the time of writing that it did work. However, it has been so long now I am not sure if it still does. If you're willing to revive it and test it out, I'd be happy to consider a PR which includes support for it. My only request would be to keep it as a soft (run-time) dependency.

Thanks for your interest!

[AliceSLee]

Hi Mark, thanks for your reply! I will indeed try to do so, most likely end Nov. Cheers :)

[trel]

The iRODS protocol hasn't changed, so with looking at the code in master...irods-integration ...

I don't see anything that shouldn't still work.

Since this integration was written, the python library should now automatically use parallel transfer for larger files and be faster than 2018. Would be happy to hear about anything you find that doesn't 'just work'.