[FIX] use bids uri for qmri derivatives #339
Conversation
There was a problem hiding this comment.
I think you're just missing the file:// prefix: https://bids-specification.readthedocs.io/en/latest/02-common-principles.html#source-vs-raw-vs-derived-data
Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org>
|
I could swear we had a conversation where we dropped the |
|
Oh, I see. That predates the BIDS URI discussion altogether. I would still use plain relative paths. Note in Examples we use relative paths (no |
|
mh then we have two ways of specifying relative paths in the spec, should we clarify that? |
yes please. 😝 |
Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org>
There was a problem hiding this comment.
Changes here look good to me @Remi-Gau except for that one comment above. Once that is resolved, I think we can merge and mirror the change to the upstream (real) data
| "sub-01/fmap/sub-01_flip-01_TB1DAM.nii.gz", | ||
| "sub-01/fmap/sub-01_flip-02_TB1DAM.nii.gz" | ||
| "bids:source:sub-01/sub-01/fmap/sub-01_flip-01_TB1DAM.nii.gz", | ||
| "bids:source:sub-01/sub-01/fmap/sub-01_flip-02_TB1DAM.nii.gz" |
There was a problem hiding this comment.
I think RawSources does not permit BIDS URIs, only dataset relative links. Also, that field is deprecated in favor of Sources (which does permit BIDS URIs)
https://bids-specification.readthedocs.io/en/latest/glossary.html#rawsources-metadata
I think we should either just use relative links here, OR turn the RawSources here into Sources
fixes #312
@sappelhoff does that make sense as a way to refer to source data in the derivatives using bids uri?