DM-14359: Fix data ID handling in ap_* #93
Conversation
| subDRList = dataRef.subItems() | ||
| if subDRList: | ||
| for subDR in subDRList: | ||
| if dataExists(subDR): |
There was a problem hiding this comment.
I am curious why this needs to be recursive. Wouldn't just datasetExists handlle all such situations? When does it not?
There was a problem hiding this comment.
I have no idea. This is just copied from pipe_base.
There was a problem hiding this comment.
It looks like @r-owen wrote this function originally.
| The type of data references to return. | ||
| level : `str` | ||
| The level of data ID at which to search. If the empty string, the | ||
| default level for ``datasetType`` shall be used. |
There was a problem hiding this comment.
Please reference wherever it is in the butler docs that level is better defined. If anywhere.
There was a problem hiding this comment.
I'm not aware of any definition of level in the docs.
There was a problem hiding this comment.
I would be surprised if you could find appropriate docs. I suggest we live with it. It was necessary in the butler when I wrote it, because of dataset types such as LSST raw images, which were per amplifier. I strongly suspect it's still necessary (though you were welcome to try removing it and see if we can still process LSST phosim data). It will go away with gen3 butler, thank heavens.
This PR factors some of the dataId handling logic out of
pipe_baseso that it can be used by code other thanArgumentParser. I've created a new module,butlerHelpers.py, for similar application logic that may be factored out in the future (in the unlikely event that it's needed before we rewrite everything for Butler 3).