Uploaded image for project: 'Request For Comments'
  1. Request For Comments
  2. RFC-907

Make astroplan available in summit containers

    XMLWordPrintable

    Details

    • Type: RFC
    • Status: Proposed
    • Resolution: Unresolved
    • Component/s: DM
    • Labels:
      None

      Description

      This would be a very useful tool to have on the summit for quite a few reasons. It's in conda-forge and seems, at first glance at least, to not be too needy/controversial.

      https://anaconda.org/conda-forge/astroplan

        Attachments

          Issue Links

            Activity

            Hide
            tjenness Tim Jenness added a comment -

            Merlin Fisher-Levine are you suggesting there is a rubinenv-summit conda environment that should be used for all of Jenkins that tests summit-x packages and that would include astroplan? Or are you asking that rubinenv itself (which is nominally solely for lsst-distrib) should start including whatever summit-x packages need?

            Show
            tjenness Tim Jenness added a comment - Merlin Fisher-Levine are you suggesting there is a rubinenv-summit conda environment that should be used for all of Jenkins that tests summit-x packages and that would include astroplan? Or are you asking that rubinenv itself (which is nominally solely for lsst-distrib) should start including whatever summit-x packages need?
            Hide
            mfisherlevine Merlin Fisher-Levine added a comment -

            I would be fine with either of those, I don't think I mind at all - practically speaking they're going to be the same for developers/end-users, though I'm sure architecture will mind about which of those we do (hence asking for KT's input).

            My only real thought/point here is that I do think that code which is included in lsst_sitcom, which ships to the summit, and which has to work for observers/commissioning, should be tested as part of Jenkins, and in order to write those tests, those packages need to be available.

            Show
            mfisherlevine Merlin Fisher-Levine added a comment - I would be fine with either of those, I don't think I mind at all - practically speaking they're going to be the same for developers/end-users, though I'm sure architecture will mind about which of those we do (hence asking for KT's input). My only real thought/point here is that I do think that code which is included in lsst_sitcom , which ships to the summit, and which has to work for observers/commissioning, should be tested as part of Jenkins, and in order to write those tests, those packages need to be available.
            Hide
            ktl Kian-Tat Lim added a comment -

            I think you're better off not testing Summit packages in Jenkins at all. (I'd like to not test Science Pipelines in Jenkins, but that's a much bigger lift.) If you're testing in GHAs, you can build your own environment, including using or not using rubin-env and augmenting it. I think (but am not certain) that you can even install all of the Science Pipelines in that environment and still fit in a GHA worker. (What you can't do is build the Science Pipelines in a GHA worker.)

            Show
            ktl Kian-Tat Lim added a comment - I think you're better off not testing Summit packages in Jenkins at all. (I'd like to not test Science Pipelines in Jenkins, but that's a much bigger lift.) If you're testing in GHAs, you can build your own environment, including using or not using rubin-env and augmenting it. I think (but am not certain) that you can even install all of the Science Pipelines in that environment and still fit in a GHA worker. (What you can't do is build the Science Pipelines in a GHA worker.)
            Hide
            ktl Kian-Tat Lim added a comment -

            Alternatively, test in the TSSW Jenkins, which is much more modern, and which can also add any Summit-needed packages.

            Show
            ktl Kian-Tat Lim added a comment - Alternatively, test in the TSSW Jenkins, which is much more modern, and which can also add any Summit-needed packages.
            Hide
            mfisherlevine Merlin Fisher-Levine added a comment -

            Well, that would be a huge departure from the current model, especially as summit code can be broken by core DM changes, and therefore should be part of that CI system, I think. Hsin-Fang Chiang is working on the long-awaited data package for making this possible too... Maybe the three (or more) of us should have a chat about the best approach for that at some point though, it's probably quite beyond the scope of this RFC

            Happy to leave this pending for the time being.

            Show
            mfisherlevine Merlin Fisher-Levine added a comment - Well, that would be a huge departure from the current model, especially as summit code can be broken by core DM changes, and therefore should be part of that CI system, I think. Hsin-Fang Chiang is working on the long-awaited data package for making this possible too... Maybe the three (or more) of us should have a chat about the best approach for that at some point though, it's probably quite beyond the scope of this RFC Happy to leave this pending for the time being.

              People

              Assignee:
              mfisherlevine Merlin Fisher-Levine
              Reporter:
              mfisherlevine Merlin Fisher-Levine
              Watchers:
              John Parejko, Kian-Tat Lim, Merlin Fisher-Levine, Tim Jenness
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Dates

                Created:
                Updated:
                Planned End:

                  Jenkins

                  No builds found.