[INFRA-11] Implement mapper to use an LSST butler with PFS Created: 18/Jul/14 Updated: 02/Jul/16 Resolved: 02/Jul/16 |
|
| Status: | Done |
| Project: | Software Development Infrastructure |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Story | Priority: | Major |
| Reporter: | rhl | Assignee: | aritter |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Story Points: | 2 |
| Sprint: | 2014-8, 2014-9, 2014-12, 2014-13 |
| Reviewers: | rhl |
| Description |
|
We should adopt the LSST/HSC "butler" interface to data. This will require writing a mapper. |
| Comments |
| Comment by bick [ 02/Sep/14 ] |
|
I put in a basic outline of an obs_pfs package in obs_pfs.git. It's very simple and follows the obs_subaru package as an template. The mapper is watered down and right now only includes a 'raw' mapper target. There's an ingest.py script which just inherits from the main LSST one and removes the need for 'filter' info in paths and such. To test all the pieces, I used an HSC raw data file, but I didn't check it in to the git repo as it's an 18M binary. You can grab it from master:~bick/HSCA00123754.fits and then try: ./bin/pfsIngestImages.py --create /path/to/HSCA00123754.fits $PWD/data (yes, you do have have dump the newly created data repo in $PWD/data/). Then you should be able to run ./tests/getRaw.py, which just does a butler.get('raw') for visit 1236 ccd 1 (i.e. the HSC data file I used for testing) and checks the width and height of the loaded image. The code to generate defects wasn't HSC specific, so I moved it over wholesale. So far, there's only a minimal SConstruct + SConscript in place. There's nothing yet needing to be compiled. I put in a pfs/pfs_geom.paf, though the notion of a 'camera' is very different here. It currently uses most of the HSC specs for CCDs, and is limited to 8 CCDs. I haven't yet considered how the geom might look for the IR devices. I think these details can easily be filled in once they've been decided upon ... I think I actually have a message from Craig outlining a naming convention ... must check that. Before doing any more, I'd like to get RHL to take a quick look and make sure this is on track for what he'd like. I'm a little worried that a 'camera' is a misplaced idea here. If this is more or less the right idea, then we can start adding mapper targets and getting all the device specifics sorted out. |
| Comment by bick [ 02/Sep/14 ] |
|
I don't consider this a final review as there's much yet to add here, but the basic framework is in place. If this is roughly on tract, you can send it back to me to complete, with a second more final review to happen then. |
| Comment by swinbank [ 02/Jul/16 ] |
|
Agreed at the meeting of 2016-07-01 that this is done. We'll file tickets to make further updates as returned. |