[INSTRM-1420] Publish final `pfsConfig` and `pfsDesign` FITS files, before taking sps/fps/mcs/agcc exposures. Created: 23/Oct/21  Updated: 02/Feb/23  Resolved: 02/Feb/23

Status: Done
Project: Instrument control development
Component/s: ics_iicActor
Affects Version/s: None
Fix Version/s: None

Type: Task Priority: Normal
Reporter: cloomis Assignee: cloomis
Resolution: Done Votes: 0
Labels: CanBeClosed
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Blocks
blocks INSTRM-1422 Publish complete pfsDesign/pfsConfig ... Open
Relates
relates to INSTRM-1427 Publish pfsConfig files when the PFI ... Done

 Description   

Before running any sps/fps/agcc/mcs command which might take an exposure, make sure the current config has been constructed and published. 

Iterate over `sps.lightSources`, and gather the appropriate `fps.pfsConfig`, `sunss.pointing` and `dcb.bundles` values. When any of those values change, trigger the generation of a new, possibly aggregated, `pfsDesign/pfsConfig`.
 
Put published FITS files in `/data/raw/$date/{pfsDesign,pfsConfig}/` – these will be copied with the raw data. We want one `pfsConfig` file for each `visit0`, not one per visit.
 
To highlight the most important configuration: when the `pfi` is the only light source, simply copy/link the `pfDesign` file from the targeting/planning software and the `pfsConfig` from the `fpsActor` output.
 

 Comments   
Comment by naoyuki.tamura [ 02/Nov/21 ]

I don't have clear ideas about what "final" means, and what "publish" means ...
Depending on the intents of tests/observations represented by pfsDesign (and Config) files, the judgement of them being "final" and "publish-iable" may change I presume, so I suspect mechanism(s) with some human intervention are supposed to be discussed here? I am sorry very likely I am just ignorant and behind, but I appreciate a bit more elaboration about intents.

Comment by cloomis [ 02/Nov/21 ]

This would be the place for the pfsDesign and pfsConfig files which actually get used and observed: the ones which are associated with some PFSx file. This is a subset of the pfsDesign files which are available to observe.

To be more concrete, when iic observePfsDesign designId=xxx and thus fps moveToPfsDesign designId=xxx get called, one design file from the location/directory which is populated by shuffle (see INSTRM-1416) is read and instantiated by the fpsActor. It is also "published" by iic by being copied to, as proposed, /data/raw/$date/pfsDesign/.

Comment by arnaud.lefur [ 02/Feb/23 ]

We now have pfsConfig created before each sps exposure which I think mostly what this ticket was aiming for.
The pfsConfig merging is implemented for dcb + SuNSS, but not for pfi+dcb|SuNSS .
Let's close this one at least, and open new tickets if necessary.

Generated at Sat Feb 10 16:35:17 JST 2024 using Jira 8.3.4#803005-sha1:1f96e09b3c60279a408a2ae47be3c745f571388b.