By "hard wired" do you mean that the input pfsSimObject files all represent flat F_nu objects? If not, can't we complete the book-keeping without waiting for Sean?

By "hardwired", I mean that the input spectrum regardless of objId is a flat F_nu object. There's no file backing those. I believe the bookkeeping has been taken care of already, and we're just waiting to make the switch once the spectra are delivered.

How hard is it to make a file of the proper input format?  Are the arcs and sky also hard-wired (I seem to remember some horrors with the sky).

Is there a ticket for Sean to provide the "real" spectra?

Here's a simulated galaxy spectrum from Sean Johnson after it's been inserted into the 2D simulator and run all the way through the pipeline. You can see the Balmer break and [OII] at 757.5nm, so z=1.03.

from lsst.daf.persistence import Butler
butler = Butler("/home/pfs/pfs/sim2d-95/INTEGRATION/rerun/integration/pipeline")
spectrum = butler.get("pfsCoadd", catId=1, tract=0, patch="0,0", objId=0x12, numExp=4, expHash=0x1eab2e92)
fig, axes = spectrum.plot(plotSky=False, plotErrors=False, show=True)

Here's how I ran the simulator:

pfs@0f30f57bf254:~/pfs/SIMS/raw $ makePfiDesign --fibers lam --pfiDesignId 1 --scienceCatId 1 --scienceObjId "18 55 71 76 93 94 105 112 115"
pfs@0f30f57bf254:~/pfs/SIMS/raw $ makePfiDesign --fibers lam --pfiDesignId 2 --scienceCatId 1 --scienceObjId "18 55 71 76 93 94 105 112 115"
pfs@0f30f57bf254:~/pfs/SIMS/raw $ makeSimRun 1 2

Of course, more extensive simulations will require a lot more work to build the necessary PfsDesign files: this is just building two exposures, each with 7 science objects (plus 2 sky and 1 flux standard).

I believe this is ready for review.

Once it merges, I'll provide a build on tiger and instructions on how to run it.

hassan: I think you signed off on this on GitHub. Is it OK to merge?

Yes, sorry. I've changed the status to 'Review Complete'.

price: feel free to merge so that this ticket can be closed.

Merged to master.