psphot -modeltest cannot do forced photometry of empty sky

[welling]

I'm looking into problems Michael has reported (see below) with doing forced photometry with psphot via the -modeltest option, and the issue seems to be a feature rather than a bug. If one tries forced photometry at a point which isn't a statistically significant peak, the tests for significant pixels at pmSourceMoments.c:136 and 137 exclude so many pixels that the test at line 167 returns false and causes photometry to fail.
I expected that the routine would calculate moments and variance

estimates for the moments, but that doesn't actually seem to be the
goal- pmSourceMoments() seems to be designed only to generate moments for a source which definitely exists. We're looking for a way to do forced photometry for locations before the SN or asteroid appears, so we need to look at empty points in space. Would it be reasonable to elaborate pmSource with some variance information and change pmSourceMoments to cease to exclude based on the tests at lines 136 and 137?

---

Hi Gene,

I'm having trouble with the force-position psphot photometry

"""
[serenity o5083g0235o.95580] psphot -file o5083g0235o.95580.wrp.
11968.skycell.064.fits -mask o5083g0235o.95580.wrp.11968.skycell.
064.mask.fits -variance o5083g0235o.95580.wrp.11968.skycell.
064.wt.fits -modeltest 817.7 213.0 -model PS_MODEL_PS1_V1 test
[...]
2009-10-13 21:33:18Z|serenity.phyast.pitt.edu|I|psphot

background sky : min -24.786289 mean 5.599239 max 14.937018 stdev

2.697644
2009-10-13 21:33:18Z|serenity.phyast.pitt.edu|5|psphot

build resampled image: 0.359058 sec

2009-10-13 21:33:19Z|serenity.phyast.pitt.edu|I|psphot

subtracted background model: 0.696939 sec

Aborting in function psphotModelTest at psphotModelTest.c:142. Error
stack:
2009-10-13 21:33:19Z|serenity.phyast.pitt.edu|A|psLib.sys

psSourceMoments error

Abort trap
"""

This worked back in May, 2009 for

"""
psphot -file o4741g0236o.36477.ch.11950.XY12.ch.fits -threads 4
-modeltest 2.63e+3 3.41e+3 test_out2
"""

but I don't have that image locally anymore. So I tried fitset, but I can't figure out what the third column is supposed to be and then it didn't seem to work for other reasons.

"""
[serenity o5083g0235o.95580] psphot -file o5083g0235o.95580.wrp.
11968.skycell.064.fits -mask o5083g0235o.95580.wrp.11968.skycell.
064.mask.fits -variance o5083g0235o.95580.wrp.11968.skycell.
064.wt.fits -fitset forcexy.txt test

psphotArguments (psphotArguments.c:242) : Expected to see one more argument; saw 3

Usage: one of the following
psphot -file fname1[,fname2,...] -mask maskfile1[,maskfile2,...]

-variance varfile1[,varfile2,...] OutFileBaseName

psphot -list FileNameList [-masklist MaskFileNameList]

-variancelist VarFileNameList OutFileBaseName

Try 'psphot -help' for more options and explanation
Error reading arguments

-> psphotArguments (psphotArguments.c:242): Incorrect arguments

Expected to see one more argument; saw 3

[serenity o5083g0235o.95580] cat forcexy.txt
817.7 213.0 1
"""

I'm happy to file this as a bug report, but I'm can't figure out what the desired usage is and so am not sure whether if it's a psphot bug or just a user bug.

Thanks,
Michael