def preproc(self,qombos):
if type(qombos) is StringType:
qombo=qombos
qombos=[qombo]
allfiles=self.science_images(qombos)
for image in allfiles:
if self.verbose:
print "Preprocessing image %s" % image
# Add BPM as a header keyword
update_head(image,'BPM',self.bpmroot+'.pl',
"Bad Pixel Mask")
# Calculate sky background
skybkg=wirc_sky(image,'!BPM')
update_head(image,'SKYBKG',skybkg,
'Value of sky background')
# Rudimentary WCS
iraf.add_wcs(image,instrument="wirc")
# Done with Step #2
update_head(image,'IQWRCSTP',2,
"Stage of IQWIRC processing")
def dummy(dum1,dum2):
# Setup ccdproc options
ccdproc=iraf.mscred.ccdproc
ccdproc.ccdtype=""
ccdproc.noproc=no
ccdproc.fixpix=no
ccdproc.oversca=no
ccdproc.trim=yes
ccdproc.zerocor=yes
ccdproc.flatcor=no
ccdproc.illumco=no
ccdproc.fringec=no
ccdproc.readaxi="line"
ccdproc.trimsec=trimsec
ccdproc.zero=biasname
# Basic sanity checks
check_exist(biasname,"r")
# Big Loop
while(1):
# Parse inputs
allfiles=glob.glob(inpat)
newfiles=[]
for image in allfiles:
if not reduced.has_key(image):
newfiles.append(image)
for image in newfiles:
if verbose:
print "Reducing new image %s" % image
# Bias subtraction
image1=biaspfx+image
s1=ccdproc(image,output=image1,Stdout=1)
# Flatfielding
filt=get_head(image1,filtkey)
flatname=flatpre+filt+".fits"
check_exist(flatname,"r")
iraf.iqflatten(image1,flatname,outpfx=flatpfx,
normflat=yes,statsec=statsec,vignflat=no,
clobber=yes,verbose=no)
image2=flatpfx+image1
# Rudimentary WCS
iraf.add_wcs(image2,instrument="p60ccd")
# Object-detection
iraf.iqobjs(image2,sigma,satval,masksfx=masksfx,
wtimage="none",clobber=yes,verbose=no)
# Refine WCS
############
# Clean up
check_exist(image1,"w")
# Done with processing
reduced[image]=yes
# Wait a little while
time.sleep(twait)
def iqp60(inpat,endhow="never",endwhen="",focmode="all",
logfile="",clobber=globclob,verbose=globver):
""" intelligent processing of P60 data files """
# Defaults
twait=30
reduced={}
filtkey="FILTER"
biaskey="IMGTYPE"
biasre="BIAS"
biasroot="Bias"
biaspfx="b"
flatkey="IMGTYPE"
flatre="DOMEFLAT"
flatpre="Flat-"
flatpfx="f"
bpm0root="BPM0"
bpmroot="BPM"
binkey="CCDSUM"
detseckey="DETSEC"
binroikey="BINROI"
binroidef="A"
nfocus=10
focuskey="IMGTYPE"
focusre="^FOCUS"
fseqkey="FOCUSSEQ"
skipkey="IMGTYPE"
skipre="SAOFOCUS"
statsec=""
reqkeys=['IMGTYPE','OBJECT','FILTER','BINROI',
'CCDSUM','DETSEC','ROISEC']
sigma=3.0
minraw=0.0
maxraw=65535.0
satval=50000.0
masksfx="mask"
rawpix=0.3787
zptkey="ZEROPT"
zpukey="ZEROPTU"
# For now:
#catmags={'B':'BMAG','V':'BMAG','g':'BMAG',
# 'R':'RMAG','r':'RMAG','I':'IMAG',
# 'i':'IMAG','ip':'IMAG','z':'IMAG',
# 'zp':'IMAG','zl':'IMAG','zs':'IMAG'}
## In the future, once Sloan mags are written to the ubcone output:
catmags={'B':'BMAG','V':'VMAG','g':'GPMAG',
'R':'RMAG','r':'RPMAG','I':'IMAG',
'i':'IMAG','ip':'IPMAG','z':'ZPMAG',
'zp':'ZPMAG','zl':'ZPMAG','zs':'ZPMAG',
'upr':'UPMAG','gpr':'GPMAG','rpr':'RPMAG',
'ipr':'IPMAG','zpr':'ZPMAG'}
exptmkey="EXPTIME"
exptmstd=60.0
minzpts={}
minskys={}
fwhm0=1.5
gain=2.3
readn=7.5
crsfx="crm"
statfile="Status_Science.txt"
# Parse end-condition "time"
if endhow=="time":
re1=re.search("^(\d+):(\d+)",endwhen)
if re1:
tnow=time.gmtime()
# Force UT calculation w/ ignoring of DST issues
reftime=time.mktime([tnow[0],tnow[1],tnow[2],
int(re1.group(1)),int(re1.group(2)),0,
tnow[6],tnow[7],0]) - time.timezone
if reftime<time.time():
reftime+=86400
if verbose:
print "Running until %s" % \
time.strftime("%d %b %H:%M",time.gmtime(reftime))
else:
print "Failed to parse %s as UT time" % endwhen
print "Running until stopped..."
endhow="never"
# Parse focmode setting
fonly=0
if focmode=="nofocus":
skipre="FOCUS"
elif focmode=="focus":
fonly=1
skipre="SCIENCE"
# Parse logfile setting
if len(logfile)>0:
check_exist(logfile,'w',yes)
try:
log=open(logfile,'w')
sys.stdout=log
except:
print "Failed to open logfile %s for writing" % logfile
# Setup ccdproc options
ccdproc=iraf.ccdred.ccdproc
ccdproc.overscan=yes
ccdproc.trim=yes
ccdproc.zerocor=no
ccdproc.interactive=no
ccdproc.function="chebyshev"
ccdproc.order=3
ccdproc.sample="*"
ccdproc.naverage=3
ccdproc.niterate=2
ccdproc.low_reject=3.0
ccdproc.high_reject=3.0
ccdproc.grow=0.0
# Demosaic all the files
allfiles=glob.glob(inpat)
alllist=','.join(allfiles)
iraf.iqmosaic(alllist,outpfx="",biassec="!BIASSEC",
trimsec="!TRIMSEC",joinkeys="CCDSIZE,DETSEC,ROISEC",
splitkeys="AMPSEC,CCDSEC,TRIM,OVERSCAN,CCDPROC",
clobber=yes,verbose=verbose)
iraf.add_wcs(alllist,instrument="p60new",binkey=binkey)
# Set ccdproc params for bias-subtraction only
ccdproc.overscan=no
ccdproc.trim=no
ccdproc.zerocor=yes
# Determine if we need to make calibration files
dobias=no
doflats=no
for image in allfiles:
# Criteria for bias image
biasval,flatval=get_head(image,[biaskey,flatkey])
if re.search(biasre,biasval,re.I):
dobias=yes
reduced[image]=yes
elif re.search(flatre,flatval,re.I):
doflats=yes
reduced[image]=yes
# Calibrations
if (dobias or doflats) and not fonly:
iraf.iqcals(alllist,dobias=dobias,biasproc=no,
biaskey=biaskey,biasre=biasre,biasroot=biasroot,
doflats=doflats,flatproc=yes,flatkey=flatkey,
flatre=flatre,filtkey=filtkey,flatpre=flatpre,
flatscale="mode",statsec="",normflat=yes,
dobpm=no,bpmroot=bpmroot,classkey=binroikey,
classdef=binroidef,mosaic=no,
clobber=clobber,verbose=verbose)
# Delete binroi-specific BPMs and flatfields
oldbpms=glob.glob(bpmroot+'-'+'*.pl')
for oldbpm in oldbpms:
iraf.imdel(oldbpm,verify=no,go_ahead=yes)
oldflats=glob.glob(flatpre+'*-*.fits')
for oldflat in oldflats:
iraf.imdel(oldflat,verify=no,go_ahead=yes)
# Clip all flatfields and update BPM as we go
bpm0name=bpm0root+'.pl'
bpmname=bpmroot+'.pl'
if os.path.exists(bpmname):
check_exist(bpm0name,'w',clobber=yes)
iraf.imrename(bpmname,bpm0name,verbose=no)
allflats=glob.glob(flatpre+'*.fits')
for flatimg in allflats:
iraf.iqclip(flatimg,lthresh=0.75,hthresh=1.25,bookend=no,
replace=1.0,maskin=bpm0name,maskout=bpmname,
maskval=1,clobber=yes,verbose=no)
iraf.imdel(bpm0name,verify=no,go_ahead=yes)
iraf.imrename(bpmname,bpm0name,verbose=no)
update_head(flatimg,"CLIPFLAT",1,"Has flatfield been clipped?")
update_head(flatimg,"CLIPREPL",1.0,
"Replacement value for clipped pixels")
iraf.imrename(bpm0name,bpmname,verbose=no)
##################################################
# Lists of focusfiles
allfocus=[]
focusfiles=[]
# Should additional calibrations be taken later on...
biasfiles=[]
flatfiles=[]
# Big Loop
done=no
while not done:
# Parse inputs
allfiles=glob.glob(inpat)
newfiles=[]
for image in allfiles:
if not reduced.has_key(image):
# Exclude Bias & Flats
if re.search(biasroot,image,re.I) or \
re.search(flatpre,image,re.I):
reduced[image]=yes
continue
# Exclude calibration files (for now)
biasval,flatval=get_head(image,[biaskey,flatkey])
if re.search(biasre,biasval,re.I):
biasfiles.append(image)
reduced[image]=yes
elif re.search(flatre,flatval,re.I):
flatfiles.append(image)
reduced[image]=yes
# Exclude "skippable" files
skipval=get_head(image,skipkey)
if re.search(skipre,skipval,re.I):
reduced[image]=yes
# Queue file for processing
else:
reduced[image]=no
newfiles.append(image)
elif not reduced[image]:
# Attempted reduction before, but failed?
newfiles.append(image)
if newfiles:
# Brief pause to avoid file conflicts (?)
time.sleep(3.0)
# Demosaic
newlist=','.join(newfiles)
ccdproc.overscan=yes
ccdproc.trim=yes
ccdproc.zerocor=no
iraf.iqmosaic(newlist,outpfx="",biassec="!BIASSEC",
trimsec="!TRIMSEC",joinkeys="DETSEC",
splitkeys="AMPSEC,CCDSEC,TRIM,OVERSCAN",
clobber=yes,verbose=verbose)
iraf.add_wcs(newlist,instrument="p60new",binkey=binkey)
# Reset ccdproc params for bias-subtraction only
ccdproc.overscan=no
ccdproc.biassec=""
ccdproc.trim=no
ccdproc.trimsec=""
ccdproc.zerocor=yes
ccdproc.flatcor=no
##############################
# Process new images
for image in newfiles:
# Required keyword check
if not check_head(image,reqkeys):
print "Image %s is missing required keywords" % image
continue
# Attempt processing only once per image
reduced[image]=yes
if verbose:
print "Reducing new image %s" % image
# Track useful status information
scistat={}
# Prepare for processing
pix=rawpix
binroi=binroidef
biasname=biasroot+'.fits'
bpmname=bpmroot+'.pl'
filt=get_head(image,filtkey)
flatname=flatpre+filt+".fits"
# Calibration files for BINROI setting
if check_head(image,binroikey):
binroi=get_head(image,binroikey)
if binroi != binroidef:
# Get the right calibration files
fullbias=biasname
biasname=biasroot+'-'+binroi+'.fits'
fullbpm =bpmroot+'.pl'
bpmname =bpmroot+'-'+binroi+'.pl'
fullflat=flatname
flatname=flatpre+filt+'-'+binroi+".fits"
# Details of this binning/roi combo
[binval,detsec]=get_head(image,[binkey,detseckey])
binning=p60parsebin(binval)
roi=p60parseroi(detsec)
# Pixel scale (no anisotropic binning, sorry)
pix=binning[0]*rawpix
# Check for existance of bias; make it if we have to
if not os.path.exists(biasname):
print ("Failed to find bias for %s=%s; "+ \
"constructing one from full-frame bias") % \
(binroikey,binroi)
p60roiflat(fullbias,biasname,binning,roi)
# Produce the flatfield if necessary
if not os.path.exists(flatname):
p60roiflat(fullflat,flatname,binning,roi)
# Produce the bad-pixel mask if necessary
if not os.path.exists(bpmname):
# Get BPM into FITS format
bpm1=iraf.mktemp("iqbpm")+".fits"
iraf.imcopy(fullbpm,bpm1,verbose=no)
# Adjust BPM for binning/ROI
bpm2=iraf.mktemp("iqbpm")+".fits"
p60roibpm(bpm1,bpm2,binning,roi)
# Convert BPM to PL format
iraf.imcopy(bpm2,bpmname,verbose=no)
# Delete temporary files
iraf.imdel(bpm1,verify=no,go_ahead=yes)
iraf.imdel(bpm2,verify=no,go_ahead=yes)
# Bias subtraction
check_exist(biasname,"r")
ccdproc.zero=biasname
image1=biaspfx+image
check_exist(image1,'w',clobber)
iraf.ccdproc(image,output=image1,Stdout=1)
# Flatfielding
check_exist(flatname,"r")
iraf.iqflatten(image1,flatname,outpfx=flatpfx,
normflat=no,clipflat=no,
statsec=statsec,subsky=yes,
vignflat=no,clobber=yes,verbose=no)
imagef=flatpfx+image1
# Clip (bookend) flattened & sky-subtracted image
skybkg=float(get_head(image1,'SKYBKG'))
iraf.iqclip(image1,lthresh=minraw-skybkg,hthresh=maxraw-skybkg,
bookend=yes,maskin="",maskout="",clobber=yes,
verbose=no)
# Rename the file, if everything is well-behaved
reim=re.search('^%s%s(.+[^r])r?\.fits' % (flatpfx,biaspfx),
imagef,re.I)
if reim:
image2=reim.group(1)+'p.fits'
check_exist(image2,"w",yes)
iraf.imrename(imagef,image2,verbose=verbose)
else:
image2=imagef
# Set bad pixels to zero
update_head(image2,'BPM',bpmname,"Bad Pixel Mask")
iraf.iqmask(image2,mask='!BPM',method='constant',value='0.0',
clobber=yes,verbose=no)
# Defringing happens later, in bunches
# Cosmic Ray Rejection
tmpimg=iraf.mktemp("iqcr")+".fits"
check_exist("%s_%s.fits" % (image2[:15],crsfx),"w",clobber)
iraf.lacos_im(image2,tmpimg,"%s_%s.fits" % (image2[:15],crsfx),
gain=gain,readn=readn,skyval=skybkg,sigclip=4.5,
sigfrac=0.5,objlim=1.0,niter=1)
iraf.imdel(image2,verify=no,go_ahead=yes)
iraf.imcopy(tmpimg,image2,verbose=no)
iraf.imdel(tmpimg,verify=no,go_ahead=yes)
# Object-detection
if scistat.has_key("SEEING_%s" % file):
try:
fwhm=float(scistat["SEEING_%s" % file])
except:
fwhm=fwhm0
else:
fwhm=fwhm0
iraf.iqobjs(image2,sigma,satval,skyval="!SKYBKG",masksfx=masksfx,
wtimage="",fwhm=fwhm,pix=rawpix,aperture=2*fwhm/rawpix,
gain=gain,minlim=no,clobber=yes,verbose=no)
# Track some status information
nstars=get_head(image2,'NSTARS')
scistat['NSTARS']=nstars
if minskys.has_key(filt):
minsky=minskys[filt]
scistat["SKYBKG_%s" % filt]=skybkg/minsky
# Is it a focus frame?
focusval=get_head(image2,focuskey)
# Focus frame actions
if re.search(focusre,focusval,re.I):
focusfiles.append(image2)
# Find its place in the sequence
focusseq=get_head(image2,fseqkey)
ref=re.search("(\d+)of(\d+)",focusseq,re.I)
if ref:
ifocus=int(ref.group(1))
nfocus=int(ref.group(2))
# Non-focus frame actions
else:
# Make a new keyword for mosaic objects
objkey='OBJECT'
if check_head(image2,'MSCSIZE'):
mscsize=get_head(image2,'MSCSIZE')
mscdims=mscsize.split(',')
if int(mscdims[0])>1 or int(mscdims[1])>1:
object=get_head(image2,'OBJECT')
mscposn=get_head(image2,'MSCPOSN')
objmsc=object+'-'+mscposn
update_head(image2,'OBJMSC',objmsc,
'Object Name/Mosaic Position')
objkey='OBJMSC'
# Refine WCS (also calculates seeing in arcsec)
iraf.iqwcs(image2,objkey=objkey,rakey='RA',
deckey='DEC',pixtol=0.05,starfile='!STARFILE',
nstar=40,catalog='web',ubhost="localhost",
diffuse=yes,clobber=yes,verbose=verbose,
nstarmax=40)
# Retry if unsuccessful
if not check_head(image2,'IQWCS'):
iraf.iqwcs(image2,objkey=objkey,rakey='RA',
deckey='DEC',pixtol=0.05,starfile="!STARFILE",
nstar=40,catalog='web',ubhost="localhost",
diffuse=yes,clobber=yes,verbose=verbose,
nstarmax=1000)
# Calculate further quantities if WCS was successful
extinct='INDEF'
if check_head(image2,'IQWCS'):
scistat['WCSGOOD'] = 1
scistat["SEEING_%s" % filt]=get_head(image2,'SEEING')
[rapt,dcpt,lenx,leny]=get_head(image2,
['RA','DEC','NAXIS1','NAXIS2'])
[[ractr,dcctr]]=impix2wcs(image2,0.5*(1+float(lenx)),
0.5*(1+float(leny)))
coopt=astrocoords(rapt,dcpt)
cooctr=astrocoords(ractr,dcctr)
[offra,offdc]=radecdiff(coopt.radeg(),coopt.dcdeg(),
cooctr.radeg(),cooctr.dcdeg())
update_head(image2,'PTOFFSET',"%.2f %.2f" % \
(offra,offdc),
"Offset from nominal pointing (arcsec)")
scistat['PTOFFSET']="%.2f,%.2f" % (offra,offdc)
# Get zero-point against SDSS/NOMAD/USNO-B
if catmags.has_key(filt):
if check_head(image2,'OBJMSC'):
object=get_head(image2,'OBJMSC')
else:
object=get_head(image2,'OBJECT')
zpcat=""
# See if attempt already made for SDSS
if not (os.path.exists("%s.sdss" % object)):
getsdss(image2, "%s.sdss" % object)
# If successful, set to SDSS
sdssfile=open("%s.sdss" % object)
if (len(sdssfile.readlines())>2):
catalog="%s.sdss" % object
zpcat="SDSS"
# If SDSS unsuccessful, update USNO-B
if not zpcat:
update_usnob("%s.cat" % object,
outfile="%s.reg" % object)
catalog="%s.reg" % object
zpcat="USNO-B"
# Update image header
update_head(image2,"ZPCAT",zpcat)
# Run the zeropoint calculation
iraf.iqzeropt(image2,catmags[filt],
starfile="!STARFILE",catalog=catalog,
pixtol=3.0,useflags=yes,maxnum=50,
method="mean",rejout=1,fencelim=0.50,
sigma=1.5,maxfrac=0.25,
zptkey=zptkey,zpukey=zpukey,
clobber=yes,verbose=verbose)
# Use that to get an extinction estimate
zeropt,zeroptu,exptime=get_head(image2,[zptkey,
zpukey,exptmkey])
zeropt=float(zeropt)
zeroptu=float(zeroptu)
exptime=float(exptime)
scistat["ZEROPT_%s" % filt] = \
zeropt - 2.5*math.log10(exptime/exptmstd)
if minzpts.has_key(filt):
extinct = zeropt - \
2.5*math.log10(exptime/exptmstd) - \
minzpts[filt]
scistat["EXTINCT_%s" % filt]=extinct
# Fix the SEEING keyword if WCS fit was not successful
else:
scistat['WCSGOOD'] = 0
try:
seepix=float(get_head(image2,'SEEPIX'))
seeing= "%.3f" % (pix*seepix)
except:
seeing='INDEF'
update_head(image2,'SEEING',seeing,
"Estimated seeing in arcsec or INDEF")
# Write Archive Keywords
update_head(image2,'PROCESSD',1,
"Image has been processed by iqp60")
update_head(image2,'PROCVER',version,
"Version number of iqp60 used")
update_head(image2,'EXTINCT',extinct,
"Estimated mags extinction or INDEF")
update_head(image2,'PROCPROB','None',
"Problems encountered in iqp60 processing")
# Update status file
p60status(statfile,image2,scistat)
# Clean up
check_exist(image1,"w",yes)
##############################
# If a set of focus images have been processed, find best focus
if len(focusfiles)>=nfocus:
print "%d focus images have been processed" % len(focusfiles)
# Pick a starfile
focusfiles.sort()
medfile=focusfiles[len(focusfiles)/2]
starfile=get_head(medfile,'STARFILE')
# Update seeing values in all images based on good
# stars in the starfile.
ffiles=','.join(focusfiles)
iraf.iqseeing(ffiles,stars=starfile,
seekey="SEEING",method="acorr",useflags=yes,
skipstars=0,usestars=10,boxsize=64,
strictbox=no,imgsec="[940:1980,64:1980]",
update=yes,clobber=yes,verbose=no)
# Grab focus positions and seeing values
focx=[]
focy=[]
for image in focusfiles:
[fpos,seep]=list2float(get_head(image,['FOCUSPOS','SEEING']))
fpos=float(fpos)
seep=float(seep)
focx.append(fpos)
pix=rawpix
if check_head(image,binkey):
binval=get_head(image,binkey)
binels=binval.split()
binning=[int(binels[0]),int(binels[1])]
# Let's hope we don't ever do asymmetric binning
pix=binning[0]*rawpix
focy.append(pix*seep)
update_head(image,'SEEING',pix*seep,
"Estimated seeing in arcsec")
# Choose best-seeing as minimum of these
bestsee=min(focy)
bestfoc=focx[focy.index(bestsee)]
# This was too much verbosity for Brad
if verbose and 0:
print "Focus settings: "+liststr(focx,"%.2f")
print "Seeing values: "+liststr(focy,"%.1f")
print
print "Best focus: %.2f" % bestfoc
# Expunge the focusfiles list
allfocus.append(focusfiles)
focusfiles=[]
##############################
# Test end conditions
if endhow=="never":
done=no
elif endhow=="once":
done=yes
elif endhow=="time":
if time.time()>reftime:
done=yes
# Wait a little while
if not done:
time.sleep(twait)
def saofocus(inpat,prepfile=yes,endhow="never",endwhen="",
clobber=globclob,verbose=globver):
""" derive best focus from an SAOFOCUS run on P60 """
# Defaults
twait=30
reduced={}
bpmname="BPM.pl"
saokey="IMGTYPE"
saore="SAOFOCUS"
fseqkey="SAOFVALS"
sigma=2.0
satval=50000.0
masksfx="mask"
pix=0.3787 # arcsec per pixel
mindiff=2 # minimum tolerable diff b/ best and worst seeing (pix)
# Parse end-condition "time"
if endhow=="time":
re1=re.search("^(\d+):(\d+)",endwhen)
if re1:
tnow=time.gmtime()
# Force UT calculation w/ ignoring of DST issues
reftime=time.mktime([tnow[0],tnow[1],tnow[2],
int(re1.group(1)),int(re1.group(2)),0,
tnow[6],tnow[7],0]) - time.timezone
if reftime<time.time():
reftime+=86400
if verbose:
print "Running until %s" % \
time.strftime("%d %b %H:%M",time.gmtime(reftime))
else:
print "Failed to parse %s as UT time" % endwhen
print "Running until stopped..."
endhow="never"
##################################################
# Big Loop
done=no
while not done:
# Parse inputs
allfiles=glob.glob(inpat)
newfiles=[]
for image in allfiles:
if not reduced.has_key(image):
# Interested only in SAOFOCUS images
saoval=get_head(image,saokey)
if re.search(saore,saoval,re.I):
reduced[image]=no
newfiles.append(image)
else:
# Skip
reduced[image]=yes
for image in newfiles:
if verbose:
print "Processing new SAOFOCUS image %s" % image
# Only attempt reduction once for each image
reduced[image]=yes
# Demosaic
ccdproc=iraf.ccdproc
ccdproc.overscan=yes
ccdproc.trim=yes
ccdproc.zerocor=no
iraf.iqmosaic(image,outpfx="",biassec="!BIASSEC",
trimsec="!TRIMSEC",joinkeys="DETSEC",
splitkeys="AMPSEC,CCDSEC,TRIM,OVERSCAN",
clobber=yes,verbose=verbose)
# Fix the WCS that IRAF has now clobbered
iraf.add_wcs(image,instrument="p60new")
# Get some important keywords
[raoff,decoff,saonseq,saoshift] = \
get_head(image,['RAOFF','DECOFF','NUMFOC',
'SAOSHIFT'])
# Type conversions
saonseq=int(saonseq)
saoshift=float(saoshift)
# No bias-subtraction or flat-fielding
# Subtract stair-step background if requested
if prepfile:
qtmp=iraf.mktemp("saofoc")+".fits"
iraf.imcopy(image,qtmp,verbose=no)
for iamp in [-1,1]:
for istep in range(saonseq):
qsec=iraf.mktemp("saofsc")+".fits"
y1=1024+iamp*(1+istep*saoshift)+max(0,iamp)
y2=y1+iamp*(saoshift-1)
if istep==0:
y1-=iamp
elif istep==saonseq-1:
y2=1+max(0,iamp)*2047
ymin=min(y1,y2)
ymax=max(y1,y2)
statsec='[*,%d:%d]' % (ymin,ymax)
iraf.iterstat(image+statsec,nsigrej=5,
maxiter=10,prin=no,verbose=no)
medreg=float(iraf.iterstat.median)
iraf.imarith(image+statsec,'-',medreg,
qsec,verbose=no,noact=no)
iraf.imcopy(qsec,qtmp+statsec,verbose=no)
iraf.imdel(qsec,verify=no,go_ahead=yes)
iraf.imdel(image,verify=no,go_ahead=yes)
iraf.imrename(qtmp,image,verbose=no)
# No renaming of the file
# No Bad Pixel Masking or rudimentary WCS
# Object-detection
doobj=1
if check_head(image,'STARFILE'):
if os.path.exists(get_head(image,'STARFILE')):
doobj=0
if doobj:
iraf.iqobjs(image,sigma,satval,skyval="0.0",
masksfx=masksfx,wtimage="none",minlim=no,
clobber=yes,verbose=verbose)
# Establish our "region of interest" for pixels
pixrough=800
pixminx=1024-float(decoff)/pix-pixrough/2
pixmaxx=pixminx+pixrough
pixmaxy=1023
pixminy=pixmaxy-float(raoff)/pix-(1+saonseq)*saoshift-pixrough/2
# The array of focus positions (from low Y to high Y)
farrkeys=[]
for i in range(saonseq):
farrkeys.append("SAOFOC%02d" % (i+1))
farr=get_head(image,farrkeys)
# Starlist from Sextractor
starfile=get_head(image,'STARFILE')
stars=Starlist(starfile)
stars.mag_sort()
# Find some of the stars in the sequence
xvals=[]
yvals=[]
good1=[]
for star in stars:
# We're going in order from bright to faint
if star.xval>pixminx and star.xval<pixmaxx and \
star.yval<pixmaxy and star.yval>pixminy and \
star.elong<3:
good1.append(star)
xvals.append(star.xval)
yvals.append(star.yval)
if len(good1)>=1.2*saonseq:
break
# Sanity checking
if len(xvals)<1:
print "No stars found in search region"
update_head(image,'SAOFOCUS',0,
"saofocus processing failed for this image")
continue
# Guess for x is median of the xvals of the goodstars
xctr=median(xvals,pick=1)
# First guess for y is median of the goodstars that fall
# within +/- (pixfine) pix of this X value
pixfine=8
yval2=[]
good2=[]
for star in good1:
if abs(star.xval-xctr)<pixfine:
good2.append(star)
yval2.append(star.yval)
# Sanity checking
if len(yval2)<1:
print "No stars found in refined search region"
update_head(image,'SAOFOCUS',0,
"saofocus processing failed for this image")
continue
yctr=median(yval2,pick=1)
# This will be our reference star
usestar=good2[yval2.index(yctr)]
[refx,refy]=[usestar.xval,usestar.yval]
# Identify additional stars, working out from the
# reference star
usestars=[usestar]
# Search increasing & decreasing in Y
# (we use a bigger box in Y, and correct for the last star
# position, because of potential for tracking errors)
for isgn in [+1,-1]:
shifty=refy
for i in range(1,saonseq):
shifty += isgn*saoshift
tgtstars=stars.starsinbox([[refx-pixfine,refx+pixfine],
[shifty-1.5*pixfine,
shifty+1.5*pixfine]])
if len(tgtstars)>0:
minmag=99.99
# Pick the brightest in the box
for tstar in tgtstars:
if tstar.mag<minmag:
svstar=tstar
minmag=tstar.mag
usestars.append(svstar)
shifty=svstar.yval
else:
# Quit searching when there are no stars in box
break
# Exclude faint stars if we have too many
# (could also drop stars until we reach the number, but
# then we would want to check that all the y-values are in
# strict succession...)
magrough=2.0
if len(usestars)>saonseq:
usemags=[]
for star in usestars:
usemags.append(star.mag)
minmag=min(usemags)
use2=[]
for star in usestars:
if star.mag<minmag+magrough:
use2.append(star)
if len(use2)==saonseq:
usestars=use2
# Check for the right number
if len(usestars)!=saonseq:
print "Failed to get right number of target stars"
update_head(image,'SAOFOCUS',0,
"saofocus processing failed for this image")
continue
# Construct the array of x- and y-values
xval3=[]
yval3=[]
for star in usestars:
xval3.append(star.xval)
yval3.append(star.yval)
# Choose a single x-value for the run
xctr2="%.3f" % median(xval3)
# Increasing y-value means increasing index in farr
yval3.sort()
ylis=','.join(map(str,yval3))
flis=','.join(map(str,farr))
# Run iqfocus
iraf.iqfocus(image,xctr2,ylis,flis,method="acorr",
boxsize=saoshift,focuskey="BESTFOC",
seekey="SEEING",fseekey="SAOSEE",
update=yes,clobber=yes,verbose=verbose)
# Check for successful iqfocus run
if not check_head(image,'BESTFOC'):
print "iqfocus run failed to find good focus"
update_head(image,'SAOFOCUS',0,
"saofocus processing failed for this image")
continue
# Grab focus positions and seeing values
bestfoc,saosee=get_head(image,['BESTFOC','SAOSEE'])
# Some information
if verbose:
print "Focus settings: "+flis
print "Seeing values: "+saosee
print "Best focus: %.3f" % bestfoc
# Check for best-focus "near edge" condition
for i in range(saonseq):
if ("%.3f" % bestfoc)==("%.3f" % float(farr[i])):
ibest=i
break
if (saonseq>3 and (ibest==0 or ibest==saonseq-1)) or \
(saonseq>5 and (ibest==1 or ibest==saonseq-2)):
update_head(image,'SAOEDGE',1,
"Best seeing is near edge of focus run")
# Check for insufficient variation across the focus run
maxsee=0
minsee=0.5*saoshift
saoseevals=eval(saosee)
for seeval in saoseevals:
minsee=min(minsee,seeval)
maxsee=max(maxsee,seeval)
if (maxsee-minsee) < mindiff:
print "Failed to observe significant variation "+\
"across focus run"
update_head(image,'SAOFOCUS',0,
"saofocus processing failed for this image")
continue
# Convert seeing value to arcsec
seepix=get_head(image,'SEEPIX')
update_head(image,'BESTSEE',seepix,
"Best seeing in pixels for this series")
try:
seeasec=float(seepix)*pix
update_head(image,'SEEING',seeasec,
"Best seeing in arcsec for this series")
except:
print "Failed to convert seeing to arcsec for %s" % image
# Successful processing
update_head(image,'SAOFOCUS',1,
"saofocus processing succeeded for this image")
##############################
# Test end conditions
if endhow=="never":
done=no
elif endhow=="once":
done=yes
elif endhow=="time":
if time.time()>reftime:
done=yes
# Wait a little while
if not done:
time.sleep(twait)
def saofocus(inpat,
prepfile=yes,
endhow="never",
endwhen="",
clobber=globclob,
verbose=globver):
""" derive best focus from an SAOFOCUS run on P60 """
# Defaults
twait = 30
reduced = {}
bpmname = "BPM.pl"
saokey = "IMGTYPE"
saore = "SAOFOCUS"
fseqkey = "SAOFVALS"
sigma = 2.0
satval = 50000.0
masksfx = "mask"
pix = 0.3787 # arcsec per pixel
mindiff = 2 # minimum tolerable diff b/ best and worst seeing (pix)
# Parse end-condition "time"
if endhow == "time":
re1 = re.search("^(\d+):(\d+)", endwhen)
if re1:
tnow = time.gmtime()
# Force UT calculation w/ ignoring of DST issues
reftime = time.mktime([
tnow[0], tnow[1], tnow[2],
int(re1.group(1)),
int(re1.group(2)), 0, tnow[6], tnow[7], 0
]) - time.timezone
if reftime < time.time():
reftime += 86400
if verbose:
print "Running until %s" % \
time.strftime("%d %b %H:%M",time.gmtime(reftime))
else:
print "Failed to parse %s as UT time" % endwhen
print "Running until stopped..."
endhow = "never"
##################################################
# Big Loop
done = no
while not done:
# Parse inputs
allfiles = glob.glob(inpat)
newfiles = []
for image in allfiles:
if not reduced.has_key(image):
# Interested only in SAOFOCUS images
saoval = get_head(image, saokey)
if re.search(saore, saoval, re.I):
reduced[image] = no
newfiles.append(image)
else:
# Skip
reduced[image] = yes
for image in newfiles:
if verbose:
print "Processing new SAOFOCUS image %s" % image
# Only attempt reduction once for each image
reduced[image] = yes
# Demosaic
ccdproc = iraf.ccdproc
ccdproc.overscan = yes
ccdproc.trim = yes
ccdproc.zerocor = no
iraf.iqmosaic(image,
outpfx="",
biassec="!BIASSEC",
trimsec="!TRIMSEC",
joinkeys="DETSEC",
splitkeys="AMPSEC,CCDSEC,TRIM,OVERSCAN",
clobber=yes,
verbose=verbose)
# Fix the WCS that IRAF has now clobbered
iraf.add_wcs(image, instrument="p60new")
# Get some important keywords
[raoff,decoff,saonseq,saoshift] = \
get_head(image,['RAOFF','DECOFF','NUMFOC',
'SAOSHIFT'])
# Type conversions
saonseq = int(saonseq)
saoshift = float(saoshift)
# No bias-subtraction or flat-fielding
# Subtract stair-step background if requested
if prepfile:
qtmp = iraf.mktemp("saofoc") + ".fits"
iraf.imcopy(image, qtmp, verbose=no)
for iamp in [-1, 1]:
for istep in range(saonseq):
qsec = iraf.mktemp("saofsc") + ".fits"
y1 = 1024 + iamp * (1 + istep * saoshift) + max(
0, iamp)
y2 = y1 + iamp * (saoshift - 1)
if istep == 0:
y1 -= iamp
elif istep == saonseq - 1:
y2 = 1 + max(0, iamp) * 2047
ymin = min(y1, y2)
ymax = max(y1, y2)
statsec = '[*,%d:%d]' % (ymin, ymax)
iraf.iterstat(image + statsec,
nsigrej=5,
maxiter=10,
prin=no,
verbose=no)
medreg = float(iraf.iterstat.median)
iraf.imarith(image + statsec,
'-',
medreg,
qsec,
verbose=no,
noact=no)
iraf.imcopy(qsec, qtmp + statsec, verbose=no)
iraf.imdel(qsec, verify=no, go_ahead=yes)
iraf.imdel(image, verify=no, go_ahead=yes)
iraf.imrename(qtmp, image, verbose=no)
# No renaming of the file
# No Bad Pixel Masking or rudimentary WCS
# Object-detection
doobj = 1
if check_head(image, 'STARFILE'):
if os.path.exists(get_head(image, 'STARFILE')):
doobj = 0
if doobj:
iraf.iqobjs(image,
sigma,
satval,
skyval="0.0",
masksfx=masksfx,
wtimage="none",
minlim=no,
clobber=yes,
verbose=verbose)
# Establish our "region of interest" for pixels
pixrough = 800
pixminx = 1024 - float(decoff) / pix - pixrough / 2
pixmaxx = pixminx + pixrough
pixmaxy = 1023
pixminy = pixmaxy - float(raoff) / pix - (
1 + saonseq) * saoshift - pixrough / 2
# The array of focus positions (from low Y to high Y)
farrkeys = []
for i in range(saonseq):
farrkeys.append("SAOFOC%02d" % (i + 1))
farr = get_head(image, farrkeys)
# Starlist from Sextractor
starfile = get_head(image, 'STARFILE')
stars = Starlist(starfile)
stars.mag_sort()
# Find some of the stars in the sequence
xvals = []
yvals = []
good1 = []
for star in stars:
# We're going in order from bright to faint
if star.xval>pixminx and star.xval<pixmaxx and \
star.yval<pixmaxy and star.yval>pixminy and \
star.elong<3:
good1.append(star)
xvals.append(star.xval)
yvals.append(star.yval)
if len(good1) >= 1.2 * saonseq:
break
# Sanity checking
if len(xvals) < 1:
print "No stars found in search region"
update_head(image, 'SAOFOCUS', 0,
"saofocus processing failed for this image")
continue
# Guess for x is median of the xvals of the goodstars
xctr = median(xvals, pick=1)
# First guess for y is median of the goodstars that fall
# within +/- (pixfine) pix of this X value
pixfine = 8
yval2 = []
good2 = []
for star in good1:
if abs(star.xval - xctr) < pixfine:
good2.append(star)
yval2.append(star.yval)
# Sanity checking
if len(yval2) < 1:
print "No stars found in refined search region"
update_head(image, 'SAOFOCUS', 0,
"saofocus processing failed for this image")
continue
yctr = median(yval2, pick=1)
# This will be our reference star
usestar = good2[yval2.index(yctr)]
[refx, refy] = [usestar.xval, usestar.yval]
# Identify additional stars, working out from the
# reference star
usestars = [usestar]
# Search increasing & decreasing in Y
# (we use a bigger box in Y, and correct for the last star
# position, because of potential for tracking errors)
for isgn in [+1, -1]:
shifty = refy
for i in range(1, saonseq):
shifty += isgn * saoshift
tgtstars = stars.starsinbox(
[[refx - pixfine, refx + pixfine],
[shifty - 1.5 * pixfine, shifty + 1.5 * pixfine]])
if len(tgtstars) > 0:
minmag = 99.99
# Pick the brightest in the box
for tstar in tgtstars:
if tstar.mag < minmag:
svstar = tstar
minmag = tstar.mag
usestars.append(svstar)
shifty = svstar.yval
else:
# Quit searching when there are no stars in box
break
# Exclude faint stars if we have too many
# (could also drop stars until we reach the number, but
# then we would want to check that all the y-values are in
# strict succession...)
magrough = 2.0
if len(usestars) > saonseq:
usemags = []
for star in usestars:
usemags.append(star.mag)
minmag = min(usemags)
use2 = []
for star in usestars:
if star.mag < minmag + magrough:
use2.append(star)
if len(use2) == saonseq:
usestars = use2
# Check for the right number
if len(usestars) != saonseq:
print "Failed to get right number of target stars"
update_head(image, 'SAOFOCUS', 0,
"saofocus processing failed for this image")
continue
# Construct the array of x- and y-values
xval3 = []
yval3 = []
for star in usestars:
xval3.append(star.xval)
yval3.append(star.yval)
# Choose a single x-value for the run
xctr2 = "%.3f" % median(xval3)
# Increasing y-value means increasing index in farr
yval3.sort()
ylis = ','.join(map(str, yval3))
flis = ','.join(map(str, farr))
# Run iqfocus
iraf.iqfocus(image,
xctr2,
ylis,
flis,
method="acorr",
boxsize=saoshift,
focuskey="BESTFOC",
seekey="SEEING",
fseekey="SAOSEE",
update=yes,
clobber=yes,
verbose=verbose)
# Check for successful iqfocus run
if not check_head(image, 'BESTFOC'):
print "iqfocus run failed to find good focus"
update_head(image, 'SAOFOCUS', 0,
"saofocus processing failed for this image")
continue
# Grab focus positions and seeing values
bestfoc, saosee = get_head(image, ['BESTFOC', 'SAOSEE'])
# Some information
if verbose:
print "Focus settings: " + flis
print "Seeing values: " + saosee
print "Best focus: %.3f" % bestfoc
# Check for best-focus "near edge" condition
for i in range(saonseq):
if ("%.3f" % bestfoc) == ("%.3f" % float(farr[i])):
ibest = i
break
if (saonseq>3 and (ibest==0 or ibest==saonseq-1)) or \
(saonseq>5 and (ibest==1 or ibest==saonseq-2)):
update_head(image, 'SAOEDGE', 1,
"Best seeing is near edge of focus run")
# Check for insufficient variation across the focus run
maxsee = 0
minsee = 0.5 * saoshift
saoseevals = eval(saosee)
for seeval in saoseevals:
minsee = min(minsee, seeval)
maxsee = max(maxsee, seeval)
if (maxsee - minsee) < mindiff:
print "Failed to observe significant variation "+\
"across focus run"
update_head(image, 'SAOFOCUS', 0,
"saofocus processing failed for this image")
continue
# Convert seeing value to arcsec
seepix = get_head(image, 'SEEPIX')
update_head(image, 'BESTSEE', seepix,
"Best seeing in pixels for this series")
try:
seeasec = float(seepix) * pix
update_head(image, 'SEEING', seeasec,
"Best seeing in arcsec for this series")
except:
print "Failed to convert seeing to arcsec for %s" % image
# Successful processing
update_head(image, 'SAOFOCUS', 1,
"saofocus processing succeeded for this image")
##############################
# Test end conditions
if endhow == "never":
done = no
elif endhow == "once":
done = yes
elif endhow == "time":
if time.time() > reftime:
done = yes
# Wait a little while
if not done:
time.sleep(twait)
