From gmoellen@cv3.cv.nrao.edu Wed Nov 14 11:20:07 2001 Date: Wed, 14 Nov 2001 11:12:25 -0700 (MST) From: George Moellenbrock To: aips2-naug@zia.aoc.NRAO.EDU Subject: recent defect fixes Hi gang, I've made the following defect fixes in the past 7-10 days. Most are available in weekly/daily. As I understand it, the new release version (frozen early Monday) will be declared stable any day now, and all will then be available. 1. calibrater.plotcal() now plots bandpass solutions for all antennas (or just those you select. 2. The calibrater tool now has a state() and reset(apply=T/F,solve=T/F) functions so you can keep track of what you have set up to apply and/or solve for, and it is no longer necessary to done() the calibrater so often; it really behaves like a "tool" is supposed to, now. Also, setapply() and setsolve() have a new option ('unset') which enables you to unset calibration types one-at-a-time (rather than reset(), which clears everything). Check the calibrater reference manual docs for more details. (I'll be improving the examples there shortly.) 3. The flagger.filter() function now (daily only, until next week) knows about editing scope. It now uses the settings reported by flagger.state() to select data that should be tested (e.g., in a 'range' [=clipping] opertion), and then edits only that data which fails the range test, and if the new params--fullpol &/| fullchan--are set T, expands the edits to all pols &/| channels. I haven't yet expanded the pre-flagging selection to time ranges yet, but this is coming soon (and will be in the release, I think). 4. The entire "Getting Results in AIPS++" document has undergone extensive editing, mainly by Neil Killeen. It is now divided into 3 sections, generic processing, standard radioastronomy processing, and specific telescopes. I have contributed major revisions of the synthesis calibration chapter (in the "standard" section) and the VLA reduction chapter (in the specific telescopes section). Both are in pretty good shape now (tho' improvements still possible, of course). The VLA suffered a bit from Neil's strong editing pass (mainly to do with handling the source model for poln calibration), but it is otherwise ok. If you like, please take a look at these chapters. I will be fixing up the poln-cal part of the VLA chapter, and bringing both chapters up-to-date w.r.t. the items listed above over the next few days, so maybe it would be good to wait until early next week to pour over them in detail. Whatever your pleasure. At the moment, I am working on providing the long-awaited LISTR 'SCAN' listing. This should be available shortly. Below, I append an email from John Hibbard. He had a few things he wanted to report in your meeting. Cheers, George On Mon, 12 Nov 2001, John E. Hibbard wrote: > > Date: Mon, 12 Nov 2001 17:49:25 -0500 > From: John E. Hibbard > To: George Moellenbrock > Subject: Re: Glish to plot from gain table > > George, > > Thanks for the script. I started with that, but > couldn't lean much from plotting everything on > top of each other, so changed it a bit. I've attached > it below in case you are interested. Still not exactly > what I want, but I am still feeling my way around glish > and may improve on this yet. > > I changed some of the terminology. Most notably, > changing the "row" descriptor to "antenna" - > to make it clear we are looking at results per > antenna, vs a row of the ms. > > Something which is unclear in the documentation > is that pgplot commands like env and sci will not > be in the aips++ documentation, and I think there > should be a link to it on the pgplotter page. But > perhaps people will find it like I did, and this is > a busy time so I decided not to bug that. > > You might mention in this weeks meeting that I have > spent 2 sessions in the new msplot, and found nothing > to bug. I have been concentrating on comparing the > results for my spectral line data set between aips++ > and AIPS (I find significant differences between > the calibration, and need compare on a step-by-step > basis to see what is going on), so haven't done > a lot with it, but it looks good. Its a nice tool, > and will be a big improvement on AIPS when it gets some > of the modifications that I know are in the works for > the next release (like plotting delta-phase, delta-amp, > and running means). > > I'll be out most of the next three weeks, so have > productive meetings! > > -john > > ================================================================= > include 'table.g' > include 'pgplotter.g' > t:=table('n4038D.bcal') > ants:=t.getcol('ANTENNA1') + 1; > Nants:=shape(ants) > > gain:=t.getcol('GAIN'); # the indicies on the gain array are > # [pol1,pol2,spw,channel,antenna] > # pol1 and pol2 are 1 or 2, and are > # meaningful for B tables only if pol1=pol2 > solok:=t.getcol('SOLUTION_OK'); > # solok contains a 3D array of booleans > # indicating which [spw,channel,antenna] > # indicies are ok > > function plot1BP(i_ant,xylimits,gaintable,okarray) > { > Nchan:=shape(gaintable)[4] > Npol:=shape(gaintable)[1] > mypg.env(xylimits[1], xylimits[2], xylimits[3], xylimits[4], 0,0); > > if (any(okarray[1,,i_ant])) # does any channel for this antenna > # have ok solutions? > { > for (pol in 1:Npol) > { > x:=seq(Nchan) # list of channel numbers, from 1..Nchan > y:=abs(gain[pol,pol,1,,i_ant]) > > # mask out the bad channels: > x:=x[okarray[1,,i_ant]]; > y:=y[okarray[1,,i_ant]]; > > mypg.sci((i_ant+pol-1)%10+2) # use colors 2-11 > mypg.pt(x,y,17) # plot y vs. x with symbol 17 > > #make and plot antenna ID labels: > xtxt:= (i_ant/Nants)*limits[2]; > ytxt:= limits[4]*0.95; > if (pol==2) ytxt:=limits[4]*0.9; > mypg.ptxt(xtxt,ytxt,0,0,spaste(ants[i_ant])); > } > } > } > > mypg:=pgplotter() > mypg.ask:=T; > > limits:= [0,Nchan+1,0,max(abs(gain))*1.1] > # first look at one plot to figure limits for rest of plots > plot1BP(1,limits,gain,solok) > > limits := [2,125,0.5,0.7] ## WARNING: HARDWIRED VALUES > for (i in 1:Nants-1) plot1BP(i,limits,gain,solok) > > #this doesn't seem to work - I want it to ask before > #plotting the next bp, so I ended up calling plot1BP > # for each antennae individually, eg. > > plot1BP(3,limits,gain,solok); > plot1BP(6,limits,gain,solok); > # etc > > # reset color to white: > mypg.sci(1) >