From gtaylor@cv3.cv.nrao.edu Fri Sep 14 13:31:15 2001 Date: Fri, 14 Sep 2001 12:56:24 -0600 (MDT) From: gtaylor To: Frazer Owen Cc: Steven Myers , Jim Ulvestad Subject: aips++ testing log for Sept. 14th Testing AIPS++ Detailed Log Sep 14, 2001, session: Started up testing to try out the advertised changes to msplot. Working on AT166 data in stable 1.6 Build 195. Running fitstoms read the data then bombed: - warning, event glishtk.fail () dropped LocalExec::SetStatus: abnormal child termination for /home/cluster/aips++/stable/linux_gnu/bin/glishtk error, new connection immediately broken Aren't we ever going to get past this? Very annoying. starting up aips++ again and running fitstoms.... This time the timer broke: timer - callback tmr1 failed : "progress.g", line 270: self.barholder is not an agent File: progress.g, Line 268 Stack: .(), servers.g line 641 .(), widgetserver.g line 747 .(), progress.g line 217 .(), progress.g line 298 .(), progress.g line 334 but it seems to have worked and didn't dump me out of aips++ The search string function is no longer doing anything useful. That or I can't figure out how to use it. I thought I knew. At the minimum it is not very intuitive. George agrees it is broken. Bug AOCso02983 aips2-namerica Submitted 010914 NEW defect report AIPS++, version 1.6 (bui 1 enclosure "search string on manager is busted" Eventually found msplot. Fired it up and managed to create a u,v plot. Now to check out some editting modes. Lets try to display data as an image... Lattice shapes inconsistent Caught an exception! Event type=create exception=Fri Sep 14 14:52:49 2001 SEVERE image::imageconcat (file /aips++/stable/code/trial/implement/Images/DOimage.cc, line 363) ObjectID=[sequence=67 host=pegasus pid=12912 time=1000479167]: Lattice shapes inconsistent Failed to construct scratch image. Error was: Caught an exception! Event type=create exception=Fri Sep 14 14:52:49 2001 SEVERE image::imageconcat (file /aips++/stable/code/trial/implement/Images/DOimage.cc, line 363) ObjectID=[sequence=67 host=pegasus pid=12912 time=1000479167]: Lattice shapes inconsistent Failed to construct scratch image. Error was: Caught an exception! Event type=create exception=Fri Sep 14 14:52:49 2001 SEVERE image::imageconcat (file /aips++/stable/code/trial/implement/Images/DOimage.cc, line 363) ObjectID=[sequence=67 host=pegasus pid=12912 time=1000479167]: Lattice shapes inconsistent I went and got George and he showed me how to unset the defaults in the data selection in order to get this to work. This should not be necessary! Bug AOCso02984 aips2-namerica Submitted 010914 NEW defect report AIPS++, version 1.6 (bui 1 enclosure "displaying data fails" Trying again now. When I try to look at my data as an image it plots time on the x axis (different convention from TVFLG but ok), and kilometers on the y axis from 1 to -1. The y axis should be baselines from 1 to 351. Even if those baseline were somehow converted to km, they wouldn't be evenly spaced from 1 to -1. More important than this though is the fact that the cursor values don't reflect what is in the viewer. Also the values displayed look too consistent to be believed along each row. I don't think that the right data is getting displayed. This is certainly a show stopper for anybody trying to edit data with msplot. Bug AOCso02985 aips2-namerica Submitted 010914 NEW defect report AIPS++, version 1.6 (bui 1 enclosure "displaying data fails" (should have given this a different subject) OK, lets give up on the TVFLG-like capability for now. Trying simple editing. I'm plotting amplitude vs time and iterating through the plots one at a time. I would like to be able to edit out bad data when I see it, but msplot won't let me. It show's a picture of a watch and all I can do is go on to the next plot or stop. If I stop then it shows multiple plot/page and tells me I can't edit. The only way to edit is to make a note of the bad data, stop the iterative plot, then use the data selection to bring up only the baseline with the bad data. This is rather cumbersome. Would be much better if we could edit on the individual plots as they come up (as in Difmap). Bug AOCso02986 aips2-namerica Submitted 010914 NEW enhancement AIPS++, version 1.6 (bui 1 enclosure "can't edit while viewing plots" Another problem. I'm looking at RR but I want to flag all correlations. The flag all correlations box is greyed out. This is not good. One needs to be able to expand the scope of the edits to get rid of all bad data. I see that if I plot all the data then the box goes active, but this isn't the behaviour I want since it is harder to find the bad data with the RL and LR data overlaid. Bug AOCso02987 aips2-namerica Submitted 010914 NEW defect report AIPS++, version 1.6 (bui 1 enclosure "can't expand scope of edits" That's enough for today. I'll try another dataset next week if I have time. In my opinion the msplot tool is still quite a ways off from being ready for the general VLA user. In this morning's session I found 5 bugs, 4 of them related to msplot, and one of them an absolute show stopper. --- end of testing. Frustration meter reads: |********** | low high