Almost finished with initial round of CasaCore github checkins. Will publish cheat sheet.
Looking into C++ unit test framework.
casampi failures under RH6
plotMS now can open RO so that we can run multiple in parallel
Tiling study starting
ALMA Cycle 3 R2 Branch released
Continued working on refactoring and parallelizing cfcache generation code for AProjection.
Cal Library 4.5
Discussions with George about architecture, parser, interfaces
Researched Flex/Bison parser and started on prototype for cal library file parser
Tested CAS-7390 (virtual model data) - creates default model data but time averaging is incorrect
CAS-7601 (correlation type names) tested and resolved, so CAS-7577 (correlation types in Locate) is now RTT
Alma mosaic core dumping when undersampled pixels used
SD imaging tolerance in locating the pointing
trying to reproduce a w-projection bug on a 1.7TB data
Discussions about mosaic weights etc
Justo González Villalba
mstransform tickets fixed:
CAS-7488 : Re-index optional columns in SPW sub-table affected by SPW re-indexing (ASSOC_SPW and ASSOC_NATURE)
CAS-7149: Drop empty channels of last SPW when separating a SPW into a non-divisor number of channels.
CAS-7601: Added interface for new correlation selection methods
CAS-7390: On-going investigation of why model data is averaged to 0 in AveragingTVI
Discussions with SD team on how to transform Weight/Sigma Spectrum using mstransform framework
CAS-7207: Flux density not working for multiple beam one per channel image (ready to test)
CAS-7608: View GUI testing CAS 4.4 Mac 10.10 (ready to test, verified with Pam's help)
CAS-7185: Reset rest frequency in viewer to original value (ready to test)
CAS-7617: Profiler should post warning rather than segfaulting with single channel fit (ready to test)
CARTA: Layout work
Responded to Kana re: stats framework query
(CAS-7319) MSMD additions: completed adding a couple dozen or so new methods to msmd tool. Cannot commit until casacore freeze has been lifted.
(CAS-7040) region manager wbox function can segfault casa: Resolved as won't fix because apparently this bug was serendipitously fixed at some point.
(CAS-7584) no random number seed in ia.addnoise() ?: added "seeds" parameter to ia.addnoise(). Cannot commit until casacore freeze has been lifted.
(CAS-7207) Flux density not working for multiple beam one per channel image: pointed out that differences between 4.2.2 and current version did not indicate a bug, since the current version now uses the beam in each channel to compute the flux density for that channel whereas the 4.2.2 version used a single beam in the cube to compute approximate flux densities for multi-beam images in the spectral profile tool.
[JIRA] (CAS-7257) discrepancy between spectra extracted by imval and by spectral profiler in viewer: passed back to Dirk because I could not reproduce his discrepancy.
(CAS-7329) Casacore LatticeHistograms gives a segfault when setting a range outside of data range: Fixed. Cannot commit until casacore freeze has been lifted.
Progress on parametrized types in Cal Lib (CAS-6130)
Science: EVLA circpol obs prep
Started rough design for converting visstat and sdstat to new statistics framework and vi/vb2.
Wrote up design document describing MPI usage in redesigned tclean, and what requirements the new design would impose on the CASA MPI framework.
Got normalization sync'd between MosaicFT and AProjectFT by adding missing nx*ny norm and removing fudges from MosaicFT. Made derived version with changes. Finally... we can edit normalization rules in one place and MosFT /AProj, single/multi-term, parallel=T/F all see it.
Made the output of MosFT and AProj contain weightimage (.weight) and normalized PB image (.pb) with pb = sqrt(weight)/max(sqrt(weight))
Enabled and tested multi-term mosaicing via MosaicFT - this brings wideband mosaic imaging capability for ALMA.
Several other minor edits and fixes....
Discussions (about weights...) and some user support