Pull request #1068: CAS-14763
Merge in CASA/casa6 from CAS-14763 to master
* commit 'f2e0db200c55d4b3c26e06bf80d6211944a47756': (74 commits)
Change type to G
Add standard errors for parameters
Update xml and tests
Change default clock order and update the test data origin
Remove debugging list type, return of task zenithdelaycal is dict
Updated test to use data with all VLBA antennas
Updated test header
Add test data procedence
Adds more tests
Add per polarization clock solution
Switch to CASA measures to compute antenna elevations
Updated tests for zenithdelaycal
Name based antenna selection from MS
Add implementation to ZenithDelayJones
Add tests
Add implementation to ZenithDelayJones
Add caltable output
For CAS-14763, added comments to aid development
For CAS-14763, in ZenithDelayJones::setApply(Record&), removed some unnecessary stuff.
For CAS-14763, added skeletal ZenithDelayJones class (ZenithDelayCal.{h,cc}) to eventually support application of zenith-angle dependent macroscopic delay calibration for VLBI, and make VisCalGlobals's SolvableVisCal factories aware of it. This requires further development for it to work.
...
Pull request #1063: CAS-14797
Merge in CASA/casa6 from CAS-14797 to master
* commit 'd0332c4e3cda643f0a1b8a128c35f1a6ee2e973f':
Streamline the implementations of helper functions in just two helper functions now made available in MatrixCleaner.cc
Search only over inner square
Make MSMT-CLEAN compatible with previous runs again
repeat same accelerations for AspMatrixCleaner and for MultiTermMatrixCleaner
This code will increase the speed for MS-CLEAN. This is achieved by two main improvements: -> Subtract beam is faster than casacore functionality which creates temporary copies -> Run peak finding only over square determined by blcdirty and trcdirty rather than the full image -> again, the direct implementation is faster than casacore minMax functionality
Pull request #1062: CAS-14814
Merge in CASA/casa6 from CAS-14814 to master
* commit 'cb65c9b710137b44a5624dadade29b814c6a2a3d':
Remove dead code
labeling, pruning and blob-size counting in memory, avoid temporary copies
Optimize Automasking region labelling and pruning
Main bottleneck is binaryDilation computation. This part of the code has been accelerated now, by: 1.) Remove IPosition constructions from inner loop 2.) Use direct storage pointers 3.) Precompute structure offsets 4.) Return changed from BinaryDilationCore 5.) Remove the full image check in BinaryDilation() 6.) Avoid extra full-plane scans
Pull request #1061: CAS-14804
Merge in CASA/casa6 from CAS-14804 to master
* commit 'b21128a27d35f918b11d60dad2d1ff729e92d923':
Inflate epsilon for passing tests across platforms
Add tclean task for CAS-14804, depends on fasr dataset
Update comparison values in test_onefield_asp
Reverse to normalization with psf width again, this makes the unittest testonefieldasp performing superior to old version again
Solution by rescaling the objfunc, and amplitudes, for more details see description at Jira ticket
