Monday Morning Meeting 4/1/19

  • DIAL-IN NUMBERS & PASSCODES:
  • IP: 192.33.117.12##8110
  • Phone: (434) 817-6524

News / Management
  • VLASS has a Status Review on May 9th. In April, they will test AWProjection (from ARD) with pointing offset corrections and with the mpicasa thread-leak fix. They have agreed in writing that this is all they will require for the review. We too will do internal level 2 tests. There are many additional development details to follow up on, but that will happen on regular CASA/ARD schedules.

Development

  • Calibration and Flagging
    • N/A
  • Imaging
    • CAS-11191 : Work on mosaic gridder for VLASS is on hold for now. We will decide if/when to commit the pointing correction fix (only). No stakeholder needs it for 5.5.
    • CAS-7210 : Update from Kumar ?
  • Measurement Set
    • N/A
  • Image Analysis and Casacore
    • N/A
  • Data Visualization
    • N/A
  • CASA 6
    • N/A
  • HPC
    • Update on the mpicasa thread-leak issue ? Fix needed asap for VLASS testing.
  • casacore
    • Ger van Diepen is working to remove the casacore container classes that are duplicated by the C++ Standard Template Library (STL) and replace them with the corresponding STL classes. He has submitted a pull request for these changes on GitHub. The removed classes include:
      1. List
      2. Map
      3. SimpleOrderedMap
      4. HashMap
      5. Queue
      6. Stack
    • The initial attempt at removal of these classes has been completed on the branch for JIRA issue CAS-12364. While the departing casacore classes generally have the same behavior as the STL equivalents, there are some significant caveats:
      • std::list::insert is a no-op if the mapping already exists (no changes made!)
      • std::map has no default value
      • multiple std::list::iterator objects operating on the same list are not kept in sync
      • std::list has no numeric indexing
    • Developers who are responsible for the modified sections of code will be contacted (or get a JIRA ticket) to review these changes, add any tests that are required to verify that these changes are OK or (of course) adjust and refactor as needed or desired.
    • We should go one step further than casacore and remove our use of casacore::String, but this is yet to be done (and won't be done in the CASA 5.6 cycle)
    • We should also switch from using casacore shared pointers and threading constructs to STL shared pointers and C++11 direct support for threads

Build / Testing

  • Infrastructure
    • N/A
  • Unit
    • N/A
  • Verification
    • Started level 2 tests for VLASS Imaging
  • Validation
    • Under Validation (issues remaining in validation for CASA 5.5)
      • BLOCKER CAS-7210 (CAS-4640) Connect imaging weight gridder ( TCLEAN producing arc-like noise patterns when imaging small chunks of data when robust<1 )
      • BLOCKER CAS-11868 Fix mosweight=True to use less memory (Federico has been testing for ALMA, VLASS has tested, getting VLA dataset to finish testing)
      • Need final okay on docs/release notes:
        • CAS-9108 Linear basis-specific 'Xf' (multi-scan) improvements (and several related polarization tickets) (merged; need final okay on docs. Bjorn?)
        • CAS-8087 Support plotting pointing direction (R.A.-Dec.) by plotms (finished, just waiting on finalized doc/release note)
        • CAS-10593 Make component list prediction with beam use frequency variation -- just needs doc updates, Kumar is on it
      • George: do you want to open new tickets for CAS-12168, CAS-9108/etc followup? Does CAS-12280/CAS-12237 need doc updates on cal library CASAdocs page?
      • not yet ready to validate but still desired in 5.5:
        • CAS-12368 add depreciation warning to clean
        • CAS-12302 Updating XML files of CASA tasks
        • pointing tables for awproject, Preshanth to do, not yet in a CAS ticket (?), CAS-7727? -- this is definitely for 5.6, correct?
    • Kana: status of tickets moving to PIPE?

User Liaison Updates

  • CASA Newsletter: deadline April 12. Confirmed articles (apart from usual): Parallal imaging (Sandra/Federico), CASA 6 (Darrell), CARTA (Kuo-Song) - other articles TBD

Pipeline Updates

  • N/A

Single Dish Development
  • Wataru Kawasaki
  • Takeshi Nakazato
    • CAS-12049 replied to Sanjay's comment. looks reasonable.
    • supported Renaud's verification on macOS 10.13
    • testing Visual Studio Code
  • Renaud Miel
  • Suminori Nishie
    • CAS-8418Higher order poinitng interpolation ( PointingDirectionCalculator ) , 5.6 ,"Ready toValidate"
      • Restructuring Test-fixtutres and sub modules, clear strucure, function and interface are formed. [ Suspended]
      • Two mulfunction were found [new:Working]
        1. On the edge of main table data, decreasing accuracy - under specific UT condition - were observed.
          • Bugs in available range check of rows in Test-code were found. reviewing code logic.
        2. Fragmently less-accurate result were observed in spline interpolation on specific Measurement Set.
          • specific poiniting direction movement affects spline-coefficient table generation on edges of a continuous section.Examinating good countermeasure.
    • Obtaining higher C++skills, (OOD/OOP)
      • Attempting to used some of Design Pattern in gtest-code. Singleton, Adaptor was introduced.
      • Trying to find well-designed classes and/or interface .
Topic revision: r10 - 2019-04-01, DarrellSchiebel
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback