Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 6 Next »

Date

11:00

Attendees

Discussion items

TimeItemWhoNotes
 Review last meeting 2017-08-22 Meeting notes
 Review recent commits 

Periodic boundaries in parallel build? No, not yet

VTK writers for walls and interactions Done

New tools useful for everyone:

  • helpers::getRealTime() to output time in seconds;
  • helpers::getPath() returns the path of the executable
  • if VTK files are written, a python script is created to load the files in paraview

Masers

  • segfault still an issue?
  • two masers now
 Git-Svn bridgeAnthony Thornton

Everyone: make Bitbucket account.

Private directories will stay in svn.

 Interactions: list vs. vectorFormer user (Deleted)

Every BaseInteractable has a std::list of interactions, while getInteractionWith returns a vector. Please choose one and be consistent.

 setIndSpeciesIrana Denissen

Do we want to keep setIndSpecies deprecated? It causes a lot of warnings, and the alternative is not that much nicer.

Agreed, check for range, pointer etc.

species needs to be made before particles, break old codes!

 

 Next meeting Monday 18 Septeber 9:00

 

SVN log since last meeting:

------------------------------------------------------------------------
r2490 | irana | 2017-09-01 08:56:13 +0200 (Fri, 01 Sep 2017) | 2 lines

Bug-report: Version of Maser with Segfault.

------------------------------------------------------------------------
r2489 | irana | 2017-08-31 12:56:54 +0200 (Thu, 31 Aug 2017) | 2 lines

Changed the SubcriticalMaserBoundary s.t. particles are not removed anymore.

------------------------------------------------------------------------
r2488 | irana | 2017-08-31 09:15:57 +0200 (Thu, 31 Aug 2017) | 2 lines

MaserBoundaries, new names: fixed the include-guards and added the correct names to "read". MaserBoundary will be read as a ContinuousMassFlowMaserBoundary for backwards compatibility. Also added the test-data for SubcriticalMaserBoundarySelfTest.

------------------------------------------------------------------------
r2487 | tunuguntladr | 2017-08-30 10:12:44 +0200 (Wed, 30 Aug 2017) | 1 line

Modifed the class in the HorizontalMixerRun in the MercuryLab Demos folder, as previously it was not compiling. The code was created by Thomas however the version Thomas used was a slightly different one. The updated one now compiles and works as expected.
------------------------------------------------------------------------
r2486 | irana | 2017-08-30 09:50:16 +0200 (Wed, 30 Aug 2017) | 3 lines

Renaming MaserBoundary->ConstantMassFlowBoundary in documentation.


------------------------------------------------------------------------
r2485 | irana | 2017-08-30 09:49:39 +0200 (Wed, 30 Aug 2017) | 5 lines

Renaming MaserBoundary->ConstantMassFlowBoundary,
MaserBoundarOldStyle->SubcriticalMaserBoundary, changes
in the USER directory


------------------------------------------------------------------------
r2484 | irana | 2017-08-30 09:49:00 +0200 (Wed, 30 Aug 2017) | 5 lines

Renaming:
MaserBoundary->ConstantMassFlowBoundary
MaserBoundarOldStyle->SubcriticalMaserBoundary
Also changed this in the tests.

------------------------------------------------------------------------
r2483 | irana | 2017-08-29 14:14:18 +0200 (Tue, 29 Aug 2017) | 3 lines

New feature: added back in the old-style maser, which is more suitable for
subcritical flows, as it can cope with the "back-forces".

------------------------------------------------------------------------
r2482 | weinhartt | 2017-08-24 21:55:03 +0200 (Thu, 24 Aug 2017) | 1 line

General cleanup of small bugs/formatting issues
------------------------------------------------------------------------
r2481 | weinhartt | 2017-08-24 21:37:03 +0200 (Thu, 24 Aug 2017) | 1 line

General cleanup of small bugs/formatting issues
------------------------------------------------------------------------
r2480 | weinhartt | 2017-08-24 18:44:22 +0200 (Thu, 24 Aug 2017) | 1 line

Updated the formatting of the CG files
------------------------------------------------------------------------
r2479 | weinhartt | 2017-08-24 17:04:38 +0200 (Thu, 24 Aug 2017) | 1 line

Update SiloBehzad
------------------------------------------------------------------------
r2478 | marnix | 2017-08-24 14:59:28 +0200 (Thu, 24 Aug 2017) | 5 lines

compilation fix: parallel code

Previous update messed up the parallel code compiling.
Apparently I forgot to test it.

------------------------------------------------------------------------
r2477 | marnix | 2017-08-24 14:54:52 +0200 (Thu, 24 Aug 2017) | 7 lines

bug-fix Perdioc boundaries in paralle

Periodic particles are now excluded in functions that
need periodic particles to be excluded.

I.e. computing the fasts particle

------------------------------------------------------------------------
r2476 | marnix | 2017-08-24 13:51:50 +0200 (Thu, 24 Aug 2017) | 5 lines

bug-fix: VTK writer

VTK writer can now correctly cope with periodic boundaries in
parallel build.

------------------------------------------------------------------------
r2475 | marnix | 2017-08-24 12:09:55 +0200 (Thu, 24 Aug 2017) | 10 lines

bug-fix: Parallel code

Major bug fixes to the parallel code.

-Bug fix to interactions in parallel code: Sometimes the interacting particle
was not correctly found. A brute force method is now used to find it
-The new VTK writer could not cope with fake particles, that has been fixed now
-Bug fixes to periodic boundaries in parallel
-Periodic boundaries and mpi boundaries now communicate correctly with each other (hopefully..)

------------------------------------------------------------------------
r2474 | weinhartt | 2017-08-24 12:08:37 +0200 (Thu, 24 Aug 2017) | 1 line

Fixed strange bug in CGHandler
------------------------------------------------------------------------
r2470 | marnix | 2017-08-24 10:59:29 +0200 (Thu, 24 Aug 2017) | 5 lines

clean up: BaseHandler

getObject works with index, so function input name has
changed from id to index

------------------------------------------------------------------------
r2469 | weinhartt | 2017-08-24 08:50:45 +0200 (Thu, 24 Aug 2017) | 1 line

update to TimeAveragedCG and TimeSmoothedCG: added more more file
------------------------------------------------------------------------
r2468 | weinhartt | 2017-08-24 08:36:57 +0200 (Thu, 24 Aug 2017) | 1 line

updated TimeAveragedCG and TimeSmoothedCG: it now has the same outputformat as CG, and fixed a bug in writing TimeSmoothedCG; also added a few small functions in CG to avoid code duplication
------------------------------------------------------------------------
r2467 | weinhartt | 2017-08-23 16:20:52 +0200 (Wed, 23 Aug 2017) | 1 line

Update to SiloBehzadCG: saveCount was not implemented correctly
------------------------------------------------------------------------
r2466 | weinhartt | 2017-08-23 15:59:19 +0200 (Wed, 23 Aug 2017) | 1 line

added SiloBehzadCG.cpp for statistics; this required a bugfix in MindlinRollingTorsionSpecies.cc; a corresponding fix in SpeciesHandler.cc; to read in old files; and a new setN function in BaseCG
------------------------------------------------------------------------
r2465 | weinhartt | 2017-08-23 15:07:40 +0200 (Wed, 23 Aug 2017) | 1 line

Update Silo Behzad: real time is now reported in cout; and you can scale up the problem for testing purposes using the scaleUp variable
------------------------------------------------------------------------
r2464 | weinhartt | 2017-08-23 14:00:27 +0200 (Wed, 23 Aug 2017) | 1 line

Added the matlab script and the correct settings for HorizontalMixer
------------------------------------------------------------------------
r2463 | weinhartt | 2017-08-23 13:34:22 +0200 (Wed, 23 Aug 2017) | 1 line

Paraview output file now works
------------------------------------------------------------------------
r2460 | irana | 2017-08-23 10:56:51 +0200 (Wed, 23 Aug 2017) | 9 lines

Clean-up VTKWriter, MDPM-240:

- fileCounter is now a mutable unsigned int, so that everything can stay const
- added the relevant const's
- BaseHandler::writeVTK is now deprecated, as it is empty and should not be used anymore
- extracted the method writePythonFileForVTKVisualisation, and removed the todo. I did not have any inspiration for a name, so feel free to change it.
- did some clean-up to conform to the coding standards

------------------------------------------------------------------------
r2457 | weinhartt | 2017-08-23 09:04:19 +0200 (Wed, 23 Aug 2017) | 1 line

MDPM240 added finals and overrides where necessary
------------------------------------------------------------------------
r2455 | weinhartt | 2017-08-22 22:32:08 +0200 (Tue, 22 Aug 2017) | 1 line

MDPM 240 fixed small bug
------------------------------------------------------------------------
r2454 | weinhartt | 2017-08-22 21:42:20 +0200 (Tue, 22 Aug 2017) | 1 line

MDPM 240 VTK writers are now in their separate directory
------------------------------------------------------------------------
r2453 | weinhartt | 2017-08-22 21:40:24 +0200 (Tue, 22 Aug 2017) | 1 line

MDPM 240 VTK writers are now in their separate directory
------------------------------------------------------------------------
r2452 | weinhartt | 2017-08-22 21:32:01 +0200 (Tue, 22 Aug 2017) | 1 line

  • No labels