Open
Description
This issues is to keep track of the features/bugs that needs to be fixed in MAPDL. They might be tracked somewhere else as well, but this issue should raise visibility among PyMAPDL users.
- PRIORITY [BUG] gRPC stability issues #3313
- [BUG] Parsing issues when MAPDL logger is printing to console. See CICD tests get stuck very frequently #3591
- [BUG] v251 database module crashes #3248
- [BUG] PyMAPDL performance decrease on MAPDL and on v222 onwards #3208
- [FEAT] Avoid
mapdl.amesh
inmapdl.aplot
#3613 - [FEAT] Release port before exiting Very variable time for launch_mapdl() #2564
- [FEAT] Add
mapdl.variables
for variables. #2021 - [FEAT] API method to transfer arrays to the MAPDL instance #1038
- [BUG] Array parameters retrieval fail when the size is very big #2844
- [FEAT] Attach a session ID in each gRPC message. See Performance decline during loops after update to newer version of apdl-mapdl-core #2608
- [BUG] Fix reading geometry files when using gRPC: When I want to import an IGES file and plot it, MAPDL server connection terminated #2948
- [BUG] Fix MPREAD MPREAD and gRPC error when using LIB option...bug or feature? #3333 (reply in thread)
- [BUG] Mesh.grid seems to result in losing the gRPC connection to MAPDL aaS when a non-trivial number of mixed uP elements are present. #2933
- [BUG] Unexpected MAPDL server disconnection while running a series of
aplot()
#1419 - [FEAT] Transfer areas using the gRPC, so we do not need to re-mesh. Avoid
mapdl.amesh
inmapdl.aplot
#3613 - [FEAT] Implement hash in the entities, so it changes when new geometry is added, or the existing is changed.
- [FEAT] Hashing database so we can check first if there is changes in it or not before retriving mesh or data.
- [FEAT] Lines transfering should not depend on the IGES format, since it can generate slighly diferent plotting. #3811
- [FEAT] Retrieve entities IDs, and selected components using gRPC methods. Related to Error parsing components when there are many #3652
Ideally new features should couple with the MAPDL commands, so we do not duplicate code. Meaning, we are aiming to a code refactor more than coding new stuff.