|Anonymous | Login | Signup for a new account||2021-05-11 10:48 UTC|
|Main | My View | View Issues | Change Log | Roadmap|
|View Issue Details|
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0000848||Taste||[All Projects] TASTE-IV/DV||public||2019-03-27 14:25||2019-03-28 09:37|
PI checking for SDL functions possibly broken
After a recent update (not sure which), it now appears to be possible to create PIs for SDL functions where the class of the PI is not sporadic. TASTE IV no longer reports an error.
When the SDL function containing the unprotected PI is edited, OpenGeode operates as normal but behaves as if the unprotected PI does not exist.
Is this intentional? It seems to break the SDL semantics.
|Steps To Reproduce|
Create an SDL function, give it an unprotected PI, edit the SDL function.
|Tags||No tags attached.|
It's my fault, I asked Ellidiss to release this constraint as I planned to give the possibility to expose SDL Procedures as kind of RPC - through sync PIs.
However this is not yet properly implemented, the main issue is related to the code skeletons generation. It is not possible to "declare" the procedures in the part of the SDL function that is re-generated by taste-generate-skeletons (system_structure.pr) and let the user implement them inside the model part (contrary to normal SDL signals, which are declared and connected in one place, and used in another part)
I need to think more about it.
No problem, shall we close this ticket?
No, it is a bug, so let's keep it open until it is properly handled
|2019-03-27 14:25||shd01||New Issue|
new => assigned
|2019-03-27 14:25||shd01||Assigned To||
|2019-03-27 15:32||maxime||Note Added: 0003594|
|2019-03-28 07:42||shd01||Note Added: 0003595|
|2019-03-28 09:37||maxime||Note Added: 0003596|
|2019-03-28 09:37||maxime||Assigned To||
ellidiss => maxime
|Copyright © 2000 - 2011 MantisBT Group|