Bug for NPP015 (REDEFINE length) partially fixed

[Claes65]

No description provided.

[MarkusAmshove]

In the disabled test, which variable/line do you expect a diagnostic to be raised and why? (e.g. 900 bytes > 800 bytes)
I'm a bit lost on such complex redefines :D

[Claes65]

2 #bytes2 (A1/1:850)
2 redefine #bytes2
3 #bytes-str (A750)
3 #r1 (a1/1:450)

Bytes2 is 850, but the redefinition fields are 750+450

[MarkusAmshove]

It seems to not check the redefine at all, because the top level #GRP is checked.
This is a tough change. Maybe it would be easier to pull out the length check from the addTargetToRedefine (it is never called for redefine #bytes2?) and do a pass over all variable at the end of parseInternal and check if a variable is a redefine and if yes, then check the sizes.

You can use defineData.variables() to do so, no need to manually go through all levels. That method already does return all variables (including tested).

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

92.0% Coverage
0.0% Duplication

[Claes65]

It seems to not check the redefine at all, because the top level #GRP is checked. This is a tough change. Maybe it would be easier to pull out the length check from the addTargetToRedefine (it is never called for redefine #bytes2?) and do a pass over all variable at the end of parseInternal and check if a variable is a redefine and if yes, then check the sizes.

You can use defineData.variables() to do so, no need to manually go through all levels. That method already does return all variables (including tested).

I think for now, we can merge this in, and then look to fix the more fundamental REDEFINE thingy at a later point? I can create an issue for it, if you want me to.