Trying to match to a typedef value in a receive statement causes "bad node type 44" error message

[tgonzalez89]

When I try to match a message in a receive statement I get a "bad node type 44" error message. This happens when the message's type is a typedef. The error message is rather cryptic and doesn't give much insight. I would like to know if this is a bug or the intended behavior. I don't think the documentation mentions anything related to the restriction of using struct types for doing matching in receive statements.

Code example:

typedef t {
    int i
}
init {
    chan c = [1] of {t}
    t x;
    !(c ?? [eval(x)]) // <--- ERROR
}

Link to StackOverflow question here.

[nimble-code]

that's a bug -- i'll check it out and find a fix if i can

[nimble-code]

fixed with update to pangen2.c and pangen4.c
not entirely sure that the fix covers all cases, but it does work for the example code and a few small variations of it

[tgonzalez89]

Thank you. The issue seems to be fixed.

[tgonzalez89]

I played around with the fixed code and in some cases where the struct type is more complex there seems to be some issues. I attached two examples. In complex_id.pml I see the following error: "invalid end state". In test.pml I see this other error: "missing pars in receive". I think both files should pass the simulation. Should I open a new issue or can this one be reused/reopened? I think there might be some issues when there are more than one eval()'s that evaluate a typedef variable in the receive statements and/or when the typedef is composed of other typedefs.

testcases.zip

[nimble-code]

made a further update -- 6 files changed -- that should now cover all cases.
i don't think the invalid endstate is in error, but let me know if that's not the case

[tgonzalez89]

Hmm... I still see the same issues. I simplified test.pml even further. The asserts here should pass but they are failing: pan:1: assertion violated Q_has(c, 1, receiver_id.t2.t1a.i, 1, sender_id1.t2.t1a.i, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
test.txt

[nimble-code]

agreed -- good example -- that'll take some more checking

[nimble-code]

i have a fix -- will update once i've tested it a little more

[nimble-code]

fix uploaded -- 7 files changed -- this now passes the test.txt example and everything else I could think of. fingers crossed....