Closed ruoruoniao closed 6 months ago
Every time I think I have wrapped my head around idlc
's inside well enough to review PRs and do so maintenance on it, it seems to catch me in yet another bit of weirdness. Who would have thought that this runs into an infinite recursion between alternating between y = x->type_spec
and x = y->parent
... I did not anyway.
That means what seemed like the less surprising solution doesn't work and your original proposal is infinitely better, thanks to the absence of infinite recursion. Can you please revive it? I am sorry to have caused you this much trouble.
Every time I think I have wrapped my head around
idlc
's inside well enough to review PRs and do so maintenance on it, it seems to catch me in yet another bit of weirdness. Who would have thought that this runs into an infinite recursion between alternating betweeny = x->type_spec
andx = y->parent
... I did not anyway.That means what seemed like the less surprising solution doesn't work and your original proposal is infinitely better, thanks to the absence of infinite recursion. Can you please revive it? I am sorry to have caused you this much trouble.
Sure, I am quite surprised when I found the pipeline return error...I don't have a good solution either, unless we can record the path that the function runs on, and terminate it when it encounters the same node, but that's too strange...
Of course ( ̄▽ ̄)" @eboasson
except branch idl_is_type_spec(node)
As you say, change a fix way~ @eboasson