Fix unused variables that only appear in assert
#1036
Conversation
There was a problem hiding this comment.
Still needs formatting fixes, and #1037 should help so that the formatting guidance is provided in the GitHub action log.
|
I'll wait for #1037 to be merged and then will rebase this. |
This replaces instances of the form ``` x = f(..); assert(x); ``` with ``` x = f(..); INVARIANT(x, "..."); ```
mgudemann
force-pushed
the
fix/unused-vars-in-assert
branch
from
c8f7102 to
675461f
Compare
mgudemann
force-pushed
the
fix/unused-vars-in-assert
branch
from
bbff9ef to
fccec66
Compare
mgudemann
force-pushed
the
fix/unused-vars-in-assert
branch
from
24951de to
e851938
Compare
src/ic3/aux_types.hh
Outdated
| @@ -6,8 +6,14 @@ Author: Eugene Goldberg, eu.goldberg@gmail.com | |||
|
|
|||
| ******************************************************/ | |||
|
|
|||
| #pragma once | |||
There was a problem hiding this comment.
We don't usually use this in this code base. Why is it necessary here?
There was a problem hiding this comment.
Not sure it is necessary, I am not too familiar with this code to be honest.
Is it better to replace this with
#ifndef ...
#define ..
...
#endif
?
Or omit altogether if it compiles without such a guard?
There was a problem hiding this comment.
None of those ic3 headers appear to use a header guard, so omit altogether if it compiles, I'd say.
There was a problem hiding this comment.
it seems that only dnf_io.hh requires a guard. The PR is update accordingly.
This fixes instances in the IC3 implementation where variables are only used for
assert, leading to errors when compiling withg++13.3.0.This also removes
assert(false)in two instances in the IC3 implementation.