The only nagging worry I have about abort
for this is that if the other abort ends up having a more prominent role in the spec, both being named the same will be confusing.
(I edited this post a million times. Sorry.)
The only nagging worry I have about abort
for this is that if the other abort ends up having a more prominent role in the spec, both being named the same will be confusing.
(I edited this post a million times. Sorry.)
I agree that the overlap with process-level abort
is the main downside to abort!
. I still think dealing with that ambiguity is easier than ambiguity around “failure” (again, since Result
is becoming increasingly common), though.
@brson, what do you think about panic!
or die!
? (To me, I’d expect that latter to be the same as process abort, though…)
We could just also rename process abort, that solves both problems. panic!
is growing on me, though
panic!
is somewhat vague though. It’s not clear what is happening, as opposed to abort!
. Example:
foo.unwrap_or_panic(); // oh, dear!
I dislike all the “cute” names like panic
or die
. I prefer names that describe exactly what is happening without fluff or silly personifications.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.