[cfe-commits] r56096 - in /cfe/trunk: lib/Sema/SemaStmt.cpp test/SemaCXX/condition.cpp

[Daniel Dunbar]

Argiris is right this convention is being widely used. I tend to agree
with Mike though that if we are going to use non-obvious non-const
reference parameters it would be great to make it explicit in the
name. This has already caused inadvertent regressions.

Personally I often use pointer parameters for output argument in such
situations (i.e. Expr**) to make the code more explicit; that is something
of a matter of taste however.

 - Daniel

On Thu, Sep 11, 2008 at 2:05 PM, Mike Stump <mrs at apple.com> wrote:
> On Sep 11, 2008, at 12:09 PM, Argiris Kirtzidis wrote:
>> CheckCXXBooleanCondition follows the convention of other check
>> functions
>> (CheckInitializerTypes, CheckSingleInitializer, CheckCastTypes, etc)
>
> CheckFoo implies that the routine just Check for the condition Foo (to
> me).  Maybe we should pick a new naming scheme for these types of
> routines.  ConvertFoo at least implies an action.  ConvertAndCheck,
> though, that is a little wordy.
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>