[cfe-commits] r76531 - in /cfe/trunk: lib/Analysis/CFG.cpp test/Analysis/dead-stores.c

Ted Kremenek kremenek at apple.com
Tue Jul 21 12:54:10 PDT 2009


On Jul 21, 2009, at 11:32 AM, Eli Friedman wrote:

> On Tue, Jul 21, 2009 at 10:15 AM, Ted Kremenek<kremenek at apple.com>  
> wrote:
>> Right.  Note that we want to keep ChooseExpr in the CFG (and not  
>> just inline
>> its effects) so that we maintain the structure of the control- 
>> flow.  This is
>> important for clients that render diagnostics that incorporate  
>> control-flow
>> (e.g., the static analyzer).
>
> Essentially, you want to treat it as a node in the CFG so that the
> static analyzer can print which side is the evaluated side?  That
> seems like it'll generally just lead to longer paths without providing
> useful information, considering the ways it's normally used.  I don't
> feel strongly about it, though.
>
> -Eli

That's exactly right.  ChooseExpr is used so rarely I don't think it's  
an issue, so I think it actually provides a net improvement in user  
experience to leave it this way.




More information about the cfe-commits mailing list