Re: [Boost-build] Sun toolset

From: K. Noel Belcourt (kbelco_at_[hidden])
Date: 2006-12-02 20:48:37

On Dec 2, 2006, at 6:32 PM, Caleb Epstein wrote:

> On 12/2/06, K. Noel Belcourt <kbelco_at_[hidden]> wrote:
>>  From Sun's CC manpage,
>>
>>                 [no%]tmplife   (Standard mode only) [Do not] Clean
>>                                up the temporary objects that are
>>                                created by an expression at the end
>>                                of the full expression, as defined
>>                                in the ANSI/ISO C++ Standard.
>>                                (When -features=no%tmplife is in
>>                                effect, most temporary objects are
>>                                cleaned up at the end of their
>>                                block.) The default is no%tmplife.
>>
>> Which I read to mean that -features=tmplife leaves temporary files
>> scattered around.
>
> Nothing of the sort.  It has to do with the lifetime of temporaries
> created in expressions like:
>
>    void foo (const std::string& arg) { ... }
>
>    void func () {
>       foo ("bar"); // [1]
>       // more stuff...
>   } // [2]
>
> With Sun CC's default (-no%tmplife), the unnamed temporary std::string
> created for the call to foo would not be destroyed until the enclosing
> scope exits ([2] above).  A conforming compiler (and Sun CC with
> -tmplife) will destroy the temporary at [1].  In the case of
> std::string this side-effect is harmless, but for more "interesting"
> types the default behavior could be disastrous.

Oh, cool.  Thanks for the explanation (I've never used the option as  
you can tell).

-- Noel