On 18/09/10 05:33, Alex Rousskov wrote:
> On 09/17/2010 01:19 AM, Amos Jeffries wrote:
>> see IRC discussions 2010-09-17 for the reasons.
>>
>> Boils down to it not actually being needed and causing semantic problems
>> when provided by SBuf().
>
> I would not remove terminate() completely because it does provide needed
> functionality. I would just make it private, document non-persistency
> caveat, and call it from c_str() and other applicable methods. If it is
> only used once, this becomes less important, of course, but I would
> still keep the code isolated.
The way I look at it exportTRef()** as it exists today can be used
internally to implement terminate(). This patch alters the (one) use of
terminate to do exactly that.
Whether the post-transition code keeps exportTRef()** privately for
this purpose is outside the scope.
** current name used for clarity.
Amos
-- Please be using Current Stable Squid 2.7.STABLE9 or 3.1.8 Beta testers wanted for 3.2.0.2Received on Sat Sep 18 2010 - 02:20:39 MDT
This archive was generated by hypermail 2.2.0 : Sat Sep 18 2010 - 12:00:06 MDT