Boost logo

Boost :

From: David Abrahams (dave_at_[hidden])
Date: 2004-08-05 10:48:32


Stefan Slapeta <stefan_at_[hidden]> writes:

> David Abrahams wrote:
>
>> Well, in principle the patch is OK, but whatever you're doing that's
>> causing that code to be compiled with CW8 is emphatically *NOT* OK.
>> At best it represents a deadly ODR trap.
>> So I'd rather not see you make any changes to detail/iterator.hpp --
>> that part of the code is not meant for conforming compilers.
>>
>
> Sorry, as I thought it was harmless I as I wanted to see the results
> for it in the next run, I commited it before I read your mail. Shall I
> remove it again?

It is basically harmless, so you can leave it there unless it causes
problems.

> What I'll do anyway is to disable this workaround test for all
> compilers that support partial specialization.

I'm still highly suspicious that something is wrong whether or not
the test is enabled. What could cause that code to be compiled by
CW?

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk