|
Boost : |
From: Chris Little (cslittle_at_[hidden])
Date: 2002-02-23 22:09:36
on 2/23/02 1:46 PM, Samuel Krempp at krempp_at_[hidden] wrote:
> On Sat, 2002-02-23 at 17:48, Chris Little wrote:
>> I tried this new version on Metrowerks CWP 7.2 and I am seeing some warnings
>> and two errors.
>
> great, it's one more compiler.
>
>> I am getting a warning about a possible unwanted ;. Could you remove the
>> empty else statement and just have a comment instead.
>>
>> I'm also seeing a number of unused argument warnings due to the functions in
>> group.hpp on lines 45 and 79.
>
> okay, fixed those.
>
>
>> format_test1.cpp compiles and runs fine.
>> format_test2.cpp compiles but fails two tests.
>>
>> The test on line 80 fails.
>>
>> Expected: "__16/9"
>> Received: "____16/9"
>
> hum, I think that was a bug in the version '1.40' (that I thought final)
> I had managed to get rid of the MSVC overload problem, but I realised
> later that groups were then simply processed as any argument, and this
> line is exactly meant to test against that.
>
> Was those results obtained with this version ?
> if so, I'm aware of that, and it is fixed at the same time as the MSVC
> overload trouble.
> I'm experimenting a bit on the solution proposed by Aleksey, and will
> upload my updated code shortly.
>
>
>> The test on line 109 fails.
>> Expected: "[ 123 ] [+WW234] [-XX345] [YY 456] [ZZZ-10]\n"
>> Received: "[ 123 ] [WW+234] [XX-345] [YY 456] [ZZZ-10]\n"
>
> It certainly is the same cause (groups not being processed correctly)
>
>
I was building with the 1.40 version. I saw in a later version that you
have posted a 1.42. I'll download it and confirm the fix.
Chris
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk