|
|
Boost Testing : |
From: Aleksey Gurtovoy (agurtovoy_at_[hidden])
Date: 2005-12-16 04:20:25
D. Kevin McGrath writes:
> On Dec 5, 2005, at 10:50 PM, Aleksey Gurtovoy wrote:
>
>> D. Kevin McGrath writes:
>>> Maybe I missed this somewhere, but can anyone tell me why compiler-
>>> status is built using the pjl-toolset, and not the main toolset?
>>
>> That's an artifact of both tools sharing a single Jamfile
>> ($BOOST_ROOT/tools/regression/build/Jamfile). Technically,
>> 'regression.py' doesn't even need 'compiler_status' executable.
>>
>>> That just seems a little backwards to me.
>>
>> Why so? It's a regression tool, not a library.
>>
>
> It seems backwards since the output from compiler_status would then
> be using the pjl-toolset, rather than the main regression toolset.
> Or am I still confused as to how the executable is used?
Most likely; like I've said in the quoted message, 'regression.py'
chain doesn't use 'compiler_status' executable at all. It's being
built only because it shares its Jamfile with 'process_jam_log'.
Hope this clarifies things,
-- Aleksey Gurtovoy MetaCommunications Engineering