Re: [boost] Boost regression notification (2005-10-25 [RC_1_33_0]) [comeau]

From: Robert Ramey (ramey_at_[hidden])
Date: 2005-10-25 14:16:47

I have made a tweak to the serialiation Jamfile in an attempt to address 
something that
prevents tests from being run on comeau.  This causes a command line error 
wiht the
como "pre-linker".  Anyway, it didn't work out so now I've removed the 
"fix".  I am having
a lot of problems with the comeau setup.  This is really irksome to me as I 
really want to test with it.

a) I the como "prelinker" complains about "c:\Program Files"... it doesn't 
like the space.  I've set all my como paths to c:\\progra~1 .. but still I 
get this.

b) on the test matrix, "prelinkage" is failing because there are identical 
template instantiations in both the boost test library as well as the boost 
serialization libraries.  I don't know how to get around this.

Robert Ramey

Caleb Epstein wrote:
> On 10/25/05, Douglas Gregor <dgregor_at_[hidden]> wrote:
>>
>> Boost regression test failures
>>> serialization|
>> test_class_info_load_binary_archive: cw-9_4 cw-9_5-darwin
>> gcc-3_4_3-sunos mingw-3_4_2
>
>
> http://tinyurl.com/84b82
>
> This is a results page for my gcc-3_4_3-sunos run, but is saying:
>
> See
> bin/boost/libs/serialization/test/test_class_info_save_text_archive.test/sunpro-5_3-sunos/debug/debug-symbols-off<http://engineering.meta-comm.com/boost-regression/CVS-RC_1_33_0/output/CalebEpstein-bin-boost-libs-serialization-test-test_class_info_save_text_archive-test-sunpro-5_3-sunos-debug-debug-symbols-off.html>
>
> Which (aside from not exisiting) is a result page for a different
> toolset that I also happen to test for. Any thoughts about what might
> be wrong here? Is something wrong with the result collection, or
> perhaps the Jamfile re-writes that have been going on?