Re: [jamboost] boost.build v2 testing failure on NT

From: Vladimir Prus (ghost_at_[hidden])
Date: 2002-11-12 02:51:37

David Abrahams wrote:
> Hi Volodya,
> 
> I hope these things look straightforward...
> 
> cd c:/boost/tools/build/test/
> set DO_DIFF=1 && python test_all.py
> --- C:\DOCUME~1\DAVIDA~1\LOCALS~1\Temp\testcmd.36expected	Mon Nov 11 16:51:26 2002
> +++ C:\DOCUME~1\DAVIDA~1\LOCALS~1\Temp\testcmd.37actual	Mon Nov 11 16:51:26 2002
> @@ -10,7 +10,7 @@
> 
> * Project root: project-test1/dir2
> * Parent project: (none)
> -* Requirements: <include>/home/ghost/build/boost-cvs
> +* Requirements: <include>/home/ghost/build\boost-cvs

Evil Jam is to blame :-)

W:\MyDocuments\build\jam_src\bin.ntx86>bjam.exe -f-
p = <include>1/2/3 ;
ECHO $(p) ;
ECHO $(p:G=) ;
^Z
<include>1/2/3
1/2\3

I've fixed that and project_test1.py now pass.

I actually have access to win2000 box (only it's slow, and
free bcc32 is the only compiler). So I went ahead with
tests.

project_test3 failed as well -- I was using "rm" on NT. I've
klu^H^H^Hfixed that as well. There's a problem: variables in
actions are looked up only in global scope, so I had to poke
RM varaible in global scope, which is somewhat ugly.

project_test4 failed too, but the reason is different. I belive
that your changes to error formatting broke it, because it tests
that a certain error message is produced.

I think we should discuss what do do with error messages. If we decide
to use 'errors.error', then we should add an option to omit backtrace,
so that we can check the error message only. I also think that if we
don't come to a conclusion fast enought, the change must be (temporary)
reverted, because developing with failing tests is hard.

- Volodya