Re: [jamboost] Two problems with msvc

From: Vladimir Prus (ghost_at_[hidden])
Date: 2003-03-07 02:13:43

David Abrahams wrote:

> Hmm, looks like it's back in a different form:
>
> python test_all.py msvc-6.5

> path_features : PASSED
> relative_sources : PASSED
> no_type : PASSED
> default_build : 
> "c:\boost\tools\build\test\..\jam_src\bin.ntx86\bjam
> -sBOOST_BUILD_PATH=c:\boost\tools\build\test\..\new;c:\boost\tools\build\te
>st -d0 --quiet " returned 1 STDOUT ============
> ...skipped <./msvc/debug>a.rsp for lack of <./msvc/debug>a.obj...
> ...skipped <./msvc/debug>a.exe for lack of <./msvc/debug>a.rsp...
> ...skipped <./msvc/release>a.rsp for lack of <./msvc/release>a.obj...
> ...skipped <./msvc/release>a.exe for lack of <./msvc/release>a.rsp...
> ...failed updating 2 targets...
>
> STDERR ============
> 'cl' is not recognized as an internal or external command,
> operable program or batch file.

It's a different problem. Now, "default-build" works in a following way:
if user gives *any* build request explicitly, the entire default build is
ignored. In order to do any testing of default build, the test
disables passing toolset name to jam. So "msvc-6.5" on the command
line has no effect. 

I'm planning to start hacking on better default properties now: so that
"msvc-6.5" on the command line will be combined with "debug release" in
default-properties. In the meantime, I guess the test can be skipped on msvc.

- Volodya