Re: [HEAD] Teach testsuite/lib/objc.exp about Darwin frameworks

[Ziemowit Laski <zlaski at apple dot com>]

On 3 Jun, 2004, at 4.42, Nicola Pero wrote:

> > > > Could you please make the linking against Foundation conditional 
> > > > for the
> > > > Apple runtime?

Here goes -- does this look OK?  I'd like to commit this, and then 
later,
once we figure out how to distinguish Mac OS X from Darwin, we can 
conditionally
use David's NSConstantString @implementation.

Thanks,

--Zem

[gcc/testsuite/ChangeLog]
2004-06-03  Ziemowit Laski  <zlaski@apple.com>

        * lib/objc.exp (objc_target_compile): When running tests
        on Mac OS X, do not pass down '-framework Foundation' if
        the GNU runtime is being used.

Index: gcc/testsuite/lib/objc.exp
===================================================================
RCS file: /cvs/gcc/gcc/gcc/testsuite/lib/objc.exp,v
retrieving revision 1.22
diff -c -3 -p -r1.22 objc.exp
*** gcc/testsuite/lib/objc.exp  2 Jun 2004 19:48:49 -0000       1.22
--- gcc/testsuite/lib/objc.exp  3 Jun 2004 21:47:34 -0000
*************** proc objc_target_compile { source dest t
*** 184,193 ****
        append ld_library_path ":${libobjc_dir}"
      }

!     # If we are running on Darwin, we'll need to point the linker at
!     # the Foundation framework, where many goodies (e.g., 
NSConstantString)
!     # reside.
!     if { [string match "*-*-darwin*" $target_triplet] } {
        lappend options "libs=-framework Foundation"
      }

--- 184,193 ----
        append ld_library_path ":${libobjc_dir}"
      }

!     # If we are running on Mac OS X and using the NeXT runtime, we'll 
need
!     # to point the linker at the Foundation framework, where many 
goodies
!     # (e.g., NSConstantString) # reside.
!     elseif { [string match "*-*-darwin*" $target_triplet] } {
        lappend options "libs=-framework Foundation"
      }



> > > Even this wouldn't be entirely satisfactory, as there are people 
> > > trying to
> > > compile and use gnustep-base with the Apple runtime, in which cases 
> > > you'd
> > > need the Apple runtime + linking against gnustep-base.
> > >
> > > I suppose the solution would be to #include a trivial implementation 
> > > of
> > > NSConstantString in the tests when compiled with the Apple runtime.  
> > > That
> > > way you don't depend on anything external and it just works, and you 
> > > can
> > > test with multiple library combinations.
> > >
> > > If you want to do this (I don't have an Apple runtime at hand), it 
> > > would
> > > be good - should be quite easy (and little bit of fun) to write a 
> > > trivial
> > > NSConstantString implementation.
> > >
> > > Else, I don't think we should be too picky and prevent improvements 
> > > - I
> > > think Zem's change is an improvement over what we have, so it's a 
> > > good
> > > change -- I'd just suggest adding comments to it about how to make it
> > > better, so that if there is someone interested in that stuff (eg 
> > > running
> > > the tests on Darwin without Foundation) in the future, they'll know 
> > > from
> > > the comments how/what to do to extend the tests to support that.
> >
> > :-), the minimal NSConstantString implementation was the first thing 
> > on
> > my mind also.  But I wasn't looking for perfection.  The problem I 
> > have,
> > is that on non OS X Darwin with the GNU runtime these test should pass
> > and now would fail.
> >
> > It would be nice if the gcc testsuite could be testing the Apple 
> > runtime
> > in the absence of Foundation, but that hasn't worked before either 
> > which
> > is why I didn't ask for the minimal NSConstantString implementation.
> > Yet if there is interest, I could hack up patch for test cases which
> > should make linking against Foundation obsolete.
>
> I personally think this (patches providing the minimal NSConstantString
> implementation to make linking against foundation libraries obsolete)
> would be best and if you had the time to produce such patches that 
> would
> be fabulous.
>
> Otherwise, adding -framework Foundation, but only if the Apple runtime 
> is
> used, and a short comment saying a minimal NSConstantString 
> implementation
> would be the real fix rather than that hack, would be my second choice 
> (I
> suppose if you supply a patch for that to Zem he'd like it).
>
> Thanks for looking into this.
--------------------------------------------------------------
Ziemowit Laski                 1 Infinite Loop, MS 301-2K
Mac OS X Compiler Group        Cupertino, CA USA  95014-2083
Apple Computer, Inc.           +1.408.974.6229  Fax .5477