Re: [PATCH] MI: Add value field to output of -var-create

[Daniel Jacobowitz <drow@false.org>]

On Thu, Jan 18, 2007 at 10:08:39PM +1300, Nick Roberts wrote:
Content-Description: message body and .signature
>  > > This patch just adds the value field to the output of var-create so that
>  > > the frontend doesn't need to issue another command to find out what it
>  > > is.
>  > > 
>  > > If approved, I'll change the tests and documentation accordingly.
>  > 
>  > Seems fine to me, after we work out what to do with the testsuite. I
>  > went to check whether this should be PRINT_ALL_VALUES or
>  > PRINT_SIMPLE_VALUES but it seems fine the way you've got it - I wish
>  > I could remember the rational for PRINT_SIMPLE_VALUES properly.
> 
> OK, I've done this now with all tests changed.  I've used Vladimir's
> mi_create_varobj as much as possible.  The three places where I've kept
> -var-create are:
> 
> 1) Where failure of -var-create is being tested.
> 2) Where a command token is used.
> 3) Where a name isn't specified for the variable object ("-" used).
> 
> I get the same test results before and after my change.

Thanks for doing this.  I get some new failures - just three though.

FAIL: gdb.mi/gdb701.exp: create fooPtr
FAIL: gdb.mi/gdb792.exp: create var for class A
FAIL: gdb.mi/gdb792.exp: create var for class C which has baseclass A

  # Why does this have a FIXME?
  setup_xfail *-*-*
  mi_gdb_test "-var-update *" \
! 	"FIXME\\^done,changelist=\\\[\{name=\"psnp->ptrs.0.next.long_ptr\",in_scope=\pe_changed=\"false\"\}\\\]" \
        "update all vars psnp->next->long_ptr (and 1.long_ptr) changed"
  clear_xfail *-*-*

Typo, I think ("in_scope=\pe_changed")?

-- 
Daniel Jacobowitz
CodeSourcery