Re: [PATCH] [PR 9514] Fixing parse error for "pointer to a function pointer"

[Keith Seitz <keiths@redhat.com>]

On 10/17/2011 03:59 AM, Abhijit Halder wrote:
>> +FAIL: gdb.base/code-expr.exp: (int ** @code)
>> +FAIL: gdb.base/cvexpr.exp: (int ** const)
>>
> With latest CVS code the patch has no regression:

I still see those as regressions (gdb 7.3.50.20111018-cvs):

$ diff -p testsuite/gdb.sum.0 testsuite/gdb.sum | grep "int \*\*"
   PASS: gdb.base/code-expr.exp: (@code int **)
! PASS: gdb.base/code-expr.exp: (int ** @code)
   PASS: gdb.base/code-expr.exp: (@code int **)
! FAIL: gdb.base/code-expr.exp: (int ** @code)
   PASS: gdb.base/cvexpr.exp: (const int **)
! PASS: gdb.base/cvexpr.exp: (int ** const)
   PASS: gdb.base/cvexpr.exp: (const int **)
! FAIL: gdb.base/cvexpr.exp: (int ** const)

Do I have an outdated version of the patch?

diff --git a/gdb/c-exp.y b/gdb/c-exp.y
index b850179..60004f3 100644
--- a/gdb/c-exp.y
+++ b/gdb/c-exp.y
@@ -944,6 +944,8 @@ abs_decl:   '*'
                         { push_type (tp_pointer); $$ = 0; }
         |       '*' abs_decl
                         { push_type (tp_pointer); $$ = $2; }
+       |       abs_decl '*'
+                       { push_type (tp_pointer); $$ = $1; }
         |       '&'
                         { push_type (tp_reference); $$ = 0; }
         |       '&' abs_decl


Am I missing something else?

Keith