This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
Re: C++/Java regressions
- From: mec dot gnu at mindspring dot com (Michael Elizabeth Chastain)
- To: mec dot gnu at mindspring dot com
- Cc: gdb at sources dot redhat dot com, ian at wasabisystems dot com
- Date: Tue, 25 Nov 2003 12:06:32 -0500 (EST)
- Subject: Re: C++/Java regressions
[Ian might not need any more of this ...]
I asked:
mec> What is the correct output when a breakpoint is taken on
mec> "A::bar(int) const" ?
After some deliberation, I decided that I want the test script
to be liberal in what it accepts.
class A {
public:
...
int foo (int arg);
int bar (int arg) const;
};
When a breakpoint is taken on A::foo, the script currently accepts:
Breakpoint N, A::foo (this=...)
Breakpoint N, A::foo(int) (this=...)
When a breakpoint is taken on A::bar, the script currently accepts:
Breakpoint N, A::bar (this=...)
Breakpoint N, A::bar(int) const (this=...)
I am going to keep all these patterns and add another pattern to
accept what gdb is printing today:
Breakpoint N, A::bar const (this=...)
If someone wants to spend time on making gdb's output better here,
that is okay with me. But I decided that it's too low priority for
my attention and the list bandwidth.
David C is that okay with you?
Patch to follow.
Michael C