Re: [RFC - Python Scripting] Add 'end' attribute to gdb.Symtab_and_line

[Siva Chandra <sivachandra@google.com>]

On Wed, Jun 13, 2012 at 12:10 AM, Siva Chandra <sivachandra@google.com> wrote:
> I have changed the behavior to [pc, end_pc] as this (at least the
> name) feels more meaningful to me. But, why was right open behavior
> chosen for internal code? Even the two-argument flavor of the
> 'disassemble' command has right open behavior.

Actually, I would like to take this back. I think, wrt 'disassemble'
command, it is essentially disassembling all instructions which fully
fall in the range provided as argument. Wrt symtab_and_line.end, if we
were to expose it as end_pc in the Python API, then I think it should
correspond to the PC value for the last instruction of that line of
source and not the last address in the range. And, if that were to be
the case, I do not think it is very practical as it does not indicate
where the next instruction starts. With the other choice, where in
end_pc indicates the last address in the range, the name end_pc
becomes a misnomer.

From my personal point of view, I see two options:

1. Call this new attribute 'last' and let it indicate the last address
in the range.
   or
2. Call this new attribute 'next_start' and let it indicate the start
PC of the next instruction after the last instruction of the current
source line.

Either of the above is good enough for the use case I have in mind.

Thanks,
Siva Chandra