Re: [RFA] 12843

[Daniel Jacobowitz <drow@false.org>]

On Mon, Aug 29, 2011 at 5:12 PM, Keith Seitz <keiths@redhat.com> wrote:
> On 08/29/2011 12:46 PM, Tom Tromey wrote:
>
>> My first reaction is against this, but I don't have a particularly good
>> explanation for that.  I will think about it.
>
> Is there actually any reason to require a flag? Couldn't we simply allow
> either a flag-based location _or_ a linespec, but not both, i.e.,
>
> -insert-break -file foo.c -function my_function
>
> OR
>
> -insert-break foo.c:my_function
>
> but not
>
> -insert-break -file foo.c my_function
>
> This would allow backward compatibility and new/better functionality.
>
> I wouldn't be surprised if IDEs adopted the flag-based version very quickly.

Heck, sometimes I'd use this from the command line if I could.

-- 
Thanks,
Daniel