This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch][python] 2 of 5 - Frame filter MI code changes.
- From: Phil Muldoon <pmuldoon at redhat dot com>
- To: Tom Tromey <tromey at redhat dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>
- Date: Tue, 07 May 2013 09:23:46 +0100
- Subject: Re: [patch][python] 2 of 5 - Frame filter MI code changes.
- References: <51876882 dot 3010301 at redhat dot com> <87ppx3x2k3 dot fsf at fleche dot redhat dot com>
On 06/05/13 21:42, Tom Tromey wrote:
>>>>>> "Phil" == Phil Muldoon <pmuldoon@redhat.com> writes:
>
> Phil> @@ -52,14 +78,43 @@ mi_cmd_stack_list_frames (char *command, char **argv, int argc)
>
> Phil> + if ((argc > 3) || (argc == 2 && oind) || (argc == 1 && ! oind))
> Phil> + error (_("-stack-list-frames: Usage: [--no-frame-filters] [FRAME_LOW FRAME_HIGH]"));
>
> Phil> - if (argc == 2)
> Phil> + if (argc == 3 || argc == 2)
>
> It seems to me that these checks should use oind rather than argc.
> Then they would be simpler.
>
> As it is I think they are wrong for the weird but accepted input:
>
> -stack-list-frames --no-frame-filters --no-frame-filters 0 15
I'm not sure how I would write them just using oind. oind will be an
integer index to the last option in the argv list parsed.
-stack-list-frames --no-frame-filters 0 15
mi_getopt stops parsing when it reaches the first non-option argument.
So in this case it would be 1. argc is the total number of arguments,
which is in that case is 3.
Your example:
-stack-list-frames --no-frame-filters --no-frame-filters 0 15
Would fail on the first "if" condition check:
if ((argc > 3) || (argc == 2 && oind) || (argc == 1 && ! oind))
As argc is > 3. Typically from what I have found all MI commands do
an argc check for the accepted number of arguments. Because MI relies
on integer arguments for ranges (i.e. FRAME LOW - FRAME HIGH), and also
some MI commands allow integer substitutions for some options (i.e., 1
or --all-values), and mi_getopt stops parsing on the first non-option
it encounters, MI arguments are inherently limited to strict
positional adherence.
Anyway, that "if" statement just checks for permutations of
-stack-list-frames --no-frame-filters 1 2 3 4
-stack-list-frames --no-frame-filters 1
-stack-list-frames 1
It looks like I am missing a check for:
if (argc == 3 && ! oind).
I will add that.
If there is a way to write the test solely using oind I will happily
do it, but I must admit to being puzzled how to do it.
Cheers
Phil