Re: [RFC] Fix the MI result of -break-insert with multiple locations

[Vladimir Prus <vladimir@codesourcery.com>]

On 29.01.2013 20:04, Marc Khouzam wrote:

>> (gdb)
>> -break-insert foobar
>> ^done,bkpt={number="1",type="breakpoint",disp="keep",enabled="
>> y",addr="<MULTIPLE>",times="0",original-location="foobar"},{nu
> mber="1.1",enabled="y",addr="0x00000000004008d7",func="foobar()",file="overload.cpp",fullname=".../src/test-bktp-mi-> overload/overload.cpp",line="22",thread-groups=["i1"]},{number
> ="1.2",enabled="y",addr="0x00000000004008e0",func="foobar(int)",file="overload.cpp",fullname=".../src/test-bktp-mi-> overload/overload.cpp",line="26",thread-groups=["i1"]}
>> (gdb)
>>
>> So no square brackets. Therefore not a list and IMHO not a
>> valid MI output.
>
> You're right.  The output I took was from -break-list, but
> for -break-insert, and probably some others, this does not
> look to be valid syntax.
>
> I like your proposal of
>    bkpt={number="1", ... , locations=[{number="1.1", ...}, ...]}
>
> which would allow to easily differentiate between the "main bp
> entry" and the "sub entries".

I think it would help to see complete output, without ellipsis.

(And when we agree that the proposed output is good, the patch would need testsuite changes to be accepted)

Thanks,

-- 
Vladimir Prus
CodeSourcery / Mentor Graphics
http://www.mentor.com/embedded-software/