Re: [PATCH] Add fullname field for MI -break-info command

[Daniel Jacobowitz <drow@false.org>]

On Sat, Jan 28, 2006 at 04:09:42PM +0300, Vladimir Prus wrote:
> Revised patch attached. I did verify that it causes no regressions for MI
> tests.
> 
> 
> Note, however, that I could not verify that it does not cause any
> regressions globally, because something seems wrong with my configuration
> or the test system. Sporadically, I get errors like this:

What system are you testing on?  Presumably a bug in the host or in the
port of expect to the host.

I've verified no regressions on x86_64-pc-linux-gnu.

The patch looks OK.  There's a couple of formatting errors, and
it needs ChangeLog entries, and then it can be applied.

Oh, and Nick was right: you don't need the loc_type check any more.
Everything in that case statement will be a breakpoint and have
a valid address.

> +                struct symtab_and_line sal = find_pc_line (b->loc->address, 0);
> +                char* fullname = symtab_to_fullname (sal.symtab);

char *fullname, please, for consistency.

> +                if (fullname)
> +                  {
> +                    ui_out_field_string (uiout, "fullname", fullname);
> +                  }

Don't need the extra braces here.

> @@ -75,25 +78,27 @@
>      # -break-insert -t srcfile:$line_callee4_head
>      # -break-list
>  
> +
> +
>      mi_gdb_test "222-break-insert -t main" \
> -             "222\\^done,bkpt=\{number=\"1\",type=\"breakpoint\",disp=\"del\",enabled=\"y\",addr=\"$hex\",func=\"main\",file=\".*basics.c\",line=\"$line_main_body\",times=\"0\"\}" \
> +             "222\\^done,bkpt=\{number=\"1\",type=\"breakpoint\",disp=\"del\",enabled=\"y\",addr=\"$hex\",func=\"main\",file=\".*basics.c\",${fullname},line=\"$line_main_body\",times=\"0\"\}" \
>               "break-insert -t operation"

And don't need the extra blank lines.

-- 
Daniel Jacobowitz
CodeSourcery