Re: [PATCH v2 2/5] Further cleanup/modernization of gdb.base/commands.exp

[Pedro Alves <palves@redhat.com>]

On 11/09/2016 04:09 PM, Simon Marchi wrote:
> On 2016-11-09 10:59, Pedro Alves wrote:
>> And I'm wondering whether this below as well would be a good idea,
>> or whether it'd obfuscate?  If a good idea, maybe we'd put the new
>> variable in gdb.exp.
> 
> ...
> 
>> @@ -45,7 +50,7 @@ proc_with_prefix gdbvar_simple_if_test {} {
>>           "else" \
>>           "  p/x 0xdeadbeef" \
>>           "end"] \
>> -    "\\\$\[0-9\]* = 0xdeadbeef" \
>> +    "$valnum_re = 0xdeadbeef" \
>>      "#1"
> 
> 
> That's not bad.  I was going to suggest using {} instead of "" to get
> rid of most backslashes (untested):
> 
> -"\\\$\[0-9\]* = 0xdeadbeef"
> +{\$[0-9]* = 0xdeadbeef}

I strikes me that I can use {}when building the input strings.  I was
already doing it in the new tests added by the following patches, but
not in this one, for some reason.  Let me give that a try.

> 
> but with the variable it looks good as well.
> 

Thanks,
Pedro Alves