Re: [resubmit] gdb.base, r*.exp thru w*.exp

[Michael Snyder <msnyder@vmware.com>]

Joel Brobecker wrote:
>> Well, this is a very common and ancient idiom that is used
>> everywhere throughout the test suite.
>>
>> You have to assume that we don't care about anything between
>> the ".*" and the "$gdb_prompt $".  If we do, the test is wrong,
>> but if it consumes more output than it was meant to, the
>> following tests will fail.
> 
> To give more context about the whole effort, I have to say that this was
> one psychologically tough patch to review, because there are lots of
> little things that I would have wanted to improve. But Pedro is right
> that we should limit ourselves to semi-mechanical changes and not get
> carried away by trying to make things perfect.  If the original test was
> already erroneous, we can fix that as a followup patch.  That could be
> another mechanical patch...
> 

Ditto.  For instance, there are lots of places where I could
substitute your new "gdb_test_no_output" routine, but I am
refraining, unless I have to touch the test anyway.

You may note that I am using it for the "sevenbit" business.