Re: [RFC] TI msp430 architecture support

[Stan Shebs <stanshebs@earthlink.net>]

On 6/25/13 8:51 AM, Kevin Buettner wrote:
> On Tue, 25 Jun 2013 08:25:20 -0600
> Tom Tromey <tromey@redhat.com> wrote:
> 
>>>>>>> "Kevin" == Kevin Buettner <kevinb@redhat.com> writes:
>>
>> Kevin> I fixed the formatting issues that you noticed and then, for
>> Kevin> good measure, I ran the file through gdb_indent.sh.  I should've
>> Kevin> run gdb_indent.sh over msp430-tdep.c prior to submitting it.
>>
>> I didn't know anybody used gdb_indent.sh.  Was this a one-off thing or
>> have you run it regularly on other source files?
> 
> There was a time when the output of GNU indent was considered the
> standard for GDB indentation.  If there was any disagreement about
> correct indentation, a contributor would be asked to run GNU indent
> over the code.  The problem is, of course, that different options
> produce different results which is how gdb_indent.sh came into being.
> 
> My recollection is that there were at least two occasions when
> significant portions of the gdb source tree were reindented.  (This
> may have pre-dated the creation of gdb_indent.sh, however.) I don't
> think that mass reindentation should be done very often because it
> causes a lot more work for folks who have developed significant
> changes prior to the the reindenation occurring.  I think it can make
> sense for newly contributed files, however.
> 
> Kevin
> 

In the past when I've run GNU indent on GDB, its results were
reasonable, but subtly different from Emacs region indents; so while
indent was a good tool to clean up messes, Emacs users were going to
find code formatting changing around them unexpectedly.

The script came later, I don't know if its options fix that problem.
(I never discovered a combination of options that exactly matched
Emacs behavior.)

Stan
stan@codesourcery.com