Robert Treat wrote:
> On Thursday 16 February 2006 00:27, Tom Lane wrote:
> > Robert Treat <xzilla@users.sourceforge.net> writes:
> > > ! <li>The patch should be generated in contextual diff format and
> > > should ! be applicable from the root directory. If you are unfamiliar
> > > with ! this, you might find the script
> > > <I>src/tools/makediff/difforig</I> ! useful. Unified diffs are only
> > > preferrable if the file changes are ! single-line changes and do not
> > > rely on the surrounding lines.</li>
> >
> > I'd like the policy to be "contextual diffs are preferred", full stop.
> > Unidiffs are more compact but they sacrifice readability of the patch
> > (IMHO anyway) and when you are preparing a patch you should be thinking
> > first in terms of making it readable for the reviewers/committers.
> >
> > Some things that follow along with the readability mandate, and should
> > be brought out somewhere here:
> > * avoid unnecessary whitespace changes. They just distract the
> > reviewer, and your formatting changes will probably not survive
> > the next pgindent run anyway.
>
> would diff -c --ignore-space-change be better?
No, because some whitespace changes are important. For example when you
indent a piece of code one level higher. The submitter should eyeball
the patch (in diff form) and clean things up when something unexpected
appears, like a no-op whitespace change.
--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.