This is the mail archive of the mailing list for the GDB project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: gdb.texinfo patch for -var-list-children (2)

> Date: Sun, 21 Jun 2009 20:18:32 -0700
> From: Joel Brobecker <>
> Cc: Vladimir Prus <>, 	Chris Genly <>,
> > > I suggest you use unified diffs for patches (cvs diff -u). The
> > > default "context" format is some historically-inflicted thing that
> > > is hard to read.
> > 
> > I'm fine with both context and unified diffs.
> That's very kind of you to accept context diffs, but I do feel that
> most reviewers are more comfortable with unified diffs - so I suggest
> we keep asking for unified for future patches. I, for one, cannot
> read context diffs. I usually don't ask a resend, and convert the patch
> from one format to the next, but I like to ask that future patches
> be in unified format.

I'm surprised by such strong feelings, since the difference between
context and unified diffs is not so large.

Anyway, all GNU projects I ever contributed to accept both context and
unified diffs, and in fact so does GDB, because gdb/CONTRIBUTE says

    o	Submitting Patches

	    The patch itself. If you are accessing the CVS repository use
	    "cvs update; cvs diff -cp"; else, use "diff -cp OLD NEW" or
	    "diff -up OLD NEW". If your version of diff does not support
	    these options, then get the latest version of GNU diff.

So if from now on we are going to request only unified diffs, we
should at least change this text.  FWIW, I don't think we should
change this policy because it may prove inconvenient in some
situations (e.g., the Patch utility produces reject files in context
diff format).  But that's me.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]