Re: stripping [PATCH] without losing later tags from mailed patches (Re: [ 02/12] Remove COMPAT_IA32 support)

From: Jonathan Nieder <jrnieder@gmail.com>
Date: Mar 12, 2012, 16:57 UTC
Message-ID: <20120312165703.GB18791@burratino>
In-Reply-To: <87aa3l4vqq.fsf@thomas.inf.ethz.ch>

Thomas Rast wrote:

Show 8 quoted lines

> The problem with -b is that it's a backwards-compatibility shorthand for
> --binary, which used to pass --allow-binary-replacement (or --binary) to
> git-apply.  However, that option was obsoleted in 2b6eef9 (Make apply
> --binary a no-op., 2006-09-06) and has been a no-op for over 5 years.
> It has also not been documented since cb3a160 (git-am: ignore --binary
> option, 2008-08-09).
>
> So perhaps we can safely claim -b for --keep-non-patch, like so:

Thanks.

It we want to be extra friendly to people who have been using "format-patch --binary" with "am -b" in their scripts, we could have a transitional period during which -b is treated as a usage error.

Luckily, a quick Google code search does not reveal any users for "am -b", so I am not too worried and would not mind your patch that just switches over right away, though. After all, the failure modes are:

 - if my current script using "am -b" gets run using ancient git, it
   will accept binary patches and will strip out too many brackets
   in the subject line

 - if my ancient script using "am -b" gets run using current git, it
   will helpefully keep [IA64] brackets in the subject line

Neither seems terribly painful.

The manual would need to mention that this once meant --binary to avoid confusion when that happens.

Hope that helps, Jonathan