On Mon, 2013-02-18 at 15:04 -0500, David Miller wrote:
> From: Ben Hutchings <[email protected]>
> Date: Sun, 17 Feb 2013 19:08:42 +0000
> 
> > On Tue, 2013-02-12 at 12:50 -0500, David Miller wrote:
> >> [ Ian, there is a note for you below, I need your help. ]
> >> 
> >> Please queue up the following networking bug fixes for 3.0.x,
> >> 3.2.x, 3.4.x, and 3.7.x respectively.  Thanks!
> >> 
> >> Ian, I was able to backport the xen-netback CVE fixes for every
> >> release except 3.0.x.  But 3.0.x was beyond my capabilities.  Could
> >> you please do a 3.0.x backport of those 4 patches and submit them to
> >> -stable for me?  Thanks!
> > 
> > Added to the queue for 3.2.
> > 
> > By the way, for commit messages with non-ASCII characters the headers in
> > your patch mboxes say:
> > 
> > Content-Type: text/plain; charset=UTF-8
> > Content-Transfer-Encoding: 8bit
> > 
> > However the commit messages are actually encoded as ISO 8859-1 (and the
> > MIME headers for the attachment agree with that).  Could you try to make
> > them consistent, please?
> 
> It's what emacs MEW auto-detects the content as, does it actually cause
> a problem?

It seems to be converting, not auto-detecting.  And it doesn't do
anything about the headers inside the mbox.  So if I don't fix these up
then git will warn, for example:

Applying: tcp: fix for zero packets_in_flight was too broad
Warning: commit message does not conform to UTF-8.
You may want to amend it after fixing the message, or set the config
variable i18n.commitencoding to the encoding your project uses.

and the commit message has accented characters corrupted.

> I bet you just save the attachment and no changes happen to the
> content whatsoever.

Indeed: the text is not (AFAIK) converted at my end, so I get an mbox
file where each message says it's in UTF-8 but it's actually in ISO
8859-1.

Ben.

-- 
Ben Hutchings
Sturgeon's Law: Ninety percent of everything is crap.

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to