On Tue, Jan 23, 2018 at 4:46 PM, Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote:
> On Tue, Jan 23, 2018 at 12:49:22PM -0800, Linus Torvalds wrote:
>>
>> I suspect you're editing your mailbox, or you're using some broken
>> mailer setup, or something (using some broken "export" function?),
>> because you're corrupting Christian's name all the time:
>
> I'm not editing my mailbox, but I am using stgit. The "From:" header
> uses encoded-word syntax for Christian's name, and I think stgit is
> not decoding that when applying the patch.
Ahh. Ok, that would explain it.
> When I save the patch email message from mutt (or gmail, for that
> matter), I see:
>
> From: "=?UTF-8?q?Christian=20K=C3=B6nig?=" ...
Yes. That's the standard locale encoding format for header values
(mainly subject and author), since the
Content-Type: text/plain; charset=utf-8
only applies to the *body* of the email.
So then the email handling tools have to handle that correctly, and as
you noticed, "git am" does so:
> Applying that patch with "git am" results in the correct
>
> Author: Christian König <ckoenig.leichtzumerken@xxxxxxxxx
But apparently stgit does not:
> Applying the identical patch with "stg import -M" results in
>
> Author: =?UTF-8?q?Christian=20K=C3=B6nig?= <ckoenig.leichtzumerken@xxxxxxxxx>
I'm adding Peter Grayson and Catalin Marinas to the participants,
since they seem to be the main stgit developers.
Presumably there is either a flag for this (which should presumably be
the default). Or if not, maybe stgit can just be fixed?
[ I go off looking at the stgit sources. ]
Oddly, stgit seems to even have a *testcase* for this:
- t/t1800-import/email-qp:
From: Inge =?utf-8?q?Str=C3=B6m?= <inge@xxxxxxxxx>
and that test-case harkens back to 2006. I wonder why it doesn't seem
to work for you and Christian.
Bjorn, maybe you can send Catalin an example mbox?
Linus
