Problems submitting patch containing UTF-8 characters

Michael Gmelin freebsd at
Mon Oct 1 11:23:24 UTC 2012

On Sun, 30 Sep 2012 16:33:15 +0200
Michael Gmelin <freebsd at> wrote:

> On 30 Sep 2012, at 16:19, Shaun Amott <shaun at> wrote:
> > On Sun, Sep 30, 2012 at 05:08:03AM +0200, Michael Gmelin wrote:
> >> I recently ran into a problem submitting a PR containing UTF-8
> >> characters, they ended up garbled, so the maintainer couldn't
> >> apply the patch cleanly.
> > 
> > GNATS, unfortunately, records no information about character
> > encoding. To make matters worse, it actually removes many headers
> > from e-mail replies, meaning query-pr.cgi can usually only work on
> > guesses and assumptions.
> > 
> > Headers are however preserved inside MIME parts, and if there are
> > any of these headers, they are used; but in this case your MUA
> > hasn't included encoding information here.
> > 
> > Shaun
> > 
> > -- 
> > Shaun Amott // PGP: 0x6B387A9A
> > "A foolish consistency is the hobgoblin
> > of little minds." - Ralph Waldo Emerson
> ...

I tries setting the character encoding of the attachment, which fixes
the problem only for viewing the patch on the PR website (so one could
copy and paste the patch). The download links will still deliver
a corrupted patch though, so basically it is impossible to submit a
patch containing UTF-8 characters that can be downloaded from the PR
website correctly*.

I created a bug describing/demonstrating the issue, maybe a member of
portmgr or www could check this.


* unless one encodes it into a special ASCII format
like base64 and attaches this as text/plain us ascii and of course
tells people how to use it - this is neither practical nor desirable
since it makes it impossible to discuss a patch by just looking at it
on the website.

Michael Gmelin

More information about the freebsd-ports mailing list