On Thu, Oct 01, 2020 at 03:26:00PM -0400, Konstantin Ryabitsev wrote:

> Hrmph, apparently my assumption on what headers will be generated by 
> policy=8bit was pretty wrong. I pushed a commit to master and 
> stable-0.5.y that should properly set the charset and generate the 
> necessary headers.

This fix seems to have caused some issues for me - using msmtp with
kernel.org on 1594818110-786-1-git-send-email-brent.lu@intel.com I'm
getting headers with:

To: alsa-devel@alsa-project.org, Brent Lu <brent.lu@intel.com>
Cc: =?utf-8?q?linux-kernel=40vger=2Ekernel=2Eorg=2C_Kai_Vehmanen_=3Ckai=2Eve?=
 =?utf-8?q?hmanen=40linux=2Eintel=2Ecom=3E=2C_Pierre-Louis_Bossart_=3Cpierre?=
 =?utf-8?q?-louis=2Ebossart=40linux=2Eintel=2Ecom=3E=2C_Takashi_Iwai_=3Ctiwa?=
 =?utf-8?q?i=40suse=2Ecom=3E=2C_Guennadi_Liakhovetski_=3Cguennadi=2Eliakhove?=
 =?utf-8?q?tski=40linux=2Eintel=2Ecom=3E=2C_Liam_Girdwood_=3Clgirdwood=40gma?=
 =?utf-8?q?il=2Ecom=3E=2C_Amadeusz_S=C5=82awi=C5=84ski_=3Camadeuszx=2Eslawin?=
 =?utf-8?q?ski=40linux=2Eintel=2Ecom=3E?=

(in message ID 160167211500.21762.13369928255363041989.b4-ty@kernel.org
which I'll manage to send) which /usr/sbin/sendmail -t from both msmtp
and exim has a lot of trouble parsing and since b4 has generated the
thanks I now can't ask the database (though I don't immediately see a
way to extract it anyway).  I think we need a sidecar file with the
unencoded e-mail addresses in it?
ask b4 what addresses are buried in there even if