Re: ncurses 6.6 regression: setupterm() rejects valid PTY, breaks tmux attach

--0000000000006ca93a0652c928c3
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

In addendum, I'd like to remind the list that if this is somehow my fault,
in a patch I have completely forgotten, I'm in an airport without a laptop,
and have no easy way to verify who is at fault here.

Further discussion will have to include Ethan Hunt.

Johann

 I'm not from the internet, I just work there.

On Wed, May 27, 2026, 5:14=E2=80=AFPM Johann 'Myrkraverk' Oskarsson <
[email protected]> wrote:

> Please excuse the bullshit, we are having an ncurses episode.
>
>
> So which Single Unix Specification is at fault, and which email list
> should we include in this discussion?  I can add myself as
> [email protected], but that's just an alias for this email, and I
> have never worked on the Single Unix Specification myself.
>
>
>
> Johann
>
>  I'm not from the internet, I just work there.
>
> On Wed, May 27, 2026, 5:03=E2=80=AFPM Liam Huang <[email protected]> wrote:
>
>> Hi Johann,
>>
>> Just a quick clarification =E2=80=94 the imsg library I'm referring to i=
s the
>> OpenBSD IPC library (imsg_init, imsg_compose, etc.), not the macOS
>> iMessage CLI tool at github.com/openclaw/imsg. They share a name but
>> are completely unrelated projects.
>>
>> The OpenBSD imsg library provides structured message passing over Unix
>> domain sockets, and is used by many OpenBSD daemons (OpenSMTPD,
>> OpenBGPD, etc.). tmux bundles a compat copy of it in its source tree
>> (compat/imsg.c, compat/imsg-buffer.c) so it can build on non-OpenBSD
>> systems like Linux and macOS.
>>
>> This is not a Linux-vs-SUS issue. It's simply that tmux upgraded its
>> bundled copy of this library between releases 3.5a and 3.6b, and the
>> new version changed the wire format in a way that's incompatible with
>> the old one. When an old server and new client try to communicate, the
>> fd-passing mechanism breaks.
>>
>> Best,
>> Liam
>>
>>
>> 2026=E5=B9=B45=E6=9C=8827=E6=97=A5 16:59=EF=BC=8CJohann 'Myrkraverk' Osk=
arsson <[email protected]>
>> =E5=86=99=E9=81=93=EF=BC=9A
>>
>> Johann
>>
>>
>>

--0000000000006ca93a0652c928c3
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div>In addendum, I&#39;d like to remind the list that if=
 this is somehow my fault, in a patch I have completely forgotten, I&#39;m =
in an airport without a laptop, and have no easy way to verify who is at fa=
ult here.</div><div dir=3D"auto"><br></div><div dir=3D"auto">Further discus=
sion will have to include Ethan Hunt.</div><div><br></div><div data-smartma=
il=3D"gmail_signature">Johann<br><br> =C2=A0I&#39;m not from the internet, =
I just work there.</div></div><br><div class=3D"gmail_quote gmail_quote_con=
tainer"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, May 27, 2026, 5:14=E2=
=80=AFPM Johann &#39;Myrkraverk&#39; Oskarsson &lt;<a href=3D"mailto:johann=
@myrkraverk.com">[email protected]</a>&gt; wrote:<br></div><blockquote =
class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid=
;padding-left:1ex"><div dir=3D"auto"><div>Please excuse the bullshit, we ar=
e having an ncurses episode.</div><div dir=3D"auto"><br></div><div dir=3D"a=
uto"><br></div><div dir=3D"auto">So which Single Unix Specification is at f=
ault, and which email list should we include in this discussion?=C2=A0 I ca=
n add myself as <a href=3D"mailto:[email protected]" target=3D"_bla=
nk" rel=3D"noreferrer">[email protected]</a>, but that&#39;s just a=
n alias for this email, and I have never worked on the Single Unix Specific=
ation myself.</div><div dir=3D"auto"><br></div><div dir=3D"auto"><br></div>=
<div><br></div><div data-smartmail=3D"gmail_signature">Johann<br><br> =C2=
=A0I&#39;m not from the internet, I just work there.</div></div><br><div cl=
ass=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, May 27, 2=
026, 5:03=E2=80=AFPM Liam Huang &lt;<a href=3D"mailto:[email protected]" target=
=3D"_blank" rel=3D"noreferrer">[email protected]</a>&gt; wrote:<br></div><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc s=
olid;padding-left:1ex"><div style=3D"line-break:after-white-space"><div>Hi =
Johann,</div><div><br></div><div>Just a quick clarification =E2=80=94 the i=
msg library I&#39;m referring to is the</div><div>OpenBSD IPC library (imsg=
_init, imsg_compose, etc.), not the macOS</div><div>iMessage CLI tool at <a=
 href=3D"http://github.com/openclaw/imsg" rel=3D"noreferrer noreferrer" tar=
get=3D"_blank">github.com/openclaw/imsg</a>. They share a name but</div><di=
v>are completely unrelated projects.</div><div><br></div><div>The OpenBSD i=
msg library provides structured message passing over Unix</div><div>domain =
sockets, and is used by many OpenBSD daemons (OpenSMTPD,</div><div>OpenBGPD=
, etc.). tmux bundles a compat copy of it in its source tree</div><div>(com=
pat/imsg.c, compat/imsg-buffer.c) so it can build on non-OpenBSD</div><div>=
systems like Linux and macOS.</div><div><br></div><div>This is not a Linux-=
vs-SUS issue. It&#39;s simply that tmux upgraded its</div><div>bundled copy=
 of this library between releases 3.5a and 3.6b, and the</div><div>new vers=
ion changed the wire format in a way that&#39;s incompatible with</div><div=
>the old one. When an old server and new client try to communicate, the</di=
v><div>fd-passing mechanism breaks.</div><div><br></div><div>Best,</div><di=
v>Liam</div><div><br></div><div><br><blockquote type=3D"cite"><div>2026=E5=
=B9=B45=E6=9C=8827=E6=97=A5 16:59=EF=BC=8CJohann &#39;Myrkraverk&#39; Oskar=
sson &lt;<a href=3D"mailto:[email protected]" rel=3D"noreferrer norefer=
rer" target=3D"_blank">[email protected]</a>&gt; =E5=86=99=E9=81=93=EF=
=BC=9A</div><br><div><span style=3D"font-family:Helvetica;font-size:12px;fo=
nt-style:normal;font-variant-caps:normal;font-weight:400;letter-spacing:nor=
mal;text-align:start;text-indent:0px;text-transform:none;white-space:normal=
;word-spacing:0px;text-decoration:none;float:none;display:inline!important"=
>Johann</span></div></blockquote></div><br></div></blockquote></div>
</blockquote></div>

--0000000000006ca93a0652c928c3--