pkgsrc-Bugs archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: pkg/54393: devel/gobject-introspection broken on macos (with devel/glib2 correction)
The following reply was made to PR pkg/54393; it has been noted by GNATS.
From: =?UTF-8?Q?cl=C3=A9ment_bouvier?= <clement.bouvier.europe%gmail.com@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: macos-pkg-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
pkgsrc-bugs%netbsd.org@localhost
Subject: Re: pkg/54393: devel/gobject-introspection broken on macos (with
devel/glib2 correction)
Date: Tue, 23 Jul 2019 15:59:31 +0400
--000000000000fb4003058e57ee45
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Hi Patrick,
You are right about DYLD_LIBRARY_PATH.
I totally forgot about that. Sorry.
So we can keep the patch initially and use the variable as it is done with
LD_LIBRARY_PATH.
Concerning time_t , I didn't touch this part, but I can remove it (and
check) if you think that does not affect another platform.
I'm pretty far away of a mac today, I will come back with corrections once
available.
I have also modified devel/glib2. So If you have comments to do...
Thank you for the comments.
Clement.
Le mar 23 juil. 2019 12:15, Patrick Welche <prlw1%cam.ac.uk@localhost> a =C3=A9crit :
> The following reply was made to PR pkg/54393; it has been noted by GNATS.
>
> From: Patrick Welche <prlw1%cam.ac.uk@localhost>
> To: gnats-bugs%netbsd.org@localhost
> Cc:
> Subject: Re: pkg/54393: devel/gobject-introspection broken on macos (with
> devel/glib2 correction)
> Date: Tue, 23 Jul 2019 09:13:17 +0100
>
> On Sun, Jul 21, 2019 at 12:15:00PM +0000,
> clement.bouvier.europe%gmail.com@localhost wrote:
> > Moreover a previous patch by patrick Welche is removed so as to
> > make the giscanner works for the girepository.gir. The patch make
> > giscanner fails on macosx.
>
> I can't claim to be the author of the original patch, but yes,
> patch-giscanner_ccompiler.py was the one I pointed you to off-list.
> I think the idea was "no one should be altering the runtime environment"=
.
>
> With that patch in place, we have to manually alter the MAKE_ENV by addi=
ng
> LD_LIBRARY_PATH+=3D${WRKSRC}/output/$PKGBASE or similar ourselves, and
> of course that won't work on macos, as the equivalent would be
> DYLD_LIBRARY_PATH or somesuch
> (http://mail-index.netbsd.org/pkgsrc-users/2019/07/19/msg029051.html)
>
> I would be happy with allowing gobject-introspection to set the
> flags itself. Maybe those with other views can speak up now?
> [Say we commit this in 10 days if no comment]
>
> I won't claim to understand the macos part, so this all looks good
> to me, apart from getting rid of the @TIMET_TYPENAME@ patches.
> time_t changed from 32-bit to 64-bit on NetBSD at some point, and
> this (at least with autoconf) did the right thing. So, please leave
>
> devel/gobject-introspection/patches/patch-giscanner_ast.py
>
> devel/gobject-introspection/patches/patch-tests_scanner_Regress-1.0-expe=
cted.gir
>
> alone as time_t is irrelevant to macos.
>
>
--000000000000fb4003058e57ee45
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
<div dir=3D"auto">Hi Patrick,<div dir=3D"auto"><br></div><div dir=3D"auto">=
You are right about DYLD_LIBRARY_PATH.</div><div dir=3D"auto">I totally for=
got about that. Sorry.</div><div dir=3D"auto">So we can keep the patch init=
ially and use the variable as it is done with LD_LIBRARY_PATH.</div><div di=
r=3D"auto"><br></div><div dir=3D"auto">Concerning time_t , I didn't tou=
ch this part, but I can remove it (and check) if you think that does not af=
fect another platform.</div><div dir=3D"auto"><br></div><div dir=3D"auto">I=
'm pretty far away of a mac today, I will come back with corrections on=
ce available.</div><div dir=3D"auto"><br></div><div dir=3D"auto">I have als=
o modified devel/glib2. So If you have comments to do...</div><div dir=3D"a=
uto"><br></div><div dir=3D"auto">Thank you for the comments.</div><div dir=
=3D"auto"><br></div><div dir=3D"auto">Clement.</div><div dir=3D"auto"><div =
dir=3D"auto"><div dir=3D"auto"><div dir=3D"auto"><br><br><div class=3D"gmai=
l_quote"><div dir=3D"ltr" class=3D"gmail_attr">Le mar 23 juil. 2019 12:15, =
Patrick Welche <<a href=3D"mailto:prlw1%cam.ac.uk@localhost";>prlw1%cam.ac.uk@localhost</a>&g=
t; a =C3=A9crit=C2=A0:<br></div><blockquote class=3D"gmail_quote" style=3D"=
margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">The followin=
g reply was made to PR pkg/54393; it has been noted by GNATS.<br>
<br>
From: Patrick Welche <<a href=3D"mailto:prlw1%cam.ac.uk@localhost"; target=3D"_blan=
k" rel=3D"noreferrer">prlw1%cam.ac.uk@localhost</a>><br>
To: <a href=3D"mailto:gnats-bugs%netbsd.org@localhost"; target=3D"_blank" rel=3D"noref=
errer">gnats-bugs%netbsd.org@localhost</a><br>
Cc: <br>
Subject: Re: pkg/54393: devel/gobject-introspection broken on macos (with<b=
r>
=C2=A0devel/glib2 correction)<br>
Date: Tue, 23 Jul 2019 09:13:17 +0100<br>
<br>
=C2=A0On Sun, Jul 21, 2019 at 12:15:00PM +0000, <a href=3D"mailto:clement.b=
ouvier.europe%gmail.com@localhost" target=3D"_blank" rel=3D"noreferrer">clement.bouvi=
er.europe%gmail.com@localhost</a> wrote:<br>
=C2=A0> Moreover a previous patch by patrick Welche is removed so as to<=
br>
=C2=A0> make the giscanner works for the girepository.gir. The patch mak=
e<br>
=C2=A0> giscanner fails on macosx.<br>
<br>
=C2=A0I can't claim to be the author of the original patch, but yes,<br=
>
=C2=A0patch-giscanner_ccompiler.py was the one I pointed you to off-list.<b=
r>
=C2=A0I think the idea was "no one should be altering the runtime envi=
ronment".<br>
<br>
=C2=A0With that patch in place, we have to manually alter the MAKE_ENV by a=
dding<br>
=C2=A0LD_LIBRARY_PATH+=3D${WRKSRC}/output/$PKGBASE or similar ourselves, an=
d<br>
=C2=A0of course that won't work on macos, as the equivalent would be<br=
>
=C2=A0DYLD_LIBRARY_PATH or somesuch<br>
=C2=A0(<a href=3D"http://mail-index.netbsd.org/pkgsrc-users/2019/07/19/msg0=
29051.html" rel=3D"noreferrer noreferrer" target=3D"_blank">http://mail-ind=
ex.netbsd.org/pkgsrc-users/2019/07/19/msg029051.html</a>)<br>
<br>
=C2=A0I would be happy with allowing gobject-introspection to set the<br>
=C2=A0flags itself. Maybe those with other views can speak up now?<br>
=C2=A0[Say we commit this in 10 days if no comment]<br>
<br>
=C2=A0I won't claim to understand the macos part, so this all looks goo=
d<br>
=C2=A0to me, apart from getting rid of the @TIMET_TYPENAME@ patches.<br>
=C2=A0time_t changed from 32-bit to 64-bit on NetBSD at some point, and<br>
=C2=A0this (at least with autoconf) did the right thing.=C2=A0 So, please l=
eave<br>
<br>
=C2=A0devel/gobject-introspection/patches/patch-giscanner_ast.py<br>
=C2=A0devel/gobject-introspection/patches/patch-tests_scanner_Regress-1.0-e=
xpected.gir<br>
<br>
=C2=A0alone as time_t is irrelevant to macos.<br>
<br>
</blockquote></div></div></div></div></div></div>
--000000000000fb4003058e57ee45--
Home |
Main Index |
Thread Index |
Old Index