Re: Ghostscript and GhostPDL 9.20 release

Chris Liddell <[email protected]> Thu, 29 Sep 2016 03:54:40 +0100
Newsgroups gmane.comp.printing.ghostscript.devel
Message-ID <CAGy8pWruEqqpsC4RT4viGhWAepk4QhxiFSiP1JCU9bmUXjQcfw@mail.gmail.com>
On 28 September 2016 at 21:49, Andreas Radke <[email protected]> wrote:
> Am Tue, 27 Sep 2016 22:54:01 +0100
> schrieb Chris Liddell <[email protected]>:
>
> Redownloaded the source and now it compiles with system libs. But make
> install fails in a new way
>
> "make DESTDIR="${pkgdir}" \
>         cups_serverroot="${pkgdir}"/etc/cups \
>         cups_serverbin="${pkgdir}"/usr/lib/cups install install-so":
>
> mkdir -p /build/ghostscript/pkg/ghostscript/usr/bin
> mkdir -p /build/ghostscript/pkg/ghostscript/usr/share
> mkdir -p /build/ghostscript/pkg/ghostscript/usr/share/ghostscript
> mkdir -p /build/ghostscript/pkg/ghostscript/usr/share/ghostscript/9.20
> ./base/instcopy -c -m 755 ./bin/gs /build/ghostscript/pkg/ghostscript/usr/bin/gs
> make: *** No rule to make target 'lib/gsnd', needed by 'install-scripts'.  Stop.
>
> Any idea?

I just tried the above invocation, and it worked fine for me -
including installing the "gsnd" script. So, no, sorry, no idea.

> I'll file the requested bug tomorrow.
>
>
> Note: Trying to build the omni driver fails here:
>
> ./contrib/gomni.c:115:22: fatal error: gdevprna.h: No such file or directory
>  #include "gdevprna.h"
>                       ^
> compilation terminated.
> make[2]: *** [contrib/contrib.mak:687: soobj/gomni.o] Error 1

The omni device hasn't functioned correctly since I have been involved
with Ghostscript and with no way to tests fixes, we aren't in a
position to fix it. Thus we decided to remove it from the default
build. It also uses the (long!) deprecated async device API. We were
asked (I can't recall by whom) a few years ago to leave it as an
option in case anyone was using it. It would seem not, and now we've
removed the async device code, I think omni is going to suffer the
same fate.

Chris