Re: pkg/56497: emacs 27 fails to build on Solaris 10 (autogen.sh shell)

[Claes Nästén <pekdon%gmail.com@localhost>]

The following reply was made to PR pkg/56497; it has been noted by GNATS.

From: =?utf-8?q?Claes_N=C3=A4st=C3=A9n?= <pekdon%gmail.com@localhost>
To: gnats-bugs <gnats-bugs%netbsd.org@localhost>
Cc: 
Subject: Re: pkg/56497: emacs 27 fails to build on Solaris 10 (autogen.sh shell)
Date: Sun, 14 Nov 2021 12:04:56 +0100

 Excerpts from nia's message of 2021-11-14 10:25:01 +0000:
 > The following reply was made to PR pkg/56497; it has been noted by GNATS.
 > 
 > From: nia <nia%NetBSD.org@localhost>
 > To: gnats-bugs%netbsd.org@localhost
 > Cc: 
 > Subject: Re: pkg/56497: emacs 27 fails to build on Solaris 10 (autogen.sh
 >  shell)
 > Date: Sun, 14 Nov 2021 10:21:15 +0000
 > 
 >  On Sun, Nov 14, 2021 at 08:45:01AM +0000, Claes Nstn wrote:
 >  >  See the attached Makefile.in patch, where autogen.sh is called as is
 >  >  (and it has #!/bin/sh hardcoded). Using sh to exec it seems to solve the
 >  >  issue but a patch is more work to maintain compared to just replacing
 >  >  the shell don't you think?
 >  
 >  Can't we just use REPLACE_SH on the autogen script? Then we don't need
 >  a patch.
 >  
 
 That was my initial suggestion (although with ifdefs), and was what I
 was trying to suggest with it being more work maintaining a patch. :)
 
 Just did the second diff when trying to figure out where the autogen.sh
 was run from without using the configured shell.
 
 So yeah, REPLACE_SH on autogen.sh and we should be all set.