[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Explicit interpreter paths considered harmful
From: |
Jim Meyering |
Subject: |
Re: Explicit interpreter paths considered harmful |
Date: |
Fri, 30 Oct 2009 09:31:23 +0100 |
Ludovic Courtès wrote:
...
> because perl eventually execs /bin/sh:
>
> 4473 execve("./build-aux/gitlog-to-changelog",
> ["./build-aux/gitlog-to-changelog"], [/* 60 vars */]) = 0
> 4473 execve("/var/run/current-system/sw/bin/perl", ["perl", "-S",
> "./build-aux/gitlog-to-changelog"], [/* 59 vars */]) = 0
> 4473 execve("/bin/sh", ["/bin/sh", "-S", "./build-aux/gitlog-to-changelog"],
> [/* 59 vars */]) = 0
>
> Actually perlrun(1) reads this:
>
> #!/bin/sh -- # -*- perl -*- -p
> eval 'exec perl -wS $0 ${1+"$@"}'
> if $running_under_some_shell;
>
> ... which can’t possibly work.
>
> So I guess I’ll stick to what I suggested.
As Simon pointed out, if you start the script with #!/bin/sh,
it breaks any existing user who happens to invoke it
via the suggested $(PERL) gitlog-to-changelog
Earlier, you mentioned running announce-gen manually.
I would recommend not to do that. There is already an "announcement"
target in maint.mk, but even that requires a manual RELEASE_TYPE setting.
I invoke it only via "make beta" or "make stable".
If the existing "alpha beta stable" rule in maint.mk doesn't work for
you, please suggest an improvement -- or you can always adapt it for
your own use.
- Re: Explicit interpreter paths considered harmful, (continued)
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Paolo Bonzini, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Simon Josefsson, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Karl Berry, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/29
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Eric Blake, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Eric Blake, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Eric Blake, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Jim Meyering, 2009/10/30
- Re: Explicit interpreter paths considered harmful, Ludovic Courtès, 2009/10/30