Re: Package maintainer script policy.

1998-08-04 Thread Raul Miller
Manoj Srivastava <[EMAIL PROTECTED]> wrote: > amazed Raul, of all people, had to have this pointed out to him. I fail to see that you've demonstrated that ${1+"$@"} is different from "$@" for any posix shell. Which, I thought, was the issue we were talking about. -- Raul -- Unsubscribe? m

Re: Package maintainer script policy.

1998-08-04 Thread Ruud de Rooij
On 1998/08/04, Manoj Srivastava wrote: > Raul> Manoj Srivastava <[EMAIL PROTECTED]> wrote: > >> Well, if $@ is empty, "$@" expands to ""; ^ > > Raul> Is this mandated by the posix standard? If so, this is an example of > Raul> how debian's current

Re: Package maintainer script policy.

1998-08-04 Thread Manoj Srivastava
Hi, This just goes to prove even Debian developers may not be up to the task of passing all commandline arguments to a binary in a shell script. So, if one can write a C program that handles the task, it should be valid to have that called directly in a postinst, rather than mess up wr

Re: Package maintainer script policy.

1998-08-04 Thread Raul Miller
Manoj Srivastava <[EMAIL PROTECTED]> wrote: > Well, if $@ is empty, "$@" expands to ""; Is this mandated by the posix standard? If so, this is an example of how debian's current /bin/sh is not posix. If not, then this is just plain bogus. -- Raul -- Unsubscribe? mail -s unsubscribe

Re: Package maintainer script policy.

1998-08-04 Thread Manoj Srivastava
Hi, >>"Anselm" == Anselm Lingnau <[EMAIL PROTECTED]> writes: Anselm> Raul Miller wrote: >> > How many people actually understand what ${1+"$@"} does, really? >> >> Interesting question. I thought it meant exactly the same thing >> as "$@". Why? Anselm> There used to (?) be shells that