02 5:54 PM
Subject: Re: /usr/bin/env - Incorrect parsing of #! line?
> On Tue, Feb 12, 2002 at 05:31:03PM -0800, David Gluss wrote:
> >The problem looks to be that bash "helps out" the system by executing
> >scripts beginning with #!. In the source for bash, look in
> &
What's a command completion spec? Given a simple testcase,
I could look into it.
DG
- Original Message -
From: "Christopher Faylor" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, February 12, 2002 6:25 PM
Subject: Re: bash 2.05a command completion spec bug
> On Tue, Feb 12,
The problem looks to be that bash "helps out" the system
by executing scripts beginning with #!. In the source for
bash, look in execute_cmd.c, line 3369. Only one argument
is allowed. So e.g. #!/usr/bin/env perl -w becomes
"/usr/bin/env" "perl -w"
If I make a patch for this, should it go to th
I think it takes a while to get used to the style of this
list. Thank you for explaining what was meant by
"because... we're mean".
Perhaps a little forgiveness and tolerance all around
would (slap)...oops, never mind...
DG
PS:
% env foobar
env: foobar: No such file or directory
% env foobar -w
e
I've only tried it with csh, tcsh, sh, and bash. I put that first line in
with a colon
to put emacs into perl-mode. It could probably be a # instead...as long as
it's
not #!.
- Original Message -
From: "Peter J. Acklam" <[EMAIL PROTECTED]>
To: "Davi
I don't know if it's constructive to suggest an alternative trick, rather
than trying to fix cygwin, in this forum. However, this might work
for you:
>
>: # -*-Mode: perl;-*- use perl, wherever it is
>eval 'exec perl -wS $0 ${1+"$@"}'
> if 0;
>#!/usr/local/bin/perl -w
>
D
My fix was hasty. Here is a better one.
DG
bisonbug.tar.bz2
Description: Binary data
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.c
switching in bison.simple used to
work, and now it doesn't. The patch is very small and simple.
Indeed if you compile with gcc, you won't have a problem.
- Original Message -
From: "Andrew Markebo" <[EMAIL PROTECTED]>
To: "David Gluss" <[EMAIL PR
unt?
>
> Thanks.
>
> Randall Schulz
> Mountain View, CA USA
>
>
> At 16:54 2002-01-31, David Gluss wrote:
> >Well it took some doing...I downloaded Service Pack 5 (130 Mbytes, once
> >you find it)
>
>
> --
> Unsubscribe info: http://cygwin.com/ml/#un
oblem with bison.hairy (just at a guess) but
I've never used that.
Now that I've tried that (using the %semantic_parser line in the .y file) I
can
say there's nothing to worry about, because bison just dumps core if you use
it.
> > -Original Message-
> > From:
Well it took some doing...I downloaded Service Pack 5 (130 Mbytes, once you
find it)
- Original Message -
From: "Stephan Mueller" <[EMAIL PROTECTED]>
To: "David Gluss" <[EMAIL PROTECTED]>
Sent: Thursday, January 31, 2002 1:21 PM
Subject: RE: Bug in new
I'm at patch level 3 of Microsoft C++ 6.0. I see they
are up to patch level 5, so I will update, try it again, and
report.
- Original Message -
From: "Stephan Mueller" <[EMAIL PROTECTED]>
To: "David Gluss" <[EMAIL PROTECTED]>
Sent: Thursday, January
With version 1.31-1, Bison doesn't work with the Microsoft
compiler. A small example is enclosed. The problem is that
Microsoft doesn't have size_t in the std:: namespace, or malloc.
Included also, a patch to fix bison.simple.
DG
bisonbug.tar.bz2
Description: Binary data
--
Unsubscribe i
13 matches
Mail list logo
