Re: pushd/popd

2018-11-12 Thread David Wright
On Mon 12 Nov 2018 at 08:40:15 (-0500), Greg Wooledge wrote: > On Sat, Nov 10, 2018 at 11:58:53PM -0600, David Wright wrote: > > BTW whenever I change directory in scripts, I make sure that I'm in a > > subshell by using ( and ), which guarantees that I get back to the > > same directory however th

Re: pushd/popd

2018-11-12 Thread Greg Wooledge
On Sat, Nov 10, 2018 at 11:58:53PM -0600, David Wright wrote: > BTW whenever I change directory in scripts, I make sure that I'm in a > subshell by using ( and ), which guarantees that I get back to the > same directory however the script finishes. Depends on the script. For scripts that change d

Re: pushd/popd

2018-11-10 Thread David Wright
On Sat 10 Nov 2018 at 09:47:37 (-0800), pe...@easthope.ca wrote: > * From: pe...@easthope.ca > * Date: Fri, 09 Nov 2018 06:20:01 -0800 > > Why? Easy enough to use pushd/popd and easy enough to not [use] it but > > I'm interested in the reasoning behind this choice. > > I've read the argu

Re: pushd/popd

2018-11-10 Thread peter
* From: pe...@easthope.ca * Date: Fri, 09 Nov 2018 06:20:01 -0800 > Why? Easy enough to use pushd/popd and easy enough to not [use] it but > I'm interested in the reasoning behind this choice. I've read the arguments but no answer to my question. Is there an intrinsic reason to avoid

Re: pushd/popd

2018-11-09 Thread Jonathan Dowland
On Fri, Nov 09, 2018 at 11:26:37AM -0500, Greg Wooledge wrote: On Fri, Nov 09, 2018 at 04:20:57PM +, Jonathan Dowland wrote: One of dash's design goals is to be a purely POSIX compatible shell*, Not quite. Fear not, I'm not confusing it with posh(1). That's my reading of the dash(1) manp

Re: pushd/popd

2018-11-09 Thread Greg Wooledge
On Fri, Nov 09, 2018 at 04:20:57PM +, Jonathan Dowland wrote: > One of dash's design goals is to be a purely POSIX compatible shell*, Not quite. > * they actually *do* implement some carefully defined extensions > according to the man page Yeah. It's not meant to be a litmus test for wheth

Re: pushd/popd

2018-11-09 Thread Jonathan Dowland
On Fri, Nov 09, 2018 at 06:20:01AM -0800, pe...@easthope.ca wrote: ... not dash (by design) ... Why? Easy enough to use pushd/popd and easy enough to not it but I'm interested in the reasoning behind this choice. One of dash's design goals is to be a purely POSIX compatible shell*, so it del

Re: pushd/popd

2018-11-09 Thread peter
From: Jonathan Dowland Date: Thu, 8 Nov 2018 09:46:34 + > They are not POSIX, I don't think they are covered by any subsequent > standard either ... OK, thanks. > ... not dash (by design) ... Why? Easy enough to use pushd/popd and easy enough to not it but I'm interested in the r