Date: Mon, 01 Apr 2019 10:54:48 -0400 From: Lowell Gilbert <freebsd-questions-local@be-well.ilk.org> To: freebsd-questions@freebsd.org Subject: Re: eee-dee anyone? Message-ID: <44k1gdydvr.fsf@be-well.ilk.org> In-Reply-To: <20190401122246.89061c15.freebsd@edvax.de> (Polytropon's message of "Mon, 1 Apr 2019 12:22:46 %2B0200") References: <23e162e23288d9a2e498df5f40488bb8@kathe.in> <44muld9su4.fsf@be-well.ilk.org> <20190330035113.65fc995f.freebsd@edvax.de> <441s2o9zrr.fsf@be-well.ilk.org> <20190401122246.89061c15.freebsd@edvax.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Polytropon <freebsd@edvax.de> writes: > On Sat, 30 Mar 2019 10:56:40 -0400, Lowell Gilbert wrote: >> Because it's described by POSIX, ed(1) is with us to stay. Because it >> has non-trivial differences between POSIX and BSD versions (which have >> bitten me in the past), I use sed(1) regardless of whether ed would have >> done the job. I suspect that is a common pattern. > > I think the aspect of POSIX-compliance is one of the main reasons > that so many "old-fashioned" programs still exist in default > installs of many UNIXes. UNIX books which cover UNIX in general, > instead of concentrating on one specific Linux version, still > often cover those "legacy tools". Yes, I just checked two: ed(1) is particularly noteworthy here, because the BSD usages were sufficiently engrained in system management practices (probably on SunOS and Solaris machines at the time) that the POSIX spec actually documents the BSD behaviour, even though it refers to the non-BSD behaviour as "POSIX." This isn't the case for most other utilities: various implementations often support lots and lots of extensions that other implementations don't, but POSIX only covers the common subset that's really portable.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44k1gdydvr.fsf>