Re: FW: Documentation of POSIXLY_CORRECT

2009-11-16 Thread Ralf Wildenhues
Eric Blake writes: > | Therefore you should `export' again each environment variable that > | you update; the export can occur before or after the assignment. > > You only have to do the export once for the life of the shell, not once per > assignment. Yep, you're right. Sorry for the confusion

Re: FW: Documentation of POSIXLY_CORRECT

2009-11-16 Thread Eric Blake
Ralf Wildenhues gmx.de> writes: > > export POSIXLY_CORRECT > > POSIXLY_CORRECT= > > > > (the two lines can appear in either order). > > FWIW, you need to re-export a variable each time you change it; otherwise, > it is possible that the variable has the new value, but the environment > will pro

Re: FW: Documentation of POSIXLY_CORRECT

2009-11-16 Thread Ralf Wildenhues
Hello, Eric Blake writes: > By the way, if you want to be portable to Solaris /bin/sh, you have to use: > > export POSIXLY_CORRECT > POSIXLY_CORRECT= > > (the two lines can appear in either order). FWIW, you need to re-export a variable each time you change it; otherwise, it is possible that th

Re: FW: Documentation of POSIXLY_CORRECT

2009-11-11 Thread Eric Blake
Pádraig Brady draigBrady.com> writes: > > $ export POSIXLY_CORRECT > > That should be: export POSIXLY_CORRECT= Indeed; and that was the key piece I overlooked. 'export POSIXLY_CORRECT' merely marks POSIXLY_CORRECT for exportation, _iff_ it has a defined value; export POSIXLY_CORRECT both mar

Re: FW: Documentation of POSIXLY_CORRECT

2009-11-11 Thread Pádraig Brady
Schwarz, Konrad wrote: > > > -Original Message- > From: Schwarz, Konrad > Sent: Wednesday, November 11, 2009 2:31 PM > To: 'Eric Blake' > Subject: RE: Documentation of POSIXLY_CORRECT > > Hello Eric, > > thanks for taking this up. > &g

Re: Documentation of POSIXLY_CORRECT

2009-11-11 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [please keep the list in the loop, and please don't top-post on technical discussions] According to Schwarz, Konrad on 11/11/2009 6:31 AM: > Hello Eric, > > thanks for taking this up. > > Here is a session transcript that exhibits the bug: > > $ un

FW: Documentation of POSIXLY_CORRECT

2009-11-11 Thread Schwarz, Konrad
-Original Message- From: Schwarz, Konrad Sent: Wednesday, November 11, 2009 2:31 PM To: 'Eric Blake' Subject: RE: Documentation of POSIXLY_CORRECT Hello Eric, thanks for taking this up. Here is a session transcript that exhibits the bug: $ unset POSIXLY_CO

Re: Documentation of POSIXLY_CORRECT

2009-11-11 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Schwarz, Konrad on 11/11/2009 4:45 AM: > The current text says "[...] define the `POSIXLY_CORRECT' environment > variable." I took this to mean that the variable needs to be exported > (export POSIXLY_CORRECT), Correct. > but it needs

Documentation of POSIXLY_CORRECT

2009-11-11 Thread Schwarz, Konrad
Hi, Section 2.11, Standards conformance, of coreutils.info is worded ambigously with regards to POSIXLY_CORRECT. The current text says "[...] define the `POSIXLY_CORRECT' environment variable." I took this to mean that the variable needs to be exported (export POSIXLY_CORRECT), but it needs t