Set a variable if unset or null previously
in [Shell]
|
Prev: read multiple fields separated by : and fields separated by , in loop stored in $var
Next: Question: add a CC list to `mail` command
From: Thomas 'PointedEars' Lahn on 12 Apr 2010 18:56 Chris F.A. Johnson wrote: > Thomas 'PointedEars' Lahn wrote: >> Chris F.A. Johnson wrote: >>> Thomas 'PointedEars' Lahn wrote: >>>> Dr. David Kirkby wrote: >>>>> ${gapver:=$SAGE_ROOT/spkg/standard/newest_version gap} >>>>> [...] >>>>> Someone has said: >>>>> >>>>> "The meaning of ':=' there is to only set $gapver to $SAGE_ROOT/ >>>>> spkg/.... if >>>>> $gapver was unset or null previously. >>>>> " >>>>> >>>>> Is that POSIX compliant? >>>> The syntax: yes. The statement above: no, and that is not what >>>> happens. $gapver will be set in any case. >>> No, it will only be set (to $SAGE_ROOT/...) if it was unset or >>> null, >> >> At least that is what happens in bash 4.1.2(1). Thanks. > > That's what happens with *any* Bourne-type shell. So you have tested them *all*? For a line in SUSv2+ does not need to mean that all "Bourne-type" shells support it. (For me, it suffices to know that the current bash and POSIX-compliant shells support it like that.) Please trim your quotes to the relevant minimum. PointedEars
From: Barry Margolin on 12 Apr 2010 22:03 In article <1862438.nKmheAe9J7(a)PointedEars.de>, Thomas 'PointedEars' Lahn <PointedEars(a)web.de> wrote: > Chris F.A. Johnson wrote: > > > Thomas 'PointedEars' Lahn wrote: > >> Chris F.A. Johnson wrote: > >>> Thomas 'PointedEars' Lahn wrote: > >>>> Dr. David Kirkby wrote: > >>>>> ${gapver:=$SAGE_ROOT/spkg/standard/newest_version gap} > >>>>> [...] > >>>>> Someone has said: > >>>>> > >>>>> "The meaning of ':=' there is to only set $gapver to $SAGE_ROOT/ > >>>>> spkg/.... if > >>>>> $gapver was unset or null previously. > >>>>> " > >>>>> > >>>>> Is that POSIX compliant? > >>>> The syntax: yes. The statement above: no, and that is not what > >>>> happens. $gapver will be set in any case. > >>> No, it will only be set (to $SAGE_ROOT/...) if it was unset or > >>> null, > >> > >> At least that is what happens in bash 4.1.2(1). Thanks. > > > > That's what happens with *any* Bourne-type shell. > > So you have tested them *all*? For a line in SUSv2+ does not need to mean > that all "Bourne-type" shells support it. (For me, it suffices to know > that the current bash and POSIX-compliant shells support it like that.) I think this has been part of Bourne-style shells for decades, since before POSIX and SUS existed. They simply wrote down what was already a de facto standard. If it doesn't work in any particular shell, I'd claim that it's either not Bourne-style or it has a bug. -- Barry Margolin, barmar(a)alum.mit.edu Arlington, MA *** PLEASE post questions in newsgroups, not directly to me *** *** PLEASE don't copy me on replies, I'll read them in the group ***
From: Geoff Clare on 13 Apr 2010 09:08
Bill Marcum wrote: >> Please, what is SUSv3? >> > Single Unix Specification, version 3. > The SUSv3 standard is practically the same as POSIX, [...] The only difference between SUSv3 and POSIX.1-2001 is that SUSv3 includes an XCurses volume that is not in POSIX.1-2001. Likewise for SUSv4 and POSIX.1-2008. -- Geoff Clare <netnews(a)gclare.org.uk> |