Confounded by conditional expressions
in [Shell]
|
From: Thomas 'PointedEars' Lahn on 13 Jul 2010 04:47 Sven Mascheck wrote: > Chris F.A. Johnson wrote: >> On 2010-07-12, Kenny McCormack wrote: >>> Chris F.A. Johnson <cfajohnson(a)gmail.com> wrote: >>> ... >>>> Avoid -a and -o; use && and || [...] >>>> >>> Why? >>> >> I can't cite chapter and verse OTTOMH, but I recall seeing an Open >> Group recommendation about not using -a and -o. > > In the posix description of the "test" command these features are > marked as "obsolescent". No, they are not. BTW, the word you were looking for was "obsolete", but that does not apply here, too. > Strictly conforming applications shall not use them, and they might be > removed in a future version of posix. Complete, utter nonsense. The section "Future Directions" of the `test' command in SUSv3 (2004) says exactly "None." > The informative section "application usage" explains that some > expressions are ambiguous depending on the specific expressions > being used. But it refers to user-supplied input, and does NOT mark those features as obsolete. or "obsolescent". > A drastic example, > especially if the input is not well known: > > test -n "$a" -a -n "$b" || echo at-least-one-is-empty > > with > a==; b=x No programmer in their right mind would do that. Users might provide such values (hence "user-supplied"), but that is the only occasion they can be expected. Your From header is as invalid as your POSIX "knowledge". Please fix both. -- PointedEars
From: Janis Papanagnou on 13 Jul 2010 05:28 Chris F.A. Johnson schrieb: > On 2010-07-12, Kenny McCormack wrote: >> In article <8a15fuFhq1U1(a)mid.individual.net>, >> Chris F.A. Johnson <cfajohnson(a)gmail.com> wrote: >> ... >>> Avoid -a and -o; use && and ||: >>> >>> if [ $last_seen -gt 0 ] && [ $last_seen -lt $toolate ] ; then >> Why? >> >> (Not that I have any strong opinion on the matter, and I even agree with >> you, but you usually have some obscure reason for your pronouncements) > > I can't cite chapter and verse OTTOMH, but I recall seeing an Open > Group recommendation about not using -a and -o. Yes. There's this comment: :> Unless the application writer knows all the cases that produce input :> to the script, invocations like: :> :> test "$1" -a "$2" :> :> should be written as: :> :> test "$1" && test "$2" and also, WRT test \( expr1 -a expr2 \) -o expr3 :> This command is not always portable outside XSI-conformant systems. In POSIX the -a and -o and the parenthesis are marked as "extensions". Apparently, there are enough reasons to avoid those options. Janis
From: Thomas 'PointedEars' Lahn on 13 Jul 2010 06:14 Janis Papanagnou wrote: > Chris F.A. Johnson schrieb: >> On 2010-07-12, Kenny McCormack wrote: >>> In article <8a15fuFhq1U1(a)mid.individual.net>, >>> Chris F.A. Johnson <cfajohnson(a)gmail.com> wrote: >>> ... >>>> Avoid -a and -o; use && and ||: >>>> >>>> if [ $last_seen -gt 0 ] && [ $last_seen -lt $toolate ] ; then >>> Why? >>> >>> (Not that I have any strong opinion on the matter, and I even agree with >>> you, but you usually have some obscure reason for your pronouncements) >> >> I can't cite chapter and verse OTTOMH, but I recall seeing an Open >> Group recommendation about not using -a and -o. > > Yes. There's this comment: > > :> Unless the application writer knows all the cases that produce input > :> to the script, IOW, unless they are providing the input and not the user. > invocations like: > :> > :> test "$1" -a "$2" > :> > :> should be written as: > :> > :> test "$1" && test "$2" > > and also, WRT test \( expr1 -a expr2 \) -o expr3 > > :> This command is not always portable outside XSI-conformant systems. > > In POSIX the -a and -o and the parenthesis are marked as "extensions". | expression1 -a expression2 | [XSI] [X> | True if both expression1 and expression2 are true; otherwise, false. | The -a binary primary is left associative. It has a higher precedence | than -o. <X] | expression1 -o expression2 | [XSI] [X> | True if either expression1 or expression2 is true; otherwise, false. | The -o binary primary is left associative. <X] | [...] | ( expression ) | [XSI] [X> True if expression is true. False if expression is false. | The parentheses can be used to alter the normal precedence and | associativity. <X] | [...] | [XSI] [X> Extension <X] | | The functionality described is an XSI extension. Functionality marked XSI | is also an extension to the ISO C standard. Application writers may ^^^^^^^^^^^^^^^^^^^^^^^ | confidently make use of an extension on all systems supporting the X/Open ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | System Interfaces Extension. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | | If an entire SYNOPSIS section is shaded and marked XSI, all the | functionality described in that reference page is an extension. See | XSI Conformance [1] [1] <http://www.opengroup.org/onlinepubs/009695399/basedefs/xbd_chap02.html#tag_02_01_04> > Apparently, there are enough reasons to avoid those options. Apparently, many people like to jump to conclusions. -- PointedEars
From: Geoff Clare on 13 Jul 2010 08:26 Thomas 'PointedEars' Lahn wrote: > Sven Mascheck wrote: > >> Chris F.A. Johnson wrote: >>> On 2010-07-12, Kenny McCormack wrote: >>>> Chris F.A. Johnson <cfajohnson(a)gmail.com> wrote: >>>> ... >>>>> Avoid -a and -o; use && and || [...] >>>>> >>>> Why? >> >> In the posix description of the "test" command these features are >> marked as "obsolescent". > > No, they are not. Yes they are. You are looking at an out-of-date version of the standard. > BTW, the word you were looking for was "obsolete", but > that does not apply here, too. XBD7 section 1.7.1 Codes: [OB] Obsolescent The functionality described may be removed in a future version of this volume of POSIX.1-2008. >> Strictly conforming applications shall not use them, and they might be >> removed in a future version of posix. > > Complete, utter nonsense. The section "Future Directions" of the `test' > command in SUSv3 (2004) says exactly "None." SUSv3 was superseded by SUSv4 in 2008. Sven was referring to the current standard, not the out-of-date one you were looking at. -- Geoff Clare <netnews(a)gclare.org.uk>
From: Kenny McCormack on 13 Jul 2010 13:18
In article <tk2ug7-jtg.ln1(a)leafnode-msgid.gclare.org.uk>, Geoff Clare <geoff(a)clare.See-My-Signature.invalid> wrote: .... >SUSv3 was superseded by SUSv4 in 2008. Sven was referring to the >current standard, not the out-of-date one you were looking at. Don't confuse the pointed out with facts. They only confuse and annoy him. -- > No, I haven't, that's why I'm asking questions. If you won't help me, > why don't you just go find your lost manhood elsewhere. CLC in a nutshell. |