Array slice indexing bug (when mixing positive and negative notation)

[p5pRT]

Migrated from rt.perl.org#9533 (status was 'resolved')

Searchable as RT9533$

[p5pRT]

From torbenh@mips.com

This is a bug report for perl from torbenh@​mips.com,
generated with the help of perlbug 1.33 running under perl v5.6.1.

#!/usr/local/bin/perl -w

# I hope the bug is obvious​: I expect all the 4 last cases to be the same
# I am using perl 5.6.1 on Sun/Solaris 5.8 - I also tried perl 5.6.0 on
# i386/Linux with same result
#
# /Torben Vendt Hansen

@​a = qw(1 2 3 4 5);

print("\@​a​: @​a\n");
print("OK​: \$a[3] $a[3]\n");
print("OK​: \$a[-2] $a[-2]\n");
print("OK​: \@​a[3 .. 4] @​a[3 .. 4]\n");
print("OK​: \@​a[-2 .. -1] @​a[-2 .. -1]\n");
print("BUG​: \@​a[3 .. -1] @​a[3 .. -1]\n");
print("BUG​: \@​a[-2 .. 4] @​a[-2 .. 4]\n");

---

Flags​:
  category=core
  severity=high

---

Site configuration information for perl v5.6.1​:

Configured by steve at Wed Apr 25 02​:36​:44 EDT 2001.

Summary of my perl5 (revision 5.0 version 6 subversion 1) configuration​:
  Platform​:
  osname=solaris, osvers=2.8, archname=sun4-solaris
  uname='sunos solaris 5.8 generic sun4u sparc sunw,ultra-5_10 '
  config_args='-Dcc=gcc -de'
  hint=recommended, useposix=true, d_sigaction=define
  usethreads=undef use5005threads=undef useithreads=undef usemultiplicity=undef
  useperlio=undef d_sfio=undef uselargefiles=define usesocks=undef
  use64bitint=undef use64bitall=undef uselongdouble=undef
  Compiler​:
  cc='gcc', ccflags ='-fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
  optimize='-O',
  cppflags='-fno-strict-aliasing -I/usr/local/include'
  ccversion='', gccversion='2.95.3 20010315 (release)', gccosandvers='solaris2.8'
  intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=4321
  d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
  ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
  alignbytes=8, usemymalloc=y, prototype=define
  Linker and Libraries​:
  ld='gcc', ldflags =' -L/usr/local/lib '
  libpth=/usr/local/lib /usr/lib /usr/ccs/lib
  libs=-lsocket -lnsl -lgdbm -ldb -ldl -lm -lc
  perllibs=-lsocket -lnsl -ldl -lm -lc
  libc=/lib/libc.so, so=so, useshrplib=false, libperl=libperl.a
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags=' '
  cccdlflags='-fPIC', lddlflags='-G -L/usr/local/lib'

Locally applied patches​:
 

---

@​INC for perl v5.6.1​:
  /usr/local/lib/perl5/5.6.1/sun4-solaris
  /usr/local/lib/perl5/5.6.1
  /usr/local/lib/perl5/site_perl/5.6.1/sun4-solaris
  /usr/local/lib/perl5/site_perl/5.6.1
  /usr/local/lib/perl5/site_perl
  .

---

Environment for perl v5.6.1​:
  HOME=/home/torbenh
  LANG=C
  LANGUAGE (unset)
  LD_LIBRARY_PATH=/mips/sw/lsf42/sparcOS5_64/lib​:/mips/sw/lsf42/sparcOS5_64/lib​:/usr/local/lib​:/usr/openwin/lib​:/usr/dt/lib​::/usr/local/freetds/lib
  LOGDIR (unset)
  PATH=/usr/local/lsf/sparcOS5_64/bin​:.​:/home/torbenh/bin​:/home/torbenh/bin/Solaris​:/usr/local/bin​:/bin​:/usr/bin​:/sbin​:/usr/sbin​:/usr/ucb​:/usr/ccs/bin​:/usr/openwin/bin​:/usr/dt/bin​:/home/eng/bin
  PERL_BADLANG (unset)
  SHELL=/usr/local/bin/tcsh

[p5pRT]

From [Unknown Contact. See original ticket]

On Jun 11, Torben Vendt Hansen said​:

# I hope the bug is obvious​: I expect all the 4 last cases to be the same
# I am using perl 5.6.1 on Sun/Solaris 5.8 - I also tried perl 5.6.0 on
# i386/Linux with same result

It is not a bug.

@​a = qw(1 2 3 4 5);

print("BUG​: \@​a[3 .. -1] @​a[3 .. -1]\n");
print("BUG​: \@​a[-2 .. 4] @​a[-2 .. 4]\n");

@​a[-2 .. 4] is merely shorthand for @​a[-2, -1, 0, 1, 2, 3, 4], which is
consistant and proper. @​a[3 .. -1] should yield a warning; the range
operator takes two operands, a low and a high, in that order. It is NOT
magically linked to arrays somehow, and it does not work backwards.

--
Jeff "japhy" Pinyan japhy@​pobox.com http​://www.pobox.com/~japhy/
RPI Acacia brother #734 http​://www.perlmonks.org/ http​://www.cpan.org/
** Look for "Regular Expressions in Perl" published by Manning, in 2002 **
<stu> what does y/// stand for? <tenderpuss> why, yansliterate of course.
[ I'm looking for programming work. If you like my work, let me know. ]

[p5pRT]

From @schwern

On Tue, Jun 11, 2002 at 10​:41​:58AM -0400, Jeff 'japhy' Pinyan wrote​:

@​a[-2 .. 4] is merely shorthand for @​a[-2, -1, 0, 1, 2, 3, 4], which is
consistant and proper. @​a[3 .. -1] should yield a warning; the range
operator takes two operands, a low and a high, in that order. It is NOT
magically linked to arrays somehow, and it does not work backwards.

Maybe the bug is that it doesn't produce a warning when used backwards.

$ perl -wle '@​a = qw(1 2 3 4 5); print @​a[3 .. -1]'

Or maybe it should just work backwards. :)

--
This sig file temporarily out of order.

[p5pRT]

From [Unknown Contact. See original ticket]

-----BEGIN PGP SIGNED MESSAGE-----

Moin,

From​: Michael G Schwern <schwern@​pobox.com>
To​: japhy@​pobox.com
Cc​: Torben Vendt Hansen <torbenh@​mips.com>, perl5-porters@​perl.org
Date​: Tue, 11 Jun 2002 15​:43​:07 -0400
Message-id​: <20020611194307.GM31617@​ool-18b93024.dyn.optonline.net>

On Tue, Jun 11, 2002 at 10​:41​:58AM -0400, Jeff 'japhy' Pinyan wrote​:

@​a[-2 .. 4] is merely shorthand for @​a[-2, -1, 0, 1, 2, 3, 4], which is
consistant and proper. @​a[3 .. -1] should yield a warning; the range
operator takes two operands, a low and a high, in that order. It is NOT
magically linked to arrays somehow, and it does not work backwards.

Maybe the bug is that it doesn't produce a warning when used backwards.

$ perl -wle '@​a = qw(1 2 3 4 5); print @​a[3 .. -1]'

Or maybe it should just work backwards. :)

You could overload it. Oh wait, I never managed to bribe^Wlobby someone to
implement this...​:)

- --
perl -MDev​::Bollocks -e'print Dev​::Bollocks->rand(),"\n"'
ambassadorially transform visionary m-commerce

http​://bloodgate.com/perl My current Perl projects
PGP key available on http​://bloodgate.com/tels.asc or via email

-----BEGIN PGP SIGNATURE-----
Version​: GnuPG v1.0.6 (GNU/Linux)
Comment​: When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

iQEVAwUBPQZvGXcLPEOTuEwVAQHNsQf+Ml4PTgnv4B2OhAFsyJBmg4k1GtCQN4/1
vVzEGTdMcN8koijMFFWlNJ6lf85JFMQDh0UJnhoCloplgciWSgqGr6nJXAZT/q7p
v1C5GB6J+75xZLKfFLJfMvB32Ec0kIZp92X391byGh/q7GgJaAtjJMQs6clvUbe3
Tp22x6czvrI3n4+2ZizYwkFkAQqpIzhd7+Qnx9uS1sNkjoyt28Ag0STlnxRdWJry
/whmxGKzRlpxWlj9E+ROSgCkstfdq4ndAtc8VFMYn/q8rs2CEQ8kMsCeIc7L8QSn
XSoNf4BiqtJE2yrQYGVNc85/kQwaH9zm6vG6QJrx8y6HKJMkVWFeZw==
=N4yf
-----END PGP SIGNATURE-----

[p5pRT]

From @Abigail

On Tue, Jun 11, 2002 at 03​:43​:07PM -0400, Michael G Schwern wrote​:

On Tue, Jun 11, 2002 at 10​:41​:58AM -0400, Jeff 'japhy' Pinyan wrote​:

@​a[-2 .. 4] is merely shorthand for @​a[-2, -1, 0, 1, 2, 3, 4], which is
consistant and proper. @​a[3 .. -1] should yield a warning; the range
operator takes two operands, a low and a high, in that order. It is NOT
magically linked to arrays somehow, and it does not work backwards.

Maybe the bug is that it doesn't produce a warning when used backwards.

I don't think so. I find the behaviour useful and I often exploit it.
Not with 2 constants in the range of course, but with one of them
variable.

  # Deal with the rest of the array.
  foreach my $elem (@​array [$i .. $#array]) { ... }

And this neatly skips doesn't run the loop if the range is empty,
that is, if $i &gt; $#array.

$ perl -wle '@​a = qw(1 2 3 4 5); print @​a[3 .. -1]'

Or maybe it should just work backwards. :)

 
That's why we have $#array and @​array in scalar context.

Abigail

[p5pRT]

From @iabyn

On Wed, Jun 12, 2002 at 02​:59​:10AM -0700, Abigail wrote​:

On Tue, Jun 11, 2002 at 03​:43​:07PM -0400, Michael G Schwern wrote​:

On Tue, Jun 11, 2002 at 10​:41​:58AM -0400, Jeff 'japhy' Pinyan wrote​:

@​a[-2 .. 4] is merely shorthand for @​a[-2, -1, 0, 1, 2, 3, 4], which is
consistant and proper. @​a[3 .. -1] should yield a warning; the range
operator takes two operands, a low and a high, in that order. It is NOT
magically linked to arrays somehow, and it does not work backwards.

Maybe the bug is that it doesn't produce a warning when used backwards.

I don't think so. I find the behaviour useful and I often exploit it.
Not with 2 constants in the range of course, but with one of them
variable.

\# Deal with the rest of the array\.
foreach my $elem \(@​array \[$i \.\. $\#array\]\) \{  \.\.\.  \}

Maybe a warning only when both values are constant?
(Of course, this would still annoy those people using eval.)

Dave.

--
"Emacs isn't a bad OS once you get used to it.
It just lacks a decent editor."