scalar //g match on overloaded object causes string corruption

[p5pRT]

Migrated from rt.perl.org#77084 (status was 'resolved')

Searchable as RT77084$

[p5pRT]

From @cpansprout

use Data​::Dumper;
use overoad q/""/=>sub{"foo​:bar"};
bless [], =~ /(.*)/g;
$Data​::Dumper​::Useqq++;
print Dumper "$1";
__END__

The output varies, but it’s usually something like​:

$VAR1 = "\t\0\0\0\263\337!";

---

Flags​:
  category=core
  severity=medium

---

Site configuration information for perl 5.13.3​:

Configured by sprout at Sat Jul 31 05​:50​:29 PDT 2010.

Summary of my perl5 (revision 5 version 13 subversion 3 patch v5.13.3-140-g79504a3) configuration​:
  Snapshot of​: 79504a3
  Platform​:
  osname=darwin, osvers=10.4.0, archname=darwin-2level
  uname='darwin pint.local 10.4.0 darwin kernel version 10.4.0​: fri apr 23 18​:28​:53 pdt 2010; root​:xnu-1504.7.4~1release_i386 i386 '
  config_args='-de -Dusedevel'
  hint=recommended, useposix=true, d_sigaction=define
  useithreads=undef, usemultiplicity=undef
  useperlio=define, d_sfio=undef, uselargefiles=define, usesocks=undef
  use64bitint=undef, use64bitall=undef, uselongdouble=undef
  usemymalloc=n, bincompat5005=undef
  Compiler​:
  cc='cc', ccflags ='-fno-common -DPERL_DARWIN -no-cpp-precomp -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include',
  optimize='-O3',
  cppflags='-no-cpp-precomp -fno-common -DPERL_DARWIN -no-cpp-precomp -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'
  ccversion='', gccversion='4.2.1 (Apple Inc. build 5664)', gccosandvers=''
  intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
  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, prototype=define
  Linker and Libraries​:
  ld='env MACOSX_DEPLOYMENT_TARGET=10.3 cc', ldflags =' -fstack-protector -L/usr/local/lib'
  libpth=/usr/local/lib /usr/lib
  libs=-ldbm -ldl -lm -lutil -lc
  perllibs=-ldl -lm -lutil -lc
  libc=/usr/lib/libc.dylib, so=dylib, useshrplib=false, libperl=libperl.a
  gnulibc_version=''
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
  cccdlflags=' ', lddlflags=' -bundle -undefined dynamic_lookup -L/usr/local/lib -fstack-protector'

Locally applied patches​:
 

---

@​INC for perl 5.13.3​:
  /usr/local/lib/perl5/site_perl/5.13.3/darwin-2level
  /usr/local/lib/perl5/site_perl/5.13.3
  /usr/local/lib/perl5/5.13.3/darwin-2level
  /usr/local/lib/perl5/5.13.3
  /usr/local/lib/perl5/site_perl
  .

---

Environment for perl 5.13.3​:
  DYLD_LIBRARY_PATH (unset)
  HOME=/Users/sprout
  LANG=en_US.UTF-8
  LANGUAGE (unset)
  LD_LIBRARY_PATH (unset)
  LOGDIR (unset)
  PATH=/usr/bin​:/bin​:/usr/sbin​:/sbin​:/usr/local/bin​:/usr/X11/bin​:/usr/local/bin
  PERL_BADLANG (unset)
  SHELL=/bin/bash

[p5pRT]

From @Abigail

On Sun, Aug 08, 2010 at 02​:36​:31PM -0700, Father Chrysostomos wrote​:

# New Ticket Created by Father Chrysostomos
# Please include the string​: [perl #77084]
# in the subject line of all future correspondence about this issue.
# <URL​: http​://rt.perl.org/rt3/Ticket/Display.html?id=77084 >

use Data​::Dumper;
use overoad q/""/=>sub{"foo​:bar"};
bless [], =~ /(.*)/g;
$Data​::Dumper​::Useqq++;
print Dumper "$1";
__END__

The output varies, but it’s usually something like​:

$VAR1 = "\t\0\0\0\263\337!";

I can confirm this.

Included a TODO test for this in commit a9f6cb1.

Abigail

[p5pRT]

The RT System itself - Status changed from 'new' to 'open'

[p5pRT]

From @iabyn

On Thu, Aug 19, 2010 at 06​:38​:08PM +0200, Abigail wrote​:

On Sun, Aug 08, 2010 at 02​:36​:31PM -0700, Father Chrysostomos wrote​:

use Data​::Dumper;
use overoad q/""/=>sub{"foo​:bar"};
bless [], =~ /(.*)/g;
$Data​::Dumper​::Useqq++;
print Dumper "$1";
__END__

The output varies, but it’s usually something like​:

$VAR1 = "\t\0\0\0\263\337!";

I can confirm this.
Included a TODO test for this in commit a9f6cb1.

This is now partially fixed with commit

  f427b55

More to the point, this fixes the original bug report, but doesn't pass
all the tests in t/re/overload.t. In particular, the pos() position after
the second (scalar) match is attached to $o, and re-used for the third
match (void), which then matches the zero-length string at the end.

I think this is correct behaviour and that the third test is wrong.
Opinions?

PS Abigail​: why was a new test file created for this rather than just
adding to lib/overload.t ?

--
You never really learn to swear until you learn to drive.

[p5pRT]

From @Abigail

On Wed, Sep 01, 2010 at 05​:16​:07PM +0100, Dave Mitchell wrote​:

On Thu, Aug 19, 2010 at 06​:38​:08PM +0200, Abigail wrote​:

On Sun, Aug 08, 2010 at 02​:36​:31PM -0700, Father Chrysostomos wrote​:

use Data​::Dumper;
use overoad q/""/=>sub{"foo​:bar"};
bless [], =~ /(.*)/g;
$Data​::Dumper​::Useqq++;
print Dumper "$1";
__END__

The output varies, but it’s usually something like​:

$VAR1 = "\t\0\0\0\263\337!";

I can confirm this.
Included a TODO test for this in commit a9f6cb1.

This is now partially fixed with commit

f427b557f0f598ee6ff57dfc24080cf33d959242

More to the point, this fixes the original bug report, but doesn't pass
all the tests in t/re/overload.t. In particular, the pos() position after
the second (scalar) match is attached to $o, and re-used for the third
match (void), which then matches the zero-length string at the end.

I think this is correct behaviour and that the third test is wrong.
Opinions?

You are right. I've fixed this by setting pos ($o) = 0 after
the second test. (commit f3a0def).

PS Abigail​: why was a new test file created for this rather than just
adding to lib/overload.t ?

No particular reason. I was looking for an appropriate test file in t/re,
and didn't find one. Hence a new file. But since it's an issue about the
interaction of regexes and overloading, putting in in lib/overload.t works
as well.

Abigail

[p5pRT]

@iabyn - Status changed from 'open' to 'resolved'