From @cpansprout

This one-liner hangs during regexp compilation​:
perl5.13.5 -e 'qr/\88/'

In perl5.13.4 it compiles successfully, but qr/\9/ dies, because it is an invalid backreference.

I think the correct behaviour is for qr/\88/ to die the same way.

Note also that it does not hang if the second digit is a valid octal digit​:
$ perl5.13.5 -le 'print qr/\87/'
(?-xism​:\87)

This was caused by​:

From c99e91e Mon Sep 17 00​:00​:00 2001
From​: Karl Williamson <khw@​khw-desktop.(none)>
Date​: Sun, 18 Jul 2010 12​:28​:14 -0600
Subject​: [PATCH] Fix /[\8]/ to not match NULL; give correct warning

8 and 9 are not treated as alphas in parsing as opposed to illegal
octals.

This also adds tests to verify that 1-3 digits work in char classes.

I created an isOCTAL macro in case that lookup gets moved to a bit
field, as I plan to do later, for speed.

Flags​:
  category=core
  severity=high
  category=parsing
  category=regex

Site configuration information for perl 5.13.5​:

Configured by sprout at Sun Sep 19 17​:34​:01 PDT 2010.

Summary of my perl5 (revision 5 version 13 subversion 5 patch v5.13.5-8-g8e58c70) configuration​:
  Snapshot of​: 8e58c70
  Platform​:
  osname=darwin, osvers=10.4.0, archname=darwin-thread-multi-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 -DDEBUGGING -Duseithreads'
  hint=recommended, useposix=true, d_sigaction=define
  useithreads=define, usemultiplicity=define
  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 -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include',
  optimize='-O3 -g',
  cppflags='-no-cpp-precomp -fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -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 5.13.5​:
  /usr/local/lib/perl5/site_perl/5.13.5/darwin-thread-multi-2level
  /usr/local/lib/perl5/site_perl/5.13.5
  /usr/local/lib/perl5/5.13.5/darwin-thread-multi-2level
  /usr/local/lib/perl5/5.13.5
  /usr/local/lib/perl5/site_perl
  .

Environment for perl 5.13.5​:
  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

From @khwilliamson

The attached patch fixes this. Note that the circumstances when this
bug happened are if the first digit is any of [1-9], and the second
digit is [89]. I think it doesn't happen if the number of capture
groups is at least this two digit number.

Note that qr/\88/ means qr/\00088/. I consider this a bug, but others
says it has to be this way for backward compatiblity.

From @khwilliamson

From 039c541858ae9a7b3fb2962ade008b1b2a704b57 Mon Sep 17 00:00:00 2001
From: Karl Williamson <public@khwilliamson.com>
Date: Tue, 28 Sep 2010 14:44:08 -0600
Subject: [PATCH] regcomp.c [perl #78058] qr/\18/ loops

This patch restores the behavior of /\18/ to mean /\0018/, which was
inadvertently changed by c99e91e919b4bb89bab7829a9026ee01b1fff2a1.

This bug happens when t the first digit is [1-9], and the
second [89].
---
 regcomp.c |    2 +-
 t/re/re.t |    9 +++++++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/regcomp.c b/regcomp.c
index ff9f87b..bcdc65f 100644
--- a/regcomp.c
+++ b/regcomp.c
@@ -7601,7 +7601,7 @@ tryagain:
 		    case '0': case '1': case '2': case '3':case '4':
 		    case '5': case '6': case '7': case '8':case '9':
 			if (*p == '0' ||
-			    (isOCTAL(p[1]) && atoi(p) >= RExC_npar))
+			    (isDIGIT(p[1]) && atoi(p) >= RExC_npar))
 			{
 			    I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
 			    STRLEN numlen = 3;
diff --git a/t/re/re.t b/t/re/re.t
index 76835f0..ecbcb29 100644
--- a/t/re/re.t
+++ b/t/re/re.t
@@ -53,13 +53,18 @@ if ('1234'=~/(?:(?<A>\d)|(?<C>!))(?<B>\d)(?<A>\d)(?<B>\d)/){
     is(regnames_count(),3);
 }
 
-    { # Keep this test last, as whole script will be interrupted if times out
+    { # Keep these tests last, as whole script will be interrupted if times out
         # Bug #72998; this can loop 
         watchdog(2);
         eval '"\x{100}\x{FB00}" =~ /\x{100}\N{U+66}+/i';
         pass("Didn't loop");
+
+        # Bug #78058; this can loop
+        watchdog(2);
+        eval 'qr/\18/';
+        pass("qr/\18/ didn't loop");
     }
 
 # New tests above this line, don't forget to update the test count below!
-BEGIN { plan tests => 19 }
+BEGIN { plan tests => 20 }
 # No tests here!
-- 
1.5.6.3

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

From @cpansprout

On Tue Sep 28 13​:56​:09 2010, public@​khwilliamson.com wrote​:

The attached patch fixes this. Note that the circumstances when this
bug happened are if the first digit is any of [1-9], and the second
digit is [89]. I think it doesn't happen if the number of capture
groups is at least this two digit number.

Note that qr/\88/ means qr/\00088/. I consider this a bug, but others
says it has to be this way for backward compatiblity.

Thank you. Applied as ca67da4.

From [Unknown Contact. See original ticket]

On Tue Sep 28 13​:56​:09 2010, public@​khwilliamson.com wrote​:

The attached patch fixes this. Note that the circumstances when this
bug happened are if the first digit is any of [1-9], and the second
digit is [89]. I think it doesn't happen if the number of capture
groups is at least this two digit number.

Note that qr/\88/ means qr/\00088/. I consider this a bug, but others
says it has to be this way for backward compatiblity.

Thank you. Applied as ca67da4.

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