Re: [Bug 1952] Re: Basic perl error in readline.pm makes CPAN exit go crazy.

[Steve Peters]

I think this was fixed in the upstream module somewhere between
versions 1.0204 and 1.0208.  I can no longer reproduce the problem and
the code causing the problem has been completely rewritten.

Steve

On Tue, Nov 25, 2008 at 2:15 AM, Adam Niedling <[EMAIL PROTECTED]> wrote:
> Any update on this bug?
>
> ** Changed in: libterm-readline-perl-perl (Ubuntu)
>   Status: Confirmed => Incomplete
>
> --
> Basic perl error in readline.pm makes CPAN exit go crazy.
> https://bugs.launchpad.net/bugs/1952
> You received this bug notification because you are a direct subscriber
> of the bug.
>

-- 
Basic perl error in readline.pm makes CPAN exit go crazy.
https://bugs.launchpad.net/bugs/1952
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 297437] [NEW] Perl 5.10.0 crash with a segmentation fault when using module Tcl

[Steve Peters]

On Wed, Nov 12, 2008 at 7:36 PM, spk <[EMAIL PROTECTED]> wrote:
> Public bug reported:
>
> I'm not sure if this is a perl bug or a tcl-dev bug
>
>
> Compiling Tcl module from source
> ...
> Testing Tcl module
>
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ make test
> PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" 
> "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
> t/call.ok
> t/constantsok
> t/createcmdok
> t/eval.ok
> t/info.ok
> t/result...ok
> t/subclass.ok
> t/traceok
> t/unicode..ok
> t/var..ok
> All tests successful.
> Files=10, Tests=53,  1 wallclock secs ( 0.27 cusr +  0.08 csys =  0.35 CPU)
>
> Tests are successfull
>
>
> Locating any previous presence of the module in the system
>
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ sudo updatedb
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ locate Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/backup/Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/blib/lib/Tcl.pm
>
> No library in library paths
>
> Installing the module anche checking the position
>
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ sudo make install
> Files found in blib/arch: installing files in blib/lib into architecture 
> dependent library tree
> Installing /usr/local/lib/perl/5.10.0/Tcl.pm
> Installing /usr/local/lib/perl/5.10.0/prova.pm
> Writing /usr/local/lib/perl/5.10.0/auto/Tcl/.packlist
> Appending installation info to /usr/local/lib/perl/5.10.0/perllocal.pod
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ sudo updatedb
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ locate Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/backup/Tcl.pm
> /home/giacomo/Desktop/Tcl-0.97/blib/lib/Tcl.pm
> /usr/local/lib/perl/5.10.0/Tcl.pm
>
>
> Cat the file and execute
>
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ cat prova.pm
> use Tcl;
>
> [EMAIL PROTECTED]:~/Desktop/Tcl-0.97$ perl prova.pm
> Segmentation fault
>
>
> Backtrace of the gdb
>
> (gdb) bt
> #0  0xb7f5b528 in NpInitialize (my_perl=0xbfda67d8, X=0x9663930) at Tcl.xs:445
> #1  0xb7f5bcfc in boot_Tcl (my_perl=0x965f008, cv=0x96a9dd0) at Tcl.xs:1733
> #2  0x080b3f02 in Perl_pp_entersub ()
> #3  0x080b22e9 in Perl_runops_standard ()
> #4  0x080afdb8 in Perl_call_sv ()
> #5  0x080b019f in Perl_call_list ()
> #6  0x08064daf in ?? ()
> #7  0x08072f37 in Perl_newATTRSUB ()
> #8  0x08071d60 in Perl_utilize ()
> #9  0x08161ee8 in Perl_yyparse ()
> #10 0x080ae2e6 in ?? ()
> #11 0x080b0cc5 in perl_parse ()
> #12 0x08063e81 in main ()
>
>
> Same bug in multiple intrepid postations, but not on different distributions
> Perl is working correctly with other modules
>
> ** Affects: perl (Ubuntu)
> Importance: Undecided
> Status: New
>
> --
> Perl 5.10.0 crash with a segmentation fault when using module Tcl
> https://bugs.launchpad.net/bugs/297437
> You received this bug notification because you are subscribed to perl in
> ubuntu.
>

This is not a Perl or Tcl bug.  This is a bug in the Tcl Perl module
available from CPAN.  A debugging Perl causes the module to fail its
tests as follows.

t/call.ok 1/10Assertion ((svtype)((_svi)->sv_flags & 0xff)) >=
SVt_RV failed: file "Tcl.xs", line 653 at t/call.t line 38.

Please contact the module author from the contact information on
http://search.cpan.org/dist/Tcl for filing a bug report.

Steve Peters
[EMAIL PROTECTED]

-- 
Perl 5.10.0 crash with a segmentation fault when using module Tcl
https://bugs.launchpad.net/bugs/297437
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 336361] Re: perl crashed with SIGSEGV in NpInitialize()

[Steve Peters]

*** This bug is a duplicate of bug 336048 ***
https://bugs.launchpad.net/bugs/336048

Actually, this is not a bug in Perl, but as the backtrace clearly
shows, the Tcl module that is installed is what's causing the core
dump.

Steve Peters
st...@fisharerojo.org

0x7f73ae4b9232 in NpInitialize (my_perl=0x13af800, X=0xdc1f10) at Tcl.xs:445
445 Tcl.xs: No such file or directory.
in Tcl.xs
(gdb) bt full
#0  0x7f73ae4b9232 in NpInitialize (my_perl=0x13af800, X=0xdc1f10)
at Tcl.xs:445

On Sun, Mar 1, 2009 at 1:35 PM, Richard Eames  wrote:
> *** This bug is a duplicate of bug 336048 ***
>    https://bugs.launchpad.net/bugs/336048
>
>
> ** Attachment added: "gdb backtrace"
>   http://launchpadlibrarian.net/23267515/gdb-perl.txt
>
> ** Attachment added: "Dependencies.txt"
>   http://launchpadlibrarian.net/23267516/Dependencies.txt
>
> ** Attachment added: "ProcMaps.txt"
>   http://launchpadlibrarian.net/23267517/ProcMaps.txt
>
> ** Attachment added: "ProcStatus.txt"
>   http://launchpadlibrarian.net/23267518/ProcStatus.txt
>
> ** This bug has been marked a duplicate of bug 336048
>   perl crashed with SIGSEGV in NpInitialize()
>
> --
> perl crashed with SIGSEGV in NpInitialize()
> https://bugs.launchpad.net/bugs/336361
> You received this bug notification because you are subscribed to perl in
> ubuntu.
>

-- 
perl crashed with SIGSEGV in NpInitialize()
https://bugs.launchpad.net/bugs/336361
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 336361] Re: perl crashed with SIGSEGV in NpInitialize()

[Steve Peters]

*** This bug is a duplicate of bug 336048 ***
https://bugs.launchpad.net/bugs/336048

That said, this is not a problem in the Perl core.  This is a problem
in the perl extension called Tcl that allows Perl code to interface
with the Tcl libraries.  My guess is that the big listed at
http://rt.cpan.org/Public/Bug/Display.html?id=21181 is the same
problem you are occurring.  The bug should be put in the perl-tcl
queue, not the perl queue.

Steve Peters
st...@fisharerojo.org

On Sun, Mar 1, 2009 at 2:39 PM, Richard Eames  wrote:
> *** This bug is a duplicate of bug 336048 ***
>    https://bugs.launchpad.net/bugs/336048
>
> The problem is, I have Tcl installed properly. I've done some
> experimenting with this, and it looks like Tkx wont install on linux at
> all, unless through activestate 32bit (doesn't work on 64bit at all).
> I've had to go to my rarely used windows install on my laptop in order
> to use Tkx :(
>
> --
> perl crashed with SIGSEGV in NpInitialize()
> https://bugs.launchpad.net/bugs/336361
> You received this bug notification because you are subscribed to perl in
> ubuntu.
>

-- 
perl crashed with SIGSEGV in NpInitialize()
https://bugs.launchpad.net/bugs/336361
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 328519] [NEW] cannot use LOG_EMERG

[Steve Peters]

Nice! Trying to get a patch into Ubuntu around the maintainers back.
This ticket should be immediately closed.

Steve Peters
st...@fisharerojo.org

On Thu, Feb 12, 2009 at 7:39 AM, Fredrik Ljunggren  wrote:
> Public bug reported:
>
> Reference: https://rt.cpan.org/Public/Bug/Display.html?id=17518
>
> Bug in Syslog.pm version 0.13, fixed in 0.14:
>
>
> cannot use LOG_EMERG or "emerg" as priority with following error.
>
> syslog: invalid level/facility: emerg at ./syslog.pl line 46
>
> Syslog.pm#syslog (line 632) is
> if ($_ eq 'kern' || $num <= 0) {
> but I think
> if ($_ eq 'kern' || $num < 0) {
> is correct.
>
> ** Affects: perl (Ubuntu)
> Importance: Undecided
> Status: New
>
> --
> cannot use LOG_EMERG
> https://bugs.launchpad.net/bugs/328519
> You received this bug notification because you are subscribed to perl in
> ubuntu.
>

-- 
cannot use LOG_EMERG
https://bugs.launchpad.net/bugs/328519
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 347303] [NEW] ENV{TZ} and strftime in perl not working

[Steve Peters]

On Wed, Mar 25, 2009 at 4:13 PM, Launchpad Bug Tracker
<347...@bugs.launchpad.net> wrote:
> You have been subscribed to a public bug:
>
> If you run the following:
>
> perl -e 'use POSIX;
> $ENV{TZ} = "UTC";
> my $time = strftime("%Y-%m-%d %H:%M:%S %Z", localtime(time));
> print "$time\n";
> $ENV{TZ} = "US/Eastern";
> my $time = strftime("%Y-%m-%d %H:%M:%S %Z", localtime(time));
> print "$time\n";'
>
> You should get two different times.
>
> 2009-03-23 14:15:13 UTC
> 2009-03-23 10:15:13 EDT
>
>  But in Ubuntu 8.04 server (lsb_release -rd =>Description:      Ubuntu
> 8.04.2, Release: 8.04) you instead get
>
> 2009-03-23 14:15:40 UTC
> 2009-03-23 14:15:40 EDT
>
> It does the same thing for almost any timezone you give it. It give UTC
> time but labels it as the proper time zone.
>
> This happens on my 8.04 desktop version as well, but not on any 8.10
> machines.
>
> ** Affects: perl (Ubuntu)
>     Importance: Undecided
>         Status: New
>
> --
> ENV{TZ} and strftime in perl not working
> https://bugs.edge.launchpad.net/bugs/347303
> You received this bug notification because you are subscribed to perl in 
> ubuntu.
>

I can confirm that the Perl distributed with Ubuntu 8.04 has problems.
 This problem, however, appears to be a bug in the Debian/Ubuntu perl.
 I just built the original Perl 5.8.8 and everything worked fine.
This problem is caused by some Debian/Ubuntu applied patch.

Steve Peters
st...@fisharerojo.org

-- 
ENV{TZ} and strftime in perl not working
https://bugs.launchpad.net/bugs/347303
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 57061] Re: [Bug 57061] $p->{"type"} tries to call a method named type

[Steve Peters]

On Mon, Aug 21, 2006 at 08:22:37AM -, Nikolaus Rath wrote:
> Public bug reported:
> 
>   DB<1> s
> main::valid_answer(/home/nikratio/Projekte/Archiver/Archiver.pm:314):
> 314:if($p->{"type"} eq "string") {
>   DB<1> x $p
> 0  HASH(0x892a01c)
>'default' => 'never'
>'emptyok' => 1
>'query' => 'Expiration date'
>'type' => 'timeoff'
>   DB<2> x $p->{"type"}
> 0  'timeoff'
>   DB<3> s
> main::(/home/nikratio/bin/aadd:94):
> 94: if($@) {
>   DB<3> x $@
> 0  'Can\'t call method "type" on unblessed reference at 
> /home/nikratio/Projekte/Archiver/Archiver.pm line 314,  line 10.
> '
>   DB<4>
> 
> ** Affects: perl (Ubuntu)
>  Importance: Untriaged
>  Status: Unconfirmed
> 
> -- 
> $p->{"type"} tries to call a method named type

My guess would be that somewhere in the object hierarchy, something has
an AUTOLOAD defined.  This is the expected behavior in that case.

Steve Peters
[EMAIL PROTECTED]

-- 
$p->{"type"} tries to call a method named type
https://launchpad.net/bugs/57061

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 57061] Re: [Bug 57061] Re: Wrong line number in error message

[Steve Peters]

On Tue, Sep 05, 2006 at 06:44:30PM -, Nikolaus Rath wrote:
> 
> ** Summary changed:
> 
> - $p->{"type"} tries to call a method named type
> + Wrong line number in error message
> 
> ** Description changed:
> 
> -   DB<1> s
> - main::valid_answer(/home/nikratio/Projekte/Archiver/Archiver.pm:314):
> - 314:if($p->{"type"} eq "string") {
> -   DB<1> x $p
> - 0  HASH(0x892a01c)
> -'default' => 'never'
> -'emptyok' => 1
> -'query' => 'Expiration date'
> -'type' => 'timeoff'
> -   DB<2> x $p->{"type"}
> - 0  'timeoff'
> -   DB<3> s
> - main::(/home/nikratio/bin/aadd:94):
> - 94: if($@) {
> -   DB<3> x $@
> - 0  'Can\'t call method "type" on unblessed reference at 
> /home/nikratio/Projekte/Archiver/Archiver.pm line 314,  line 10.
> - '
> -   DB<4>
> + nokile:~/Work$ ./bug.pl
> + Can't call method "type" on unblessed reference at ./bug.pl line 18.
> + 
> + but the error is in line 27.
> 
> ** Attachment added: "testcase"
>http://librarian.launchpad.net/4136476/bug.pl
> 

Unfortunately, I don't believe this bug will ever be fixed.  The problem
is how Perl code is represented internally by the perl executable.  This
is because the optree does not contain line number information on the 
elsif lines.  It is not a trivial problem and would like cause 
performance to suffer it is fixed.  See the Perl bug queue for more
information on this problem at 
<http://rt.perl.org/rt3/Ticket/Display.html?id=37302>.

Steve Peters
[EMAIL PROTECTED]

-- 
Wrong line number in error message
https://launchpad.net/bugs/57061

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 52106] Re: default CPAN.pm installation does not install other modules

[Steve Peters]

The most recent released version of CPAN is 1.87.  Perl-5.8.8 was
released with a bug fixed version unavailable elsewhere on CPAN
(1.76_02).  When 5.8.9 is released, it will likely include 1.87 or a
later version.

-- 
default CPAN.pm installation does not install other modules
https://launchpad.net/bugs/52106

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 52150] Re: Error Parsing Debugger Variables with EPIC

[Steve Peters]

This has been fixed in the development version of Perl.  The change is
available here .

** Changed in: perl (Ubuntu)
   Status: Unconfirmed => Confirmed

-- 
Error Parsing Debugger Variables with EPIC
https://launchpad.net/bugs/52150

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 57061] Re: Wrong line number in error message

[Steve Peters]

** Changed in: perl (Ubuntu)
   Status: Unconfirmed => Confirmed

-- 
Wrong line number in error message
https://launchpad.net/bugs/57061

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 192055] Re: Failure when using getpwnam core function from perl

[Steve Peters]

Whenever getpwnam() returns undef, $! should be set.  Without properly
checking $!, you will not have a way to understand what the root cause
error is.  There will be no way to determine whether the problem is in
Perl or glibc.

Steve

On Feb 18, 2008 10:54 AM, Alceu Rodrigues de Freitas Junior
<[EMAIL PROTECTED]> wrote:
> Hello Steve,
>
> No, the addsmb script is not testing the result from getpwnam(). Even if
> it were, we could only generate a better error message or use the
> sleep() for waiting the user to be added to /etc/passwd correctly. While
> is a good programming practice to test the result from calling a
> function, I believe we could do better to understand why the error is
> happening before trying a hack in the Perl script.
>
> Unfornately I'm unable to generate a test with the Ubuntu version that
> we detected the error, we will need to wait for an update from Tiago.
>
>
> --
> Failure when using getpwnam core function from perl
> https://bugs.launchpad.net/bugs/192055
> You received this bug notification because you are a bug contact for
> perl in ubuntu.
>

-- 
Failure when using getpwnam core function from perl
https://bugs.launchpad.net/bugs/192055
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 132702] Conflicts ... (<< ..._...) which is illegal syntax

[Steve Peters]

On Wed, Aug 15, 2007 at 03:10:03PM -, Ian Jackson wrote:
> Public bug reported:
> 
> Binary package hint: perl
> 
> Package: perl-modules
> Version: 5.8.8-7ubuntu2
> Conflicts: libpod-parser-perl (<< 1.32-1), libansicolor-perl (<< 1.10-1), 
> libfile-temp-perl (<< 0.16-1), libnet-perl (<= 1:1.19-3), 
> libattribute-handlers-perl (e-temp-perl (<< 0.16-1), libnet-perl (<= 
> 1:1.19-3), libattribute-handlers-perl (<< 0.78_02-1), libcgi-pm-perl (<< 
> 3.15-1), libi18n-langtags-perl (<< 0.35-1), liblocale-maketext-perl (<< 
> 1.08-1), libmath-bigint-perl (<< 1.77-1), libnet-ping-perl (<< 2.31-1), 
> libtest-harness-perl (<< 2.56-1), libtest-simple-perl (<< 0.62-1), 
> liblocale-codes-perl (<< 2.06.1-1)
> 
> Note the dependency on libattribute-handlers-perl.  _ is not a legal
> character in version numbers and this control file has a syntax error.
> 
> This is a surprisingly serious bug: some automated package processing
> programs may well reject this package.
> 

_ is valid in module version numbers in Perl.  My guess is that this is simply
a problem with the package files.

Steve Peters
[EMAIL PROTECTED]

-- 
Conflicts ... (<< ..._...) which is illegal syntax
https://bugs.launchpad.net/bugs/132702
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 160693] [perl] [cve-2007-5116] heap overflow

[Steve Peters]

The patch applied to the Perl core is the following...


 //depot/maint-5.8/perl/regcomp.c#109 (text) 

@@ -117,7 +117,10 @@
 I32extralen;
 I32seen_zerolen;
 I32seen_evals;
-I32utf8;
+I32utf8;   /* whether the pattern is utf8 or not */
+I32orig_utf8;  /* whether the pattern was originally 
in utf8 */
+   /* XXX use this for future optimisation of case
+* where pattern must be upgraded to utf8. */
 #if ADD_TO_REGEXEC
 char   *starttry;  /* -Dr: where regtry was called. */
 #define RExC_starttry  (pRExC_state->starttry)
@@ -143,6 +146,7 @@
 #define RExC_seen_zerolen  (pRExC_state->seen_zerolen)
 #define RExC_seen_evals(pRExC_state->seen_evals)
 #define RExC_utf8  (pRExC_state->utf8)
+#define RExC_orig_utf8 (pRExC_state->orig_utf8)

 #defineISMULT1(c)  ((c) == '*' || (c) == '+' || (c) == '?')
 #defineISMULT2(s)  ((*s) == '*' || (*s) == '+' || (*s) == '?' || \
@@ -1720,15 +1724,17 @@
 if (exp == NULL)
FAIL("NULL regexp argument");

-RExC_utf8 = pm->op_pmdynflags & PMdf_CMP_UTF8;
+RExC_utf8 = RExC_orig_utf8 = pm->op_pmdynflags & PMdf_CMP_UTF8;

-RExC_precomp = exp;
 DEBUG_r({
 if (!PL_colorset) reginitcolors();
 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s `%s%*s%s'\n",
   PL_colors[4],PL_colors[5],PL_colors[0],
   (int)(xend - exp), RExC_precomp, PL_colors[1]);
 });
+
+redo_first_pass:
+RExC_precomp = exp;
 RExC_flags = pm->op_pmflags;
 RExC_sawback = 0;
@@ -1730,7 +1730,7 @@
 if (!PL_colorset) reginitcolors();
 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s `%s%*s%s'\n",
   PL_colors[4],PL_colors[5],PL_colors[0],
-  (int)(xend - exp), RExC_precomp, PL_colors[1]);
+  (int)(xend - exp), exp, PL_colors[1]);
 });

 redo_first_pass:
@@ -1754,6 +1760,25 @@
RExC_precomp = NULL;
return(NULL);
 }
+if (RExC_utf8 && !RExC_orig_utf8) {
+/* It's possible to write a regexp in ascii that represents unicode
+codepoints outside of the byte range, such as via \x{100}. If we
+detect such a sequence we have to convert the entire pattern to utf8
+and then recompile, as our sizing calculation will have been based
+on 1 byte == 1 character, but we will need to use utf8 to encode
+at least some part of the pattern, and therefore must convert the whole
+thing.
+XXX: somehow figure out how to make this less expensive...
+-- dmq */
+STRLEN len = xend-exp;
+DEBUG_r(PerlIO_printf(Perl_debug_log,
+   "UTF8 mismatch! Converting to utf8 for resizing and compile\n"));
+exp = (char*)Perl_bytes_to_utf8(aTHX_ (U8*)exp, &len);
+xend = exp + len;
+RExC_orig_utf8 = RExC_utf8;
+SAVEFREEPV(exp);
+goto redo_first_pass;
+}
 DEBUG_r(PerlIO_printf(Perl_debug_log, "size %"IVdf" ", (IV)RExC_size));

 /* Small enough for pointer-storage convention?

 //depot/maint-5.8/perl/t/op/pat.t#46 (xtext) 

@@ -3771,5 +3771,15 @@
 iseq($count,1,"should have matched once only [RT#36046]");
 }

+{
+use warnings;
+local $Message = "ASCII pattern that really is utf8";
+my @w;
+local $SIG{__WARN__}=sub{push @w,"@_"};
+my $c=qq(\x{DF});
+ok($c=~/${c}|\x{100}/);
+ok(@w==0);
+}
+
 # Don't forget to update this!
-BEGIN{print "1..1251\n"};
+BEGIN{print "1..1253\n"};

On 11/7/07, hk47 <[EMAIL PROTECTED]> wrote:
> Public bug reported:
>
> Binary package hint: perl
>
> References:
> [1] http://www.debian.org/security/2007/dsa-1400
> [2] Bug #160454
>
> >From [1]:
> "Will Drewry and Tavis Ormandy of the Google Security Team have discovered a 
> UTF-8 related heap overflow in Perl's regular expression compiler, probably 
> allowing attackers to execute arbitrary code by compiling specially crafted 
> regular expressions."
>
> ** Affects: perl (Ubuntu)
>  Importance: Undecided
>  Status: New
>
> ** Visibility changed to: Public
>
> ** CVE added: http://www.cve.mitre.org/cgi-
> bin/cvename.cgi?name=2007-5116
>
>
> --
> [perl] [cve-2007-5116] heap overflow
> https://bugs.launchpad.net/bugs/160693
> You received this bug notification because you are a bug contact for
> perl in ubuntu.
>

-- 
[perl] [cve-2007-5116] heap overflow
https://bugs.launchpad.net/bugs/160693
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 83461] Re: pack does not support Q, but perl supports 64 bit scalars

[Steve Peters]

To handle 64 bit ints, Perl would have needed to be configured correctly
to handle it.  A perl -V would help to answer that question.

-- 
pack does not support Q, but perl supports 64 bit scalars
https://bugs.launchpad.net/bugs/83461
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 192055] [NEW] Failure when using getpwnam core function from perl

[Steve Peters]

On Fri, Feb 15, 2008 at 1:27 AM, TiagoCruz <[EMAIL PROTECTED]> wrote:
> Public bug reported:
>
>  During tests of FAUS application
>  (http://faus.sourceforge.net), release 1.4.5, an issue was found by using 
> getpwnam function from Perl. This
>  release was already tested against Mandriva, Debian and Oracle Linux without 
> showing the same bug.
>
>  The bug happens when executing the script addsmb by using the web 
> application. Executing the script in a
>  shell as root does not shows the same behaviour, working as expected.
>
>  The script addsmb is a Perl script that execute the following tasks:
>
>  1 - Add an user in /etc/passwd and /etc/shadow by using the Perl module 
> Passwd::Linux.
>  2 - Adds the user in the Samba database by using IPC::Open3 Perl module, 
> smbpasswd and pdbedit
>  programs.
>
>  The Passwd::Linux uses libc functions to manipulate the Linux user 
> databases. During step one of addsmb
>  Perl script, the user is added using Passwd::Linux module exported functions 
> and then calls getpwnam to
>  retrieve the ID created for the user.
>
>  When executed by the web application, the ID is not returned by getpwnam(), 
> that returns undef, making the
>  application to fail to create the user correctly.
>
>  The script addsmb is not suid: it's executed by faus.cgi by using Sudo 
> features. The CGI executes Sudo
>  with IPC::Open3, capturing STDOUT and STDERR. Sudo then invokes addsmb with 
> the given parameters.
>
>  We were unable to detect where the failure may be happening (libc and/or 
> getpwnam from Perl is our guess
>  because some advisory file locking should be published and respected) but we 
> could provide an workaround by
>  using "sleep 3" inside addsmb script, after calling Passwd::Linux and before 
> executing getpwnam().
>
>  Possible related packages:
>  1 - Perl core (includes IPC::Open3 and getpwnam())
>  2 - Sudo
>  3 - Apache
>  4 - Passwd::Linux
>
>  Please let me know what we could test to identify where is the cause of
>  the bug.
>
>  Thank you in advance.
>

My suggestion is to run the cgi file from a command line with Perl.
Also, are you checking $! after calling getpwnam() for any error
messages?

Steve Peters
[EMAIL PROTECTED]

-- 
Failure when using getpwnam core function from perl
https://bugs.launchpad.net/bugs/192055
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 44196] Re: gnome-cups-icon uses 100% CPU

[Steve Peters]

Agreed.  Its time to fix this problem.

Steve

On 11/5/07, Peter Belew <[EMAIL PROTECTED]> wrote:
> I observed this on one of my 6.06 LTS systems a few days ago - running
> top showed that gnome-cups-icon was using 100% or nearly 100% of the
> CPU.
>
> While there is a known workaround (deleting or renaming gnome-cups-
> icon), I feel that this should be fixed in this LTS release ASAP, since
> this IS an LTS release with many months of bugfixing support ahead of
> it. This is a bug that can cause severe degradation of system
> performance. I run 6.06 LTS on some systems precisely because I expect
> them to be stable.
>
> Since apparently a bug fix exists for later releases, I would expect
> that a backport of this fix would be feasible.
>
> Cheers,
>
> Peter
>
> --
> gnome-cups-icon uses 100% CPU
> https://bugs.launchpad.net/bugs/44196
> You received this bug notification because you are a direct subscriber
> of a duplicate bug.
>

-- 
gnome-cups-icon uses 100% CPU
https://bugs.launchpad.net/bugs/44196
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 192055] Re: Failure when using getpwnam core function from perl

[Steve Peters]

On Thu, Feb 21, 2008 at 2:39 PM, TiagoCruz <[EMAIL PROTECTED]> wrote:
> Running with root/strace
>  write(1, "
> PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
>  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd";>
>  http://www.w3.org/1999/xhtml"; lang="en-US" xml:lang="en-US">
>  
>  Error message
>  
>  
>  
>  ) = 325
>  write(2, "Use of uninitialized value in pr"..., 58Use of uninitialized value 
> in print at faus.cgi line 517.
>  ) = 58
>  write(1, " bgcolor="#3684FF" border="0"> color="#ff">ERRORAn error ocurred 
> when running the program  /usr/sbin/addsmbGo back
>  
>  ) = 219
>  write(2, "error raised by line 698: Use of"..., 152error raised by line 698: 
> Use of uninitialized value in chown at /usr/sbin/addsmb line 208. cannot find 
> user in the /etc/passwd:  at faus.cgi line 520.
>  ) = 152
>  write(1, "", 7)  = 7
>  exit_group(255) = ?
>  Process 14596 detached
>

It looks like rather than having problems with getpwname(), there are
actually other bugs within this CGI script.  I would say then that
this problem has been resolved, from this bugtracker's point of view.

Steve

** Changed in: perl (Ubuntu)
   Status: New => Invalid

-- 
Failure when using getpwnam core function from perl
https://bugs.launchpad.net/bugs/192055
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 46281] Re: XS version of Scalar::List::Utils no longer included, breaking other packages

[Steve Peters]

Can anyone confirm whether this is now fixed in Ubuntu?

-- 
XS version of Scalar::List::Utils no longer included, breaking other packages
https://bugs.launchpad.net/bugs/46281
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 90334] 5.8.8-7build1 segfaults on x86

[Steve Peters]

On Wed, Mar 07, 2007 at 09:31:19AM -, Alexander Koenig wrote:
> Public bug reported:
> 
> Binary package hint: perl
> 
> Upgrading an edgy machine to feisty via 'dist-upgrade' yesterday, I ran
> into a problem with the 5.8.8-7build1 of perl. 30 or more packages
> caused a segfault in dpkg's 'Setting up' phase making apt-get fail due
> to too many errors. After some analysis I noticed it was 'perl'
> segfaulting so I downgraded to 5.8.8-7 and put 'perl' on hold. The
> following 'dist-upgrade' went smoothly and dpkg configured all packages
> successfully that had failed before. Unfortunately I don't have a trace
> as I was focused on getting the system back to work.
> 

Yes, without a simplified test case it will be difficult to really dig into
this problem.

Steve Peters
[EMAIL PROTECTED]

-- 
5.8.8-7build1 segfaults on x86
https://launchpad.net/bugs/90334

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 102638] [apport] perl crashed with SIGSEGV in rrd_test_error()

[Steve Peters]

On Wed, Apr 04, 2007 at 12:36:37AM -, Andre Menard wrote:
> You have been subscribed to a public bug:
> 
> This is present when I come back to the computer after 8 hours standing.
> 
> ProblemType: Crash
> Architecture: i386
> CrashCounter: 1
> Date: Mon Apr  2 18:40:05 2007
> Dependencies:
>  tzdata 2007b-0ubuntu1
>  belocs-locales-bin 2.4-2ubuntu2
>  locales 2.3.23
>  libc6 2.5-0ubuntu12
> DistroRelease: Ubuntu 7.04
> ExecutablePath: /usr/bin/perl
> Package: perl-base 5.8.8-7build1
> PackageArchitecture: i386
> ProcCmdline: /usr/share/munin/munin-update\ [localhost.localdomain]
> ProcCwd: /var/lib/munin
> ProcEnviron:
>  
> Signal: 11
> SourcePackage: perl
> StacktraceTop:
>  rrd_test_error () from /usr/lib/librrd.so.2
>  _rrd_update () from /usr/lib/librrd.so.2
>  rrd_update_r () from /usr/lib/librrd.so.2
>  rrd_update () from /usr/lib/librrd.so.2
>  XS_RRDs_update () from /usr/lib/perl5/auto/RRDs/RRDs.so
> Uname: Linux ubuntu 2.6.17-10-generic #2 SMP Tue Dec 5 22:28:26 UTC 2006 i686 
> GNU/Linux
> UserGroups:
> 
> ** Affects: perl (Ubuntu)
>  Importance: Undecided
>  Status: Unconfirmed
> 

It looks like the problem code is in /usr/lib/perl5/auto/RRDs/RRDs.so.  This 
does not appear to be a Perl problem, but a problem with librrds-perl instead.

Steve Peters
[EMAIL PROTECTED]

-- 
[apport] perl crashed with SIGSEGV in rrd_test_error()
https://bugs.launchpad.net/bugs/102638
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 96018] Time::Local::timelocal() is broken over GMT/BST change

[Steve Peters]

On Sun, Mar 25, 2007 at 05:43:34PM -, Crispin Flowerday wrote:
> Public bug reported:
> 
> Binary package hint: perl
> 
> The Time::Local::timelocal() function seems to be broken today after the
> GMT -> BST daylight saving time change in the UK, e.g:
> 
> $ date
> Sun Mar 25 18:28:21 BST 2007
> $  perl -MTime::Local -e 'print scalar localtime( timelocal( localtime() ) ) 
> . "\n";'
> Sun Mar 25 19:28:48 2007
> 
> Notice that going through timelocal() causes the hour to be offset by
> one hour, the above operation should be a no-op and print out the
> current local time (and indeed tomorrow it will work fine).
> 
> Another example:
> 
> use Time::Local;
> @a = ( 0, 43, 5, 25, 2, 107 );  # 5:43am, 25th March 2007
> $t = timelocal( @a ) ;
> print scalar localtime( $t ), "\n"; # Prints 6:43am
> 
> @a = ( 0, 43, 5, 26, 2, 107 );  # 5:43am, 26th March 2007
> $t = timelocal( @a ) ;
> print scalar localtime( $t ), "\n"; # Prints 5:43am
> 
> 
> This broken behaviour seems to be isolated to timezones that changed from GMT 
> (e.g. Europe/London, Europe/Dublin, Europe/Isle_of_Man all show the problem, 
> but Europe/Paris and Europe/Berlin don't).
> 
> I have reproduced this on both edgy and feisty. FWIW, Using Time::Local
> version 1.07 works, but 1.10 doesn't (I don't have 1.08 or 1.09
> available to test).
> 

The most recent version of Time::Local on CPAN is 1.17.  Perhaps using that
will cause the problems to go away.  It appears that the problem was fixed
by 1.13.

1.132006-08-09

- Switched to using Test::More for the tests.

- Fixed a bug that occurred around the DST change for Europe/London
  (and probably other time zones with a positive UTC offset). If
  given, the hour immediately after the change (2:00 AM for
  Europe/London), then the returned epoch was 3600 seconds too
  large. Reported by Roger Picton. See RT #11662.


Steve Peters
[EMAIL PROTECTED]

-- 
Time::Local::timelocal() is broken over GMT/BST change
https://launchpad.net/bugs/96018

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 576827] Re: munin-graph crashed with SIGSEGV in FcPatternDestroy()

[Steve Peters]

Looking at the stack trace, this is an RRD or munin issue, not a Perl
one.  Please reassign.

-- 
munin-graph crashed with SIGSEGV in FcPatternDestroy()
https://bugs.launchpad.net/bugs/576827
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 585042] Re: awstats.pl crashed with SIGSEGV in Perl_hv_common()

[Steve Peters]

How can this be replicated?

-- 
awstats.pl crashed with SIGSEGV in Perl_hv_common()
https://bugs.launchpad.net/bugs/585042
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 600951] [NEW] perl fails to build from source in maverick and linaro (amd64)

[Steve Peters]

A few things.  Would you be able to go to the Perl source directory and do
the following?

cd t
./perl harness -v op/getppid.t

Now paste the output in a response to this email.  Also, the output of that
Perl's "perl -V" would be helpful.

Steve

On Fri, Jul 2, 2010 at 3:57 AM, Matthias Klose  wrote:

> Public bug reported:
>
> Binary package hint: perl
>
> perl fails to build with different build errors in linaro and maverick:
>
> linaro: after the configure check?
> /usr/bin/make SHRPLDFLAGS='$(LDDLFLAGS) -Wl,-soname,libperl.so.5.10'
> libperl.so.5.10.1
> make[1]: Entering directory `/build/buildd/perl-5.10.1'
> make[1]: *** No rule to make target `libperl.so.5.10.1'.  Stop.
> make[1]: Leaving directory `/build/buildd/perl-5.10.1'
> make: *** [libperl.so.5.10.1] Error 2
> dpkg-buildpackage: error: /usr/bin/fakeroot debian/rules binary-arch gave
> error exit status 2
>
> maverick failing a test:
> # Failed at op/getppid.t line 46
> # it should not be '6856'
> # but it is.
> # Failed at op/getppid.t line 78
> #  got '6856'
> # expected '1'
> t/op/getppid..FAILED at
> test 3
>
> http://people.ubuntuwire.org/~lucas/ubuntu-
> nbs/64/perl_5.10.1-12ubuntu1_lubuntu64.buildlog
>
> ** Affects: gcc-linaro
> Importance: Undecided
> Status: New
>
> ** Affects: perl (Ubuntu)
> Importance: High
> Status: Confirmed
>
> ** Affects: perl (Ubuntu Maverick)
> Importance: High
> Status: Confirmed
>
> ** Also affects: perl (Ubuntu Maverick)
>   Importance: Undecided
>   Status: New
>
> ** Changed in: perl (Ubuntu Maverick)
>   Importance: Undecided => High
>
> ** Changed in: perl (Ubuntu Maverick)
>   Status: New => Confirmed
>
> ** Changed in: perl (Ubuntu Maverick)
>Milestone: None => maverick-alpha-3
>
> ** Also affects: gcc-linaro
>   Importance: Undecided
>   Status: New
>
> --
> perl fails to build from source in maverick and linaro (amd64)
> https://bugs.launchpad.net/bugs/600951
> You received this bug notification because you are subscribed to perl in
> ubuntu.
>

-- 
perl fails to build from source in maverick and linaro (amd64)
https://bugs.launchpad.net/bugs/600951
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 739693] Re: Configure does not find libs in /lib/x86_64-linux-gnu

[Steve Peters]

I can look into this, unfortunately not today.  There was a lot of work done
to get the correct 64-bit libs found on Linux since there didn't seem to be
a standard location for them across distributions.  My guess is that Ubuntu
currently has found another way to make this happen or there is some other
configuration that has been done on the build machine that isn't on your
machine.

Steve

On Tue, Mar 22, 2011 at 10:56 AM, TundraMan
<739...@bugs.launchpad.net>wrote:

> I am sure there are other suitable fixes and possible debate if this
> should be made a 'default' search path but here is a potential fix none
> the less.
>
> ** Patch added: "Add the new lib paths to default search paths used by
> Configure"
>
> https://bugs.launchpad.net/ubuntu/+source/perl/+bug/739693/+attachment/1930371/+files/configure-adjust-for-moving-libs.patch
>
> --
> You received this bug notification because you are subscribed to perl in
> Ubuntu.
> https://bugs.launchpad.net/bugs/739693
>
> Title:
>  Configure does not find libs in /lib/x86_64-linux-gnu
>

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/739693

Title:
  Configure does not find libs in /lib/x86_64-linux-gnu

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 790024] Re: package tzdata 2011g-0ubuntu0.11.04 failed to install/upgrade: subprocess installed post-installation script returned error exit status 128

[steve peters]

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/790024

Title:
  package tzdata 2011g-0ubuntu0.11.04 failed to install/upgrade:
  subprocess installed post-installation script returned error exit
  status 128

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 790024] [NEW] package tzdata 2011g-0ubuntu0.11.04 failed to install/upgrade: subprocess installed post-installation script returned error exit status 128

[steve peters]

Public bug reported:

Binary package hint: tzdata

installation failed

ProblemType: Package
DistroRelease: Ubuntu 11.04
Package: tzdata 2011g-0ubuntu0.11.04
ProcVersionSignature: Ubuntu 2.6.38-8.42-generic 2.6.38.2
Uname: Linux 2.6.38-8-generic x86_64
Architecture: amd64
Date: Sun May 29 18:41:54 2011
ErrorMessage: subprocess installed post-installation script returned error exit 
status 128
InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Release amd64 (20110427.1)
PackageArchitecture: all
SourcePackage: tzdata
Title: package tzdata 2011g-0ubuntu0.11.04 failed to install/upgrade: 
subprocess installed post-installation script returned error exit status 128
UpgradeStatus: No upgrade log present (probably fresh install)

** Affects: tzdata (Ubuntu)
 Importance: Undecided
 Status: New


** Tags: amd64 apport-package natty

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/790024

Title:
  package tzdata 2011g-0ubuntu0.11.04 failed to install/upgrade:
  subprocess installed post-installation script returned error exit
  status 128

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs