Perl 5 Porters Mailing List Summary: July 6th-12th
Hey everyone,
Following is the p5p (Perl 5 Porters) mailing list summary for the past week. Enjoy!
July 6th-12th
News and updates
Salvatore Bonaccorso reported a security issue in XSLoader to receive a proper CVE ID for it. The fix was already committed in perl.
Father Chrysostomos
notifies
he created a branch with work to finally remove the deprecated
${^ENCODING}
functionality and makes
encoding
die when its invocation would use that functionality. It also
recommends
Filter::Encoding
as an alternative.
Grant reports
Dave Mitchell provided the Grant #2 report for June and reports #134 and #135.
Issues
New issues
- Perl #128540:
warnings pragma silently ignores
postfixed
FATAL
import parameter. - Perl #128557: Trailing space in heredoc delimiter.
- Perl #128559: Code and documentation is inconsistent about naming UTF-8.
- Perl #128574:
Some tests call
fresh_perl*
incorrectly. - Perl #128588: Tied hash magic ignored in references the first time 'round.
- Perl #128589:
Quadmath builds fail
porting/libperl.t
. - Perl #128591:
PerlIO
refuses to readSTDOUT
or writeSTDIN
. - Perl #128597:
SEGV caused by
isLEXWARN_off
. - Perl #128607:
Confusion in
sysseek
documentation.
Shlomi Fish reported what Father Chrysostomos confirmed is a stack referencing bug.
Resolved issues
- Perl #20793: Valgrind, segfaults, Gentoo, Term::ReadLine::Gnu.
- Perl #23907: Crash during global destruction with Storable and threads.
- Perl #24645:
unlink
not working in specific piece of code, possibly an issue with threading or buffering. - Perl #28438:
Return out of first
{body}
. - Perl #33755:
utf8_heavy.pl
, tainted and insecure dependency. - Perl #34489:
perl5db.pl noTTY=1 AutoTrace=1
causes problems. - Perl #34496:
h2ph
is unable to parsedo { ... } while (0)
constructs in macro. - Perl #36113: 5.8.6 build fails on Solaris 8 with Studio 10 Compiler.
- Perl #36664: Strange behavior of shared array.
- Perl #39061: Win32 taint fork crash.
- Perl #128255:
Assert fail in
S_sublex_done
without other symptoms. - Perl #128253:
Assert fail in
S_find_uninit_var
. - Perl #128257:
Segfault in
Perl_gv_setref
. - Perl #128432:
Failures in
t/re/regexp.t
should show the expected result (more clearly). - Perl #128451:
Win32:
$Config{libpth}
is broken in 5.25.2. - Perl #128517: ExtUtils::ParseXS Please make the output reproducible.
- Perl #128525: CPAN Data::Dumper is behind blead.
- Perl #128538:
Fix copy/paste error in
Configure
. - Perl #128555:
S_reg
assertion failure. - Perl #128597:
SEGV caused by
isLEXWARN_off
.
Rejected issues
- Perl #22077:
Perl 5
sysread
not returning from interrupt. - Perl #23963: Possible buffer overrun in threads.
- Perl #38513:
DESTROY
silences failures. - Perl #121195: Unable to configure perl 5.8.8 on AIX 5.3 TL10.
- Perl #128590:
Interference of Switch and
//
.
Proposed patches
Tony Cook provides a patch in
Perl #128564
to increase parallelization for GNU make
builds on Win32.
Dan Collins provides a patch in Perl #30807 to change the text of two messages to propose that the data might also be corrupt or not a Storable binary image.
Dan Collins also provides a patch in Perl #33156 to fix a core API documentation problem.
Father Chrysostomos provides a patch in
Perl #128588
to properly handle tied hashes for references to references in
ref
calls.
Graham Knop provided a patch to fix warnings in Data::Dumper test on old perl.
Tony Cook provided a patch for Perl #128524 (Data::Dumper gets string lengths wrong when the utf8 flag is set).
Discussion
Karl Williamson started a very interesting discussion on a possible security-related use of Unicode. Since different scripts might have characters that look similar, it is possible to provide a different Unicode script of a similar-looking character as part of a string, thus mixing the Unicode scripts. Karl suggests adding a feature to regular expression to check for same-script string ("script-run"). The discussion takes place on the best syntax for it in regular expressions.
Dave Mitchell
summerized
the possible solutions for the /$empty/
issue, with a new
suggestion on how to approach it.
Maxwell Carey
asked
how to access %^H
from XS. Zefram
provided
a thorough explanation.
Leave a comment