This Week on perl5-porters - 23-29 March 2008
"Anyway, a lot of these design changes work together to produce a
smoother result. I confess I haven't thought much about whether they
could be borrowed piecemeal; I have a hard enough time keeping one
fantasy language in my head at a time." -- Larry Wall, on the
differences in exception handling between Perl 5 and Perl 6.
Topics of Interest
Perl and CPAN distributors
The thread wound up this week with Gabor Szabo asking for someone to
summarise the main points made in the discussion, for inclusion on the
Perl5 wiki. The idea is to make life easier for people who package
perl in operating system distributions, by explaining what they may,
should and must do.
Nicholas Clark noted that starting from 5.11, the infrastructure is
nearly all there for the perl5-porters to remove modules from the core
while allowing distributors and local sites to continue to keep them,
in a graceful manner.
modular ecology
http://xrl.us/biu9g
Perl @ 33536
The 5.8.9 snapshot is having problems on Vista. Jan Dubois saw
t/op/alarm.t failing but wasn't able to track down why. In the
meantime, he released "Win32" 0.35 on CPAN. This will be folded into
5.8.9 to upgrade the "Win32" therein from its current development
status.
http://xrl.us/biu9i
local $@ has an unwanted side effect
Larry Wall gave the porters a run down on the current thinking
regarding exception handling in Perl 6. Most of the pieces are in
place, except for a hierarchy of exception object types, as @Larry are
waiting to see the various implementations make of it all.
there is no try
http://xrl.us/biu9k
http://xrl.us/biu9n
pod wiki?
Reini Urban wanted to write in POD for the Perl 5 wiki (that uses
SocialText as a back-end). Michael G. Schwern explained that the
existing plug-in/markup-escape infrastructure was up to the job, and
sketched out how one might go about doing it. Reini was sufficiently
motivated to take a stab at setting it up.
Tels thought that it was silly to write Yet Another Pod-HTML
convertor, and Michael thought that "Pod::Simple::HTML" should be
close. Aristotle Pagaltzis thought that "Pod::POM::View::HTML" wasn't
too bad either.
http://xrl.us/biu9p
"left_shift" problem
Reini has started to write a document about the guts of the op-tree.
He started with the most trivial of op-codes, the one that performs a
left bitshift ("<<"). Unfortunately he became side-tracked over the
way signed and unsigned integers are used depending on whether the
"use integer" pragma is in force. Yitzchak Scott-Thoennes explained
the reason why things were why they are.
hard left turn right
http://xrl.us/biu9r
Append "64" suffix to some of the installed files
Sérgio Durigan Júnior continued his quest to get 32- and 64-bit perls
coexisting peacefully on his machine. Abigail showed how many
different perls could live in the same directory tree, sharing common
pure-Perl modules and keeping binary-incompatible files separate.
Dominic Dunlop remarked that Apple had managed to do just that
already, but unfortunately no-one outside Apple had managed to
replicate what they had done. Nicholas Clark mentioned that Edward Moy
had some raw patch files to do just that, but didn't have the time to
tidy them up (nor did Nicholas). Sounds like a nice self-contained
task...
Other discussion in the thread talked about how perl is installed on
HP-UX.
more bits
http://xrl.us/biu9t
Too late for -Co
Alberto Simões wanted to know how he could make some old scripts work
with 5.10, since they have "-Co0" on the shebang line. Aristotle
Pagaltzis recommended "use open OUT => ':utf8', ':std'" as a suitable
replacement. Rafael Garcia-Suarez explained that "-C" had received
this treatment as a consequence of a bug report, and that there was a
TODO item recorded that explained what needed to be done to have the
old behaviour reinstated (this time, without bugs).
get it right
http://rt.perl.org/rt3/Public/Bug/Display.html?id=34087
http://xrl.us/biu9v
Making pl2bat.pl more robust
Jan Dubois had been exploring approaches to make "pl2bat.pl", since
depending on whether an executable foo anything, as opposed to a
specific foo.bat, exists in the current directory or not, you can wind
up running something else that the program you thought you were going
to run. Spaces in path names add to the fun.
With a series of four fall-through tests, Jan thought he had all the
bases covered, and defied anyone to come up with a scenario that could
get past the last case and hit the "Can't find full path" error
message.
Jan also hoped that there were no longer any serious Perl users stuck
on Win9X.
going batty
http://xrl.us/biu9x
Patches of Interest
Too much sv.c consting
Nicholas Clark undid the final batch of consting from last week that
Steven Schubiger had applied to sv.c. What is more, it was the second
time he had had to roll back the same change.
Unfazed, Stephen delivered the fourteenth patch in the series. Andy
Lester explained that he had considered doing what Steven is currently
doing, when he was on his consting quest, but had considered that the
added value wasn't worth the price.
http://xrl.us/biu9z
Borg parent.pm
Rafael Garcia-Suarez edited base.pm to recommend parent.pm in its
place, and ask Max "Corion" Maischein to release a new version of
"parent" to synchronise CPAN with blead. Some minor problems cropped
up that Jerry D. Hedden tidied up.
debased
http://xrl.us/biu93
configure.com and vms.c fixes
John E. Malmberg delivered some urgent patches to get VMS to build out
of the box. H.Merijn Brand handled part of the patch through the brand
new metaconfig infrastructure. Unfortunately this caused other
platforms to complain unnecessarily. H.Merijn wasn't sure if he was
doing something wrong, or whether the approach cannot work.
http://xrl.us/biu95
Documentation fix for perlfunc/shutdown
Paul Fenwick documented the return values of "shutdown" (whose
existence I had so far managed to ignore). Nicholas Clark tried to
improve Paul's prose and Paul came back with yet a better version.
Unapplied.
and turn the lights out
http://xrl.us/biu97
Solaris build warnings
Jerry D. Hedden found some warnings when building blead on Solaris.
Andy Dougherty traced this to arrival of the "PERL_BITFIELD16" typedef
and proposed a minimal patch that cast things together correctly. Or
at least silently. Unapplied.
http://xrl.us/biu99
Revising perlhack.pod
Leon Brocard updated the documentation to "perlhack" to take into
account the changes to perlmain.c, and some notes on using "valgrind".
Applied after a resend.
http://xrl.us/bivab
Do not use "SVTYPEMASK" to prevent cleaning of "PL_fdpid" and "PL_strtab"
As part of his ongoing Kurila experiment (a fork of the perl5
codebase), Gerard Goossen traced down some problems that exist in the
Perl array implementation.
Nicholas Clark used the information to provoke an assertion failure
when running a "make minitest". He applied Gerard's change, and
suggested that there are still some things that need to be sorted out.
http://xrl.us/bivad
New and old bugs from RT
"POSIX::SigSet" double free with threads (#36653)
Steve Peters took another look at this problem from 2005, and
discovered that while 5.8.8 ran without error, blead panicked with a
free from wrong pool.
ungood
http://xrl.us/bivaf
Segfault on @ISA push after symbol table delete (#52074)
Peter Scott posted a one-liner to make blead dump core. Nicholas Clark
noted that since 5.8.8 is immune, it must be a problem in the new MRO
code.
nice self-contained task
http://xrl.us/bivah
"Text::Wrap::wrap()" generates a segfault with Cyrillic characters when
the utf8 flag is turned on (#52104)
Frédéric Buclin filed a ticket that was turned arose in the
development of Bugzilla (heh). Nicholas reduced it to a simple matter
of a regexp using "pos()" inside a substitution, and tracing back,
realised that the bug had been lying in wait, ever since it checked
in, back in 2000.
It was then a simple matter to fix it in change #33580. As a bonus,
this should make it into 5.8.9-to-be. Frédéric wondered what he could
do in the meantime. Nicholas suggested one approach and Robin Barker
another. Aristotle Pagaltzis, who wrote the current implementation of
"Text::Wrap::wrap" (and built it for speed), suggested downgrading to
a previous version.
that wraps it up
http://xrl.us/bivaj
"POSIX::strftime" hangs on %Z under cygwin (#52126)
Eric Roode reported a bug in 5.8.8 that had previously been reported,
and fixed, with change #29350. Nicholas Clark announced that said
change was included in 5.8.9. Steve Peters showed how changing the
final argument to "strftime" prevented the process from hanging in the
meantime.
tis a matter of time
http://xrl.us/bivam
Build error on AIX (#52188)
Shanthi Muthu Srinivasan was unable to build 5.8.8 on AIX, the process
blowing up when trying to link "miniperl". Dominic Dunlop suggested
that it could be due to running out of disk space, and that
Activestate have an Activeperl distribution for AIX that may be worth
a try. H.Merijn Brand suggested a special incantation for 64-bit AIX.
No further information was forthcoming.
http://xrl.us/bivao
XS error in "Win32::Console::_WriteConsoleInput" (#52216)
Ed Sanman traced down a problem of console misbehaviour to what is
probably a case of bad cut'n'paste. Not confirmed.
watch your stack
http://xrl.us/bivaq
Perl5 Bug Summary
294 new + 1509 open = 1803 (+8)
http://xrl.us/bivas
http://rt.perl.org/rt3/NoAuth/perl5/Overview.html
New Core Modules
DB_File 1.817
Paul Marquess synchronised blead and CPAN.
http://xrl.us/bivau
This is the BBC
Regexp-Optimizer 0.15
Andreas König discovered, somewhat belatedly, that change #28868
broke Dan Kogai's "Regexp::Optimizer" package.
http://xrl.us/bivaw
In Brief
Nicholas Clark applied Dominic Dunlop's patches to tidy up failing
tests on PPC64.
http://xrl.us/bivay
Vincent Pit spotted an apidoc mismatch for "Perl_magic_clearhint" and
offered a patch, so Rafael applied it.
wild cut'n'paste
http://xrl.us/biva2
Niko Tyni spotted a nasty typo in "IO::Socket::INET" where a hash key
was named "upd" instead of "udp", and also offered a patch. H.Merijn
applied the patch. I wonder if we could have tested for that.
nice catch
http://xrl.us/biva4
chromatic tweaked the usage documentation for utils/dprofpp.PL.
Applied.
being consistent
http://xrl.us/biva6
Gerard Goossen also rejigged the MAD dump XML escape regexp, thus
fixing a pile of tests that used to fail when run madly.
http://xrl.us/bivbc
Yitzchak Scott-Thoennes thinks that "Devel::Cover" in the core would
be a marvellous idea.
http://xrl.us/bivbe
Nicholas Clark wrote some code to detect "Configure" test goofs and
H.Merijn Brand folded it into the configuration infrastructure.
http://xrl.us/bivbg
David Cantrell hopes that Dave Mitchell's perl 5.10.1 plan includes
Jarkko Hietaniemi's patch for "Encode::Alias".
20 houses down from the beast
http://xrl.us/bivbi
The bug tracking system thread attracted more discussion this week.
http://xrl.us/bivbk
Sérgio Durigan Júnior pinged the list about his problems when setting
the libraries' installation path.
http://xrl.us/bivbn
Jim Cromie had a small interesting task to do, something about
separate debug info.
didn't set the world on fire
http://xrl.us/bivbp
Lincoln Stein absorbed a "CGI::Carp" typo difference between blead and
CPAN.
http://xrl.us/bivbr
Peter Scott took a bash at glob in scalar context, offering a patch to
reset the iterator in certain circumstances. Yitzchak showed how such
a change would lead to even harder to debug problems.
an oldie but a goodie
http://xrl.us/bivbt
It would appear that change #26980, regarding an improvement to
"PERL_DONT_CREATE_GVSV", would be the reason for test failures in
Math-Pari.
http://xrl.us/bivbv
TODO of the week
A task that need a little C knowledge
Weed out needless PERL_UNUSED_ARG
The C code uses the macro "PERL_UNUSED_ARG" to stop compilers warning
about unused arguments. Often the arguments can't be removed, as there
is an external constraint that determines the prototype of the
function, so this approach is valid. However, there are some cases
where "PERL_UNUSED_ARG" could be removed. Specifically
* The prototypes of (nearly all) static functions can be changed
* Unused arguments generated by short cut macros are wasteful - the
short cut macro used can be changed.
Last week's summary
Actually, two weeks ago. Sam Vilain explained how one could go about
cherry-picking patches with git and promised to create a new version
of the repository in UTF-8, so that people with non-Latin-1 names may
be represented faithfully. (Hi Slaven!)
In mail to the list, I had complained about "foo" being rendered as
``foo'' in the summaries. It turns out that this is because at some
point I switched my POD-to-HTML conversion from blead's "pod2html" to
that of maint.
It turns out that lots of bugfixes went into the 5.10 release of
"Pod::Html", but as the module is not dual-lifed, it is difficult to
upgrade older perls. So I volunteered to take care of dual-lifing it.
Look for 1.09_01 on a CPAN mirror near you soon.
This Week on perl5-porters - 9-15 March 2008
http://xrl.us/bik5f
This Week on perl5-porters - 16-22 March 2008
http://xrl.us/bivbx
About this summary
This summary was written by David Landgren.
Weekly summaries are published on http://use.perl.org/ and posted on a
mailing list, (subscription: [EMAIL PROTECTED]). The
archive is at http://dev.perl.org/perl5/list-summaries/. Corrections
and comments are welcome.
If you found this summary useful, please consider contributing to the
Perl Foundation to help support the development of Perl.
