Perl 5.9.2 is out

[Rafael Garcia-Suarez]

This word flip was weird. Every recording date of McClintic's he'd
gotten into the habit of talking electricity with the audio men and
technicians of the studio. McClintic once couldn't have cared less
about electricity, but now it seemed if that was helping him reach a
bigger audience, some digging, some who would never dig, but all
paying and those royalties keeping the Triumph in gas and McClintic
in J. Press suits, then McClintic ought to be grateful to
electricity, ought maybe to learn a little more about it. So he'd
picked up some here and there, and one day last summer he got around
to talking stochastic music and digital computers with one
technician. Out of the conversation had come Set/Reset, which was
getting to be a signature for the group. He had found out from this
sound man about a two-triode circuit called a flip-flop, which when
it turned on could be one of two ways, depending on which tube was
conducting and which was cut off: set or reset, flip or flop.

``And that,'' the man said, ``can be yes or no, or one or zero. And
that is what you might call one of the basic units, or specialized
`cells' in a big `electronic brain.' ''

``Crazy,'' said McClintic, having lost him back there someplace. But
one thing that did occur to him was if a computer's brain could go
flip or flop, why so could a musician's. As long as you were flop,
everything was cool. But where did the trigger-pulse come from to
make you flip?

-- Thomas Pynchon, V

Perl 5.9.2 is currently propagating on CPAN. Tests, of course, are
welcome, including tests of CPAN modules (the Phalanx 100 to begin
with).

Thanks to everyone who has participated in it. Thanks also to everyone
who gave me the energy(*) to do it, either directly, or indirectly, by
giving me examples of hard work, success and talent.

I plan to make releases more often until 5.10.0. Basically, the next
milestones might be :

* new core modules (integration of CPANPLUS)
* lexical pragmas
* finish assertions
* implement the (_) prototype (i.e. "default to $_")
* implement state variables

The perldelta has already been prettified on search.cpan.org :

http://search.cpan.org/~rgarcia/perl-5.9.2/pod/perl592delta.pod

More formal announcement to follow.

Enjoy,
-- rgs

--
(*) an old word for "tuits".

[Dan Kogai]

Porters

On Apr 01, 2005, at 22:07, Rafael Garcia-Suarez wrote:
> Perl 5.9.2 is currently propagating on CPAN. Tests, of course, are
> welcome, including tests of CPAN modules (the Phalanx 100 to begin
> with).

When are you going to review Encode 2.0902?

http://www.dan.co.jp/~dankogai/Encode-2.0902.tar.gz

It has been on the table for a long while. I know Nicholas is busy but
isn't there anybody else to test it and review it? I am dying to
release 2.10.

At the same time I have been pretty busy Livedoor mumbo-jumbo so I took
advantage of this silence to respond to whole bunch of mass=media in
Japan.

http://news.google.com/?q=Livedoor

I could be the most (in)?famous open source programmer in Japan
(because of|thanks to) that....

Dan the (Perl5 Porter|Encode Maintainer)

[Tels]

-----BEGIN PGP SIGNED MESSAGE-----

Moin,

Rafael Garcia-Suarez <rgarci...@mandrakesoft.com> wrote:

>Perl 5.9.2 is currently propagating on CPAN. Tests, of course, are
>welcome, including tests of CPAN modules (the Phalanx 100 to begin
>with).

[snip]

Wow, from "5.9.2 soon" to "is out" while I was not looking - congrats!

I plan to send a patch shortly with updating BigInt to v1.76, which
includes as default loading of FastCalc. In addition, a second patch will
then integrate the nec. parts of FastCalc (basically the .xs, a .pm stub
and a few tests) into blead.[1]

I had hoped to bring this outbefore 5.9.2, but you were faster :) [0]

Best wishes,

Tels

[0]: Don't worry, I was ready to send it on tuesday, and then bigrat's
trap.t segfaulted and anything was delayed because I can't figure yet out
why it crashes...
[1] As discussed previously. I tinkered a few hours with Calc optionally
loading FastCalc, but XSLoader/Dynaloader defeated me. So now I changed
BigInt to first load FastCalc, then, if that fails, Calc. Advantages are:
* Code change to BigInt is minimal. No change to Calc at all.
* You can disable FastCalc with "-Dnoextensions=FastCalc" (or so) at perl
build time, and yet always install it afterwards and it will still work
in both cases.
* Can easily disabled at runtime with "use Math::BigInt lib => 'Calc';"

To stop FastCalc repeating lot's of Calc code, I also changed FastCalc to
load Calc behind the szenes and re-use the Perl code FastCalc does not
yet have an XS equivalent for.

Hope that made sense :)

- --
Signed on Fri Apr 1 20:22:32 2005 with key 0x93B84C15.
Visit my photo gallery at http://bloodgate.com/photos/
PGP key on http://bloodgate.com/tels.asc or per email.

"Eat, eat, eat, eat the delicious sandwich!" -- Elan the Bard (Order of
the Stick)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iQEVAwUBQk2TcHcLPEOTuEwVAQEtaAf+MtlAYQLHWqwE0Gly4CZ0cEcg4visFw2R
vOrYcOpAkbHhS0N/rMeUg4AWGfT8aJchKcmxs0FVZGnNP+thQ5x1G6nO6m4EYG6I
bmTSiop1N7x6WCnIwUEEMO3Y4XUTg/I2RZzA7+fk+LYR7V09Lz576HoFESpsW77f
3NP3jPxP7O6cGtcWy0fGCClnjhYFSqQsvNMi6OYe194yxzirt7qPOTPqeYIrar6K
uvrrz3nX2X/zs2sooGZMrn5cVkE9Xox2RzXbmLkTWB5Q2c7FTBc0NCwqY4ktXzfC
8pBPLwCPYr54WXugTTlSNcXK6r5A8PN0NH5kysWYO3E5yHpR64FrhA==
=NJ00
-----END PGP SIGNATURE-----

[Rafael Garcia-Suarez]

Dan Kogai wrote:
> Porters
>
> On Apr 01, 2005, at 22:07, Rafael Garcia-Suarez wrote:
> > Perl 5.9.2 is currently propagating on CPAN. Tests, of course, are
> > welcome, including tests of CPAN modules (the Phalanx 100 to begin
> > with).
>
> When are you going to review Encode 2.0902?
>
> http://www.dan.co.jp/~dankogai/Encode-2.0902.tar.gz
>
> It has been on the table for a long while. I know Nicholas is busy but
> isn't there anybody else to test it and review it? I am dying to
> release 2.10.

2.0902 integrated in blead by change #24231, and sorry for the delay.