Fayland and Perl Programming

PSGI and Plack

2009-10-23T22:24:00.002+08:00

The most exciting thing in Perl world today is PSGI

The homepage is http://plackperl.org/

I tried that today. there are some tips for Win32 user.
* Don't try to run under Standalone server, the alarm is not supported in Win32
* after install Plack (maybe notest install Plack), try install Plack::Server::AnyEvent or Plack::Server::ServerSimple
* at last, to know a bit of the Plack, cd the cpan build dir or download the tar.gz

C:\strawberry\cpan\build\Plack-0.9006-ZEaDbK\eg\dot-psgi>plackup -a Hello.psgi -s AnyEvent
Accepting requests at http://0.0.0.0:5000/

C:\strawberry\cpan\build\Plack-0.9006-ZEaDbK\eg\dot-psgi>plackup -a Hello.psgi -s ServerSimple
Plack::Server::ServerSimple: You can connect to your server at http://localhost:5000/
127.0.0.1 - - [23/十月/2009 22:19:15] "GET / HTTP/1.1" 200 11 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3 GTB5"
127.0.0.1 - - [23/十月/2009 22:19:15] "GET /favicon.ico HTTP/1.1" 200 11 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3 GTB5"

try visit http://127.0.0.1:5000/ then you'll get the access_log like above.

well, I think the Plack::Middleware::AccessLog should use &POSIX::setlocale(&POSIX::LC_ALL); on qw( .ISO8859-1 .ISO_8859-15 .US-ASCII .UTF-8 ) to avoid 十月 the Chinese.

you can try more with the all the .psgi files under eg/dot-psgi, and you might have some ideas to write a Middleware. :)

Thanks

(Updates on Oct 24th)
miyagawa applied the patch with some changes and it works pretty well now.
Thanks very much. :)

《Perl 语言入门（第五版）》

2009-10-23T09:49:00.000+08:00

《Perl 语言入门（第五版）》

Translated by chunzi, joe.jiang and cnhackTNT

:)

Canon IXUS 95 IS

2009-09-26T14:47:00.000+08:00

I bought a digit camera today: Canon IXUS 95 IS (Chinese).

not so bad. :)

OpenSocialX::Shindig::Crypter long story

2009-09-17T16:51:00.000+08:00

The long story is

we (Zorpia) were developing OpenSocial in last days. The first choice to build a opensocial container is Shindig.

unlucky, it ONLY has PHP and Java version. yeah, I do all my code in Perl, I know nothing about PHP or Java. (well, I know a little PHP now because I was writing Service package with PHP in last days, and I must say, PHP sucks.)

lucky, the Shindig author points me to an URL: Using Shindig in a non PHP or Java environment
that I can write the 'Application' section in Perl. It IS great, really very great.

so at last, I google around and wrote my last solution: OpenSocialX::Shindig::Crypter and the php lib BasicBlobCrypter.php

and now I'm happy to share it with your guys. if you have the same situation with me, you can try that module and tell me your feeling. :)

Thanks. (OK, it's not so long.)

OpenSocialX::Shindig::Crypter short story

2009-09-17T10:07:00.001+08:00

the short story is that I released the OpenSocialX::Shindig::Crypter yesterday.

the short story is if you want to find a way that encrypt data by Perl and decrypt the encrypted data by PHP, you come to the right place.

Perl: crypt.pl

my $str = 'o=1&v=3&p=5';
my $cipher = Crypt::CBC->new( {
    'key' => 'length16length16',
    'cipher'=> 'Rijndael',
    'iv' => '1234567890abcdef',
    'literal_key' => 1,
    'padding' => 'null',
    'header' => 'none',
    keysize => 128/8
} );
my $encrypted = $cipher->encrypt($str);
print "encrypted: ".encode_base64($encrypted)."\n";
print "decrypted: ".$cipher->decrypt($encrypted)."\n";

my $hmac = Digest::SHA::hmac_sha1($encrypted, 'hmackey');
print "hmac: " . encode_base64($hmac) . "\n";
print "total: " . encode_base64($encrypted . $hmac) . "\n";

PHP: crypt.php

$str = 'o=1&v=3&p=5';
$encrypted = mcrypt_cbc(MCRYPT_RIJNDAEL_128, 'length16length16', $str, MCRYPT_ENCRYPT, '1234567890abcdef');
$decrypted = mcrypt_cbc(MCRYPT_RIJNDAEL_128, 'length16length16', $encrypted, MCRYPT_DECRYPT, '1234567890abcdef');
echo "encrypted: ".base64_encode($encrypted)."\n"; 
echo "decrypted: ".$decrypted."\n"; 

$blabla = hash_hmac('sha1', $encrypted, 'hmackey', 1);
echo "hmac: " . base64_encode($blabla) . "\n";
echo "total: " . base64_encode($encrypted . $blabla) . "\n";

both of them show the same output:

# perl crypt.pl
encrypted: rgPZe6swWb3t3yIoIZFQaQ==
decrypted: o=1&v=3&p=5
hmac: WLTGeyesYTUhwg6fjVqDudrKRz0=
total: rgPZe6swWb3t3yIoIZFQaVi0xnsnrGE1IcIOn41ag7naykc9
# php crypt.php
encrypted: rgPZe6swWb3t3yIoIZFQaQ==
decrypted: o=1&v=3&p=5
hmac: WLTGeyesYTUhwg6fjVqDudrKRz0=
total: rgPZe6swWb3t3yIoIZFQaVi0xnsnrGE1IcIOn41ag7naykc9

I'll post a long story later today.

Thanks.

use Moose; or no Moose;

2009-09-03T17:38:00.005+08:00

recently "Moose or no Moose" is spreading to the Perl world.

Moose or No Moose
Command Line Moose or Knowing what You Don't know.
Re: Moose Or No Moose
Take arms against a sea of troubles
Re: Re: Moose Or No Moose
Moose Startup Time over Time

basically I agree with Adam Kennedy that Array::Compare shouldn't based on Moose. Test::Warn is a popular module and Array::Compare without Moose can totally ease the install of the Test::Warn.

in the meantime, I would agree with Stevan Little that we shouldn't hold the usage of the Moose. it's a great module with great idea and great contributors. I use it in my daily life and it works great. once more and more people/module use Moose, the problem wouldn't become a problem.

I think Padre should use Moose, I know Adam would reject that, but I insist on that Padre will benefit from Moose even Class::XSAccessor is the fastest accessor module now. Padre will be in a better structure and much more easy to read if it's coded with Moose. Plugin/Document/Highlighter would be easily written and plugged.

another thing is that I met the problem before. Long time ago people rate down WWW::Contact because it uses Moose. but I think it's not true. Moose helps me to build the module.

at last and in short, I think simple module (like only use the accessor of Moose) shouldn't use Moose but large software/module like Padre should use Moose.

Thanks. just personal thoughts.

life updates

2009-08-24T19:06:00.002+08:00

I'm sick these days. 1 day fever, 4 days cough, 3 injections. very bad.

my modem got broken yesterday. it's not bad. I updated my book a little when there is no internet access. have a check if you haven't. :)

http://search.cpan.org/dist/Book-Chinese-MasterPerlToday/

I hope everything is fine tomorrow. God bless.

Book::Chinese::MasterPerlToday

2009-08-07T10:24:00.003+08:00

some of you may be noticed that I'm writing a free perl book in Chinese.

yeah, I released it as a CPAN module yesterday: Book::Chinese::MasterPerlToday

it's not finished yet but I promise I will. so have fun and stay tuned.

patches or suggestions are welcome. :)

The Definitive Guide to Catalyst

2009-07-26T23:21:00.002+08:00

The Definitive Guide to Catalyst: Writing Extensible, Scalable and Maintainable Perl–Based Web Applications

My boss (Jeffrey Ng) just brought this book for me (thanks to my college Calvin getting it for me from USA).

I haven't read it yet, but I will. it's a good book. I'll post a review once I read it through.

:)

Beijing Perl Workshop - Sep 19, 2009

2009-07-22T12:08:00.001+08:00

http://conference.perlchina.org/bjpw2009/

I'm afraid I will not go there, but it's still recommended to attend. :)

Thanks.

a module started with O

2009-07-11T14:21:00.002+08:00

ORDB::Unihan - An ORM for the published Unihan database

a mixin of ORLite and Unihan database. the original unihan database is txt based, and this one is sqlite.
even if you don't want to use ORLite, you still can use the Unihan.sqlite by ORDB::Unihan->sqlite_path();

bla bla, not sure how useful it is. just for fun.

Enjoy

QueryLog support for Catalyst::Model::DBIC::Schema

2009-06-17T13:12:00.002+08:00

well, I wrote Catalyst::Model::DBIC::Schema::QueryLog before. it's fine. it works.

mst released the new Catalyst::Model::DBIC::Schema today. built on top of Moose, with two traits: Caching and Replicated.
It reminds me to write Catalyst::TraitFor::Model::DBIC::Schema::QueryLog
trait is great. easy to read, configurable and sane.

Enjoy. Thanks

DBIx-Class talk

2009-05-27T11:08:00.001+08:00

I just gave a DBIx-Class talk on the annual meeting of Zorpia 1 hour ago. it's not a success one, but I still want to share with you. check:

http://www.fayland.org/talks/2009_Zorpia_DBIx_Class/dbix-class.xul

Enjoy, Thanks.

manage Task::Padre::Plugins with ORDB::CPANTS

2009-05-24T07:48:00.004+08:00

To update a Task:: module is not so hard but trivial. you need find out the new version of existing modules, and those haven't added yet.
well, do it manually is not so Perl-ish. we have Perl, we should do it automatically by running a script.

it's pretty good we have ORDB::CPANTS from Alias.
1, we get all plugins by ORDB::CPANTS through used_by
2, split them into requires and feature.
3, get the META info through YAML::LoadURI
4, generate the main part of Makefile.PL and Plugins.pm

you can find the code here: http://svn.perlide.org/padre/trunk/Task-Padre-Plugins/auto_generate.pl

I know I can generate Makefile.PL and Plugins.pm directly, but it's good enough for now.

Enjoy.

Xi'an trip next week

2009-05-22T18:07:00.002+08:00

We (Zorpia people) will hold an annual meeting next week in Xi'an.
It would be a lot of fan that I'll play basketball during the trip. 24 people from 4 different countries/regions.

if you're nearby, please drop by if you like it.

Thanks

github instead of google code

2009-05-19T11:42:00.002+08:00

yes. I do love github more than google code. it's not the problem of git or svn. svn is fine.

the point is participation.
In google code, it's very hard to contribute some code to one project if you're not the author. it disappoints me a lot.
The "fork" button in github is very helpful. people forked my repos then pull requests to ask me to merge (like perl-net-github, dist-zilla-plugin-repository). I forked someone's repos then pull back (like theschwartz-moosified, dist-zilla, git-pureperl etc).

it's awesome. go start using github right now. :)

SuperCollider Programming

2009-05-19T09:53:00.001+08:00

what the hell is SuperCollider Programming?

I'm glad that my blog is "Fayland and Perl Programming".

also I hate that blogger.com is blocked in China again. STUPID.

DBIx::Class::ResultSet::Void and TheSchwartz::Moosified

2009-05-12T06:28:00.002+08:00

Big thanks to ribasushi, the new DBIx::Class::ResultSet::Void 0.04 is using
# SELECT 1 FROM item me WHERE ( me.id = ? ) LIMIT 1: '1'
# INSERT INTO item ( id, name) VALUES ( ?, ? ): '1', 'A'
instead of COUNT(*). it's much faster now.

Another big thanks to stash from socialtext, TheSchwartz::Moosified 0.04 with PostgreSQL support and bug fixes.

Enjoy!

DBIx::Class::ResultSet::Void

2009-05-10T20:13:00.003+08:00

some time I'm really feeling sick when I use update_or_create or find_or_create, the default DBIx::Class::ResultSet return DBIx::Class::Row object by using find.
it's useful when I need the $row object, but it's dummy if I don't care.

I wrote DBIx::Class::ResultSet::Void today. It would use count instead of find if not defined wantarray. it means when I call
$rs->find_or_create( { id => 1, name => 'A' } );
instead of
my $row = $rs->find_or_create( { id => 1, name => 'A' } );
it would generate SQLs like:
# SELECT COUNT( * ) FROM item me WHERE ( me.id = ? ): '1'
# INSERT INTO item ( id, name) VALUES ( ?, ? ): '1', 'A'
it's a little better I think.

read more on the POD and Enjoy!

what am I looking forward today

2009-05-07T19:45:00.002+08:00

The Definitive Guide to Catalyst: Writing Extendable, Scalable and Maintainable Perl–Based Web Applications by Kieren Diment and Matt Trout.

I don't know which month will it be released, but I'm really looking forward to it. :)

Thanks.

irc log

2009-05-05T19:48:00.002+08:00

I didn't know that keep a public irc log is wrong until mst pointed it to me today.
Sorry for any trouble caused and it's private now.

Thanks.

new DayDayUp

2009-04-29T06:34:00.002+08:00

inspired by jrock's Multimethods and Unshortening URLs with Modern Perl, I decided to rewrite my DayDayUp CPAN module.

it's really very cool to write code with KiokuDB and MooseX::Declare. less code. more focus on real.
I'm planing to write more, but it's for now.

even it fails on Win32, it's still worth to try. :)
Enjoy

planet you can't miss today

2009-04-28T08:03:00.001+08:00

http://ironman.enlightenedperl.org/

Enjoy

CPAN updates

2009-04-21T15:50:00.002+08:00

I spent lots of time to write CPAN module in past two days. it's not so fun but enjoyable.

App::GitHub - 8 dists 3 days.
Dist::Zilla::Plugin::MetaResources - DEPERCATED, it will be supported in Dist::Zilla core.
Dist::Zilla::Plugin::Repository
Net::GitHub - also 8 dists 3 days.

To write a CPAN module is really not so hard. please go on and enjoy it.

Thanks.

App::GitHub

2009-04-20T20:25:00.002+08:00

well, App::GitHub is a command line tool wrapped by Net::GitHub.

for a brief view:
$> github.pl

Welcome to GitHub Command Tools! (Ver: 0.04)
Type '?' or 'h' for help.

github> ?
command argument description
repo :user :repo set owner/repo, eg: 'fayland perl-app-github'
login :login :token authenticated as :login
loadcfg authed by git config --global github.user|token
?,h help
q,exit,quit exit

Repos
rshow more in-depth information for the :repo in repo
rlist list out all the repositories for the :user in repo
rsearch WORD Search Repositories
watch watch repositories (authentication required)
unwatch unwatch repositories (authentication required)
fork fork a repository (authentication required)
create create a new repository (authentication required)
delete delete a repository (authentication required)
set_private set a public repo private (authentication required)
set_public set a private repo public (authentication required)
network see all the forks of the repo
tags tags on the repo
branches list of remote branches

Issues
ilist open|closed see a list of issues for a project
iview :number get data on an individual issue by number
iopen open a new issue (authentication required)
iclose :number close an issue (authentication required)
ireopen :number reopen an issue (authentication required)
iedit :number edit an issue (authentication required)
ilabel add|remove :num :label
add/remove a label (authentication required)

File/Path related
cd PATH chdir to PATH

Others
rshow :user :repo more in-depth information for a repository
rlist :user list out all the repositories for a user

github> loadcfg

github> repo fayland perl-app-github

fayland/perl-app-github> rshow
{
"owner" : "fayland",
"private" : false,
"name" : "perl-app-github",
"description" : "App::GitHub CPAN module",
"homepage" : "",
"watchers" : 2,
"forks" : 0,
"fork" : false,
"url" : "http://github.com/fayland/perl-app-github"
}

fayland/perl-app-github> repo fayland sandbox2

fayland/sandbox2> ilist
[
{
"number" : 1,
"position" : 1,
"state" : "open",
"body" : "Test Issue body.",
"created_at" : "2009/04/19 06:08:30 -0700",
"updated_at" : "2009/04/19 06:09:17 -0700",
"user" : "fayland",
"title" : "Test Issue",
"votes" : 0
},
{
"number" : 2,
"position" : 2,
"state" : "open",
"body" : "new test 3\nnew test 2",
"created_at" : "2009/04/19 18:44:49 -0700",
"updated_at" : "2009/04/19 18:50:44 -0700",
"user" : "fayland",
"title" : "new test 2",
"votes" : 0
},
{
"number" : 3,
"position" : 3,
"state" : "open",
"body" : "s",
"created_at" : "2009/04/19 18:51:17 -0700",
"updated_at" : "2009/04/19 18:51:31 -0700",
"user" : "fayland",
"title" : "sssssssss",
"votes" : 0
}
]

fayland/sandbox2> q

for now, there are ONLY repos and issue related command. but I would add more in next few days.

Thanks and Enjoy.