Atom feeds with PHP 5 Dom and XSL January 13, 2008

Posted by idimmu in php.
All blogs require silly amounts of feed generators, right? And this is a silly blog so requires a silly generator. The entire site is written using PHP5, and my automagic 'datahandler' activepage concept creates an XML document using DOM that then uses XSL as a templating engine, so I figured it wouldn't be too hard to knock up a stylesheet to turn the default datahandler for the blog in to a nice atom feed! Just make sure you set the content-type to application/atom+xml when generate the page!


<?xml version="1.0" encoding="iso-8859-1"?>

<xsl:stylesheet version = "1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">


<xsl:output indent="yes" method="xml" encoding="iso-8859-1" omit-xml-declaration="yes" doctype-system="http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd" doctype-public="-//W3C//DTD XHTML 1.1//EN" />

<xsl:template match="page">
<feed xmlns="http://www.w3.org/2005/Atom">
<link rel="alternate" type="text/html" href="http://www.idimmu.net/" />
<link rel="self" href="http://www.idimmu.net/blog/atom.php" />

<title>idimmu . net</title>
<link href="http://www.idimmu.net/"/>
<updated><xsl:value-of select="datahandler_blog/blog_list/blog/date/year"/>-<xsl:value-of select="datahandler_blog/blog_list/blog/date/month"/>-<xsl:value-of select="datahandler_blog/blog_list/blog/date/day"/>T<xsl:value-of select="datahandler_blog/blog_list/blog/date/hour"/>:<xsl:value-of select="datahandler_blog/blog_list/blog/date/minute"/>:<xsl:value-of select="datahandler_blog/blog_list/blog/date/second"/>Z</updated>
<author>
<name>idimmu</name>
</author>
<id>http://www.idimmu.net/</id>
<xsl:apply-templates select="datahandler_blog"/>
</feed>
</xsl:template>

<xsl:template match="datahandler_blog">
<xsl:apply-templates select="blog_list"/>
</xsl:template>

<xsl:template match="blog">
<entry>
<title><xsl:value-of select="title"/></title>
<link href="http://www.idimmu.net/{clonefakeurl}"/>
<id>http://www.idimmu.net/<xsl:value-of select="clonefakeurl"/></id>
<updated><xsl:value-of select="date/year"/>-<xsl:value-of select="date/month"/>-<xsl:value-of select="date/day"/>T<xsl:value-of select="date/hour"/>:<xsl:value-of select="date/minute"/>:<xsl:value-of select="date/second"/>Z</updated>
<content type="xhtml">
<div xmlns="http://www.w3.org/1999/xhtml">
<xsl:value-of select="bb_content" disable-output-escaping="yes"/>
</div>
</content>
</entry>
</xsl:template>

<xsl:template match="blog_list">
<xsl:apply-templates select="blog"/>
</xsl:template>

</xsl:stylesheet>

PHP Java Bridge in Ubuntu Gutsy with Lucene January 12, 2008

Posted by idimmu in linux, php.
The php/java bridge it a pretty awesome little protocol that basically lets us use java classes inside our own PHP applications! This lets you harness the awesome power of all the Java libraries that exist, including the popular Lucene search engine library.

I referenced two excellent blog entries here and here whilst implementing Lucene search for this blog, but I am writing up the experience anyway to compare issues and difficulties and enhance my understanding of the process.

To start with Java, Lucene and the bridge dependancies must be installed (remember to enable multiverse in your apt sources)


apt-get install sun-java6-jre sun-java6-jdk liblucene-java libitext-java
update-java-alternatives -s java-6-sun


Grab the php-java-bridge deb package from sourceforge and install it. The fact it is v4 does not reflect that it is only for PHP version 4! There are RPMs for version 5 which you could turn in to a deb package using alien but at the moment I am feeling lazy so I will see how version 4 works out first.


wget http://downloads.sourceforge.net/php-java-bridge/php-java-bridge_4.3.0-1_i386.deb
dpkg -i php-java-bridge_4.3.0-1_i386.deb


Apache should restart now, if not restart it yourself.

To check that it is working look at the output of phpinfo(), there should be a new shiny java section! Listing the running processes also is interesting!


root 20205 0.0 0.7 664520 15520 ? Sl 17:18 0:00 java -Djava.library.path=/usr/lib/php5/20060613+lfs
-Djava.class.path=/usr/lib/php5/20060613+lfs/JavaBridge.jar -Djava.awt.headless=true
-Dphp.java.bridge.base=/usr/lib/php5/20060613+lfs php.java.bridge.Standalone LOCAL:@java-bridge-4ee9 1


as does netstat


unix 2 [ ACC ] STREAM LISTENING 1913999 @java-bridge-4ee9


I think it gets started when apache starts, as java.so is loaded in to the PHP, I'm still investigating that.

As far as starting the Lucene development goes, this was a pretty good tutorial on how it all works and this site has some good Java example code that I used to work out how the PHP should work.

Below is my PHP Lucene test code, it just creates one document with a description then searches the index description for 'idi test' and outputs the match. It's pretty rad!


<?php

java_require('/usr/share/java/lucene.jar');

$analyzer = new Java('org.apache.lucene.analysis.StopAnalyzer');
$writer = new Java('org.apache.lucene.index.IndexWriter', '/path/to/store/lucene/data/in', $analyzer, true);

$doc = new Java('org.apache.lucene.document.Document');
$field = new Java('org.apache.lucene.document.Field','description','idi data test',true, true, true);
$doc->add($field);

$writer->addDocument($doc);

$writer->close();

$indexer = new Java('org.apache.lucene.search.IndexSearcher','/path/to/store/lucene/data/in');
$parser = new Java('org.apache.lucene.queryParser.QueryParser','description',$analyzer);
$query = $parser->parse('rus test');

$hits = $indexer->search($query);

for ($i = 0; $i < $hits->length(); $i++) {
$found = $hits->doc($i);
print $i.".".$found->get('description');
}
?>


Now that it's working I just have to incorperate it in to the site :)

Copying files between servers with netcat and tar January 10, 2008

Posted by idimmu in linux.
One of the quickest ways (faster than scp at any rate) of copying a large number of files between 2 servers is by abusing the awesome powers of Linux's pipeing and netcat and tar!

Basically we set up netcat listening on the server you want the files copied too which pipes it's output to tar which extracts anything sent to it.


root@tanglefoot:/exports/archive# nc -l -p 7878 | tar -xzf -


Then we set up tar on the server we want to copy from, make it create a tarball and pipe it through a netcat which connects to the other server!


fee /home/shared/people # tar -cz MC | nc -q 10 tanglefoot 7878


When the copy has finished the sending instance of netcat will then exit!

Using PowerDNS with PostgreSQL on Ubuntu Gutsy January 9, 2008

Posted by idimmu in linux.
We handle DNS for thousands of domains for our customers and whilst our existing solution worked it was very messy to maintain and work with so we decided to trial a new solution for our offices to see how it would perform. We wanted something that could be database driven for ease of maintenance and we were personally recommended PowerDNS, so we decided to trial that one first.

For the database we would normally go with MySQL but we wanted an instance of PostgreSQL to play with as we are considering moving our main platform to it at some point in the future.

Our DNS server is running on Ubuntu Gutsy and everything we need is fortunately in the repositories so installing it is as easy as:


apt-get install pdns-backend-pgsql pdns-doc pdns-recursor pdns-server postgresql postgresql-contrib postgresql-doc


After all the software is installed we need to tell PowerDNS to use our PostgreSQL server in /etc/powerdns/pdns.conf


launch=gpgsql
gpgsql-host=127.0.0.1
gpgsql-user=powerdns
gpgsql-password=password
gpgsql-dbname=powerdns


We then need to configure the database, tables and user permissions in PostgreSQL.

To create the user we must become a superuser which typically involves changing to the postgres unix user and taking advantage of the ident based authentication.


root@sambuca:~# su postgres
postgres@sambuca:/root$ psql
Welcome to psql 8.2.5, the PostgreSQL interactive terminal.

postgres=# CREATE USER powerdns WITH PASSWORD 'password';
CREATE USER


You can check the user has been created through the psql client too.


postgres=# select * from pg_shadow;
usename | usesysid | usecreatedb | usesuper | usecatupd | passwd | valuntil | useconfig
----------+----------+-------------+----------+-----------+-------------------------------------+----------+-----------
postgres | 10 | t | t | t | | |
powerdns | 16385 | f | f | f | md5e954fb1203f8da7392a0c7406f83d765 | |
(2 rows)


We then need to create and switch to the new database


postgres=# create database powerdns;
CREATE DATABASE

postgres=# l
List of databases
Name | Owner | Encoding
-----------+----------+----------
postgres | postgres | UTF8
powerdns | postgres | UTF8
template0 | postgres | UTF8
template1 | postgres | UTF8
(4 rows)

postgres=# c powerdns
You are now connected to database "powerdns".


The table structure is


create table domains (
id SERIAL PRIMARY KEY,
name VARCHAR(255) NOT NULL,
master VARCHAR(128) DEFAULT NULL,
last_check INT DEFAULT NULL,
type VARCHAR(6) NOT NULL,
notified_serial INT DEFAULT NULL,
account VARCHAR(40) DEFAULT NULL
);
CREATE UNIQUE INDEX name_index ON domains(name);

CREATE TABLE records (
id SERIAL PRIMARY KEY,
domain_id INT DEFAULT NULL,
name VARCHAR(255) DEFAULT NULL,
type VARCHAR(6) DEFAULT NULL,
content VARCHAR(255) DEFAULT NULL,
ttl INT DEFAULT NULL,
prio INT DEFAULT NULL,
change_date INT DEFAULT NULL,
CONSTRAINT domain_exists
FOREIGN KEY(domain_id) REFERENCES domains(id)
ON DELETE CASCADE
);

CREATE INDEX rec_name_index ON records(name);
CREATE INDEX nametype_index ON records(name,type);
CREATE INDEX domain_id ON records(domain_id);

create table supermasters (
ip VARCHAR(25) NOT NULL,
nameserver VARCHAR(255) NOT NULL,
account VARCHAR(40) DEFAULT NULL
);

GRANT SELECT ON supermasters TO powerdns;
GRANT ALL ON domains TO powerdns;
GRANT ALL ON domains_id_seq TO powerdns;
GRANT ALL ON records TO powerdns;
GRANT ALL ON records_id_seq TO powerdns;


And then we can look at them!


powerdns=# d
List of relations
Schema | Name | Type | Owner
--------+----------------+----------+----------
public | domains | table | postgres
public | domains_id_seq | sequence | postgres
public | records | table | postgres
public | records_id_seq | sequence | postgres
public | supermasters | table | postgres
(5 rows)


After the user is created we need to edit /etc/postgresql/8.2/main/ph_hba.conf to grant that user access to the database from localhost


host powerdns powerdns 127.0.0.0/16 md5


We then need to reload PostgreSQL for the changes to take effect.


root@sambuca:~# /etc/init.d/postgresql-8.2 reload


We then need to populate it with the important SOA and NS records. All the records take a creation date as a timestamp, so we also created a function to return the current timestamp.


create function epoch() returns int AS 'select extract(epoch from now())::int;';

insert into records (domain_id, name, type, content, ttl, prio,change_date) values (1, 'btn.com', 'NS', 'dnsserver.btn.com',600,10,epoch());

insert into records (domain_id, name, type, content, ttl, prio,change_date) values (1, 'btn.com', 'SOA', 'dnsserver 2005091301 10800 3600 604800 600',600,10,epoch());

insert into records (domain_id, name, type, content, ttl, prio,change_date) values (1, 'dnsserver.btn.com', 'A', '10.0.0.1',600,10,epoch());


Now all we need to do is edit /etc/resolv.conf to use the new nameserver


nameserver 10.0.0.1


and check that it works!


rus@absinthe:~$ host dnsserver.btn.com
dnsserver.btn.com has address 10.0.0.1

Configuring Tomcat 5.5 and Apache 2 with mod_jk January 9, 2008

Posted by idimmu in linux.
mod_jk is a conduit between a web server and Tomcat, it supports a variety of web servers including IIS. Using mod_jk to put Apache in front of Tomcat lets you use all the power of Apache (caching, gzip, mod_rewrite, etc) whilst at the same time serving content from Tomcat, also with Ubuntu it's really easy to set up!

First of all install the software, you will need to enable the backports repository on Dapper for this.


apt-get install sun-java6-bin sun-java6-jdk tomcat5.5 libapache2-mod-jk


The Tomcat 5.5 that comes with Ubuntu already has an AJP connector configured on port 8009 so there is no additional configuration to do to it's server.xml file.

We then need to configure a worker.properties file for Apache2 which tells it about the Tomcat instance, I make mine in /etc/apache2/worker.properties


worker.list=idimmu
worker.idimmu.type=ajp13
worker.idimmu.host=localhost
worker.idimmu.port=8009


Make sure mod_jk is then enabled with a2enmod jk (it probably already is).

And finally we tell Apache2 about the worker instance in /etc/apache2/apache2.conf


JkWorkersFile /etc/apache2/workers.properties
JkLogFile /var/log/apache2/mod_jk.log
JkLogLevel info
JkMount /* idimmu


This will direct any requests to the Apache2 server to the Tomcat server
  1  2  3  4  5  6  7  8  9 

Tags

Friends

twitter

    lastfm

    • Therapy? – A Moment Of Clarity
    • Therapy? – Unbeliever
    • Therapy? – Die Laughing
    • Finger Eleven – Swallowtail
    • Equilibrium – Mana
    • Equilibrium – Dämmerung
    • Equilibrium – Ruf in Den Wind
    • Equilibrium – Des Sängers Fluch
    • Equilibrium – Die Weide und der Fluß
    • Equilibrium – Heiderauche

    IdleRPG Stats

    • 1 Stu| 44
    • 3 HRH_H_Crab 43
    • 8 Jeekay 43
    • 12 Appocomaster 43
    • 27 Kumquatt 36
    • 36 idimmu 17