Happy coding

Theese days I’ve been following a thread in the php-es mailing list at lists.php.es. A subscriber noticed that static class attributes weren’t shared between different page requests. Someone explained him that the way PHP deals with the requests is the cause for it not to have an “application” variable scope.

In Java servlets, for example, static class attributes are shared by all the instances of that class in the virtual machine. Scalability (and speed) at a medium scale is achieved using object in-memory caches. In PHP that approach can’t be done without using shared memory management libraries, and never in an easy way. Developers have to resort to disk and database caches, much slower that the memory ones.

Moreover, the fact of having to parse in each request not only the target PHP file, but also all those ones recursively included, represent another of the disadvantages of the platform.

It’s strange because, as far as I know, an “execution environment” common to all the requests does exist on other languages, like Perl (Apache mod_perl) or Python. That way, data can be shared among the requests, memory caches can be done, and so on.

Something that would be nice, even being a daydream, would be to develop an “application server” consisting of:

• A library that made shared memory usage simpler for storing session variables and object instances on it (even singletons), making them available from different threads.
• A multithreaded web server implemented entirely in PHP using the socket API. The server would spawn threads with each request using the threads and processes API. It’d be a replacement for Apache.

This way, we would have the advanteges of the application servers and the syntax simplicity of PHP. The disadvantage would be that, as PHP has been developed with the “parse, create environment, process the request and dispose everything” model in mind, with the proposed approach, a great number of memory leaks would be found. That memory leaks are currently hidden thanks to the short life cycle of the requests.

That’s all. This was an idea I had some time ago and it seemed interesting to me. Now I’ve decided to share it with you. It’s surprising that nobody has already thought about it, isn’t it?

UPDATE:

I’ve just found Quercus, a PHP5 implementation written in Java that allows PHP programs to interact directly with Java libraries, taking advantage of all the benefits that Java caches, connection pools, etc. can provide. It’s not the same approach I propose, but maybe a good one anyway.

Today I’ve needed to match strings in Java against a pattern similar to the ones used in filename matching. Java already has a String.matches() method to test against a regular expression, but hasn’t one to match against more limited filename-like patterns (* and ? wildcards).

At first, I tought about implementing the matching by hand, comparing character to character, but soon found a quickier and simpler approach: transform the pattern into a regular expression. Maybe this source code could be useful for you:

    private boolean matchFilter(String sample, String filter) {
        if (sample==null || filter==null) return true;

        StringBuffer f=new StringBuffer(".*");
        
        for (StringTokenizer st=new StringTokenizer(
                filter,"%*",true);
                st.hasMoreTokens();) {
            String t=st.nextToken();
            if (t.equals("?")) f.append(".");
            else if (t.equals("*")) f.append(".*");
            else f.append(Pattern.quote(t));
        }
        f.append(".*");
        
        return sample.matches(f.toString());
    }

It’ll soon be a year for me working in a project that uses Tapestry as one of its core technologies. I’ve learnt many interesting things and also faced some challenges to do certain specific stuff. Internet has always been a good help tool, and now it’s time to give back my little bit.

Tapestry features the class RedirectException to break the execution at any time and bring the user to the URL you want. But that exception expects an URL, and if you want to reach an existing Tapestry page, you need the corresponging URL for it.

Fortunately, this post in the Tapestry wiki at apache.org showed me how to get that URL for Tapestry3. But I also needed that exception to work for Tapestry4, so I had to investigate how to do the same think using the HiveMind registry. Finally, I got the class working and I’ve contributed it to the wiki where I found the Tapestry3 one.

I hope this to be useful!

As said in their web, Linden Labs has dual licensed their Second Life client both proprietary and GPL. But that isn’t everything: a new Linux client has been developed, so we all can enjoy the experience, and there are even people talking about using Mono as a scripting engine.

I have tried Second Life, and the least I can tell you is that it’s simply amazing. You can create objects and animate them writing scripts, perform all kind of actions (riding bikes, driving cars, talking, listening music, watching videos (ok, only through QuickTime in Windows), buying and selling objects, meeting people… in summary living a life. You should give it a try.
As a sample, I show you this picture, where my avatar is editing her motorcycle and seeing the source code of the script that drives it (click to enlarge):

Hibernate is a very useful framework to do Object-Relational mapping in Java. It saves a lot of time, gives the programmer a lot of flexibility and allows you to do amazing things like mapping object inheritance.

Nevertheless, it has some “strange” behaviours that, in my opinion, seem a bit courterintuitive. I’m going to tell you about the way Hibernate translates object relations into joins.

Let’s imagine we have the class Person, with attributes id and name, and the class Car, with attributes id and brand. Class Person has a one-to-one relationship with Car named car, and there’ll be persons who have a car, and persons who haven’t.
If we want to query about the cars each people have, the “intuitive” query would be this:

select p.name, p.car from Person p

And the “intuitive” result would be: (John, Car instance #1); (Peter, null); (Mary, Car instance #2), isn’t it?

It isn’t. The real thing is that Hibernate translates the HQL sencence to a SQL one that selects from Person and does an (inner) join with Car, in order to instantiate a Car with all its attribute values set up. As a consequence of that inner join, all the people having a null car are missing from the result.

Intuitive solution: force the outer join by hand

select p.name, c.id, c.brand from Person p left join p.car c

Non intuitive (but valid) solution: use the required attributes directly

select p.name, p.car.id, p.car.brand from Person p

In this last query, I would expect a NullPointerException when the person has no car, since if there is no car, there is no attribute to evaluate, but the exception doesn’t get thrown, the problem doesn’t arise and you get the work done.

Quite strange, isn’t it?

Until a couple of days ago, I used to use Gnome network-admin’s ability of having multiple network profiles to configure my wifi card at work and at home. But yesterday, the program went crazy and started to crash without any logical reason and I decided to shift to another solution: logical network interface mapping.

Interface mapping is a feature of ifupdown (/etc/network/interfaces). You can define some logical interfaces and call a script in order to choose which one will be mapped to a physical interface. The /etc/network/interfaces could be like this one (being HomeNET and WorkNET two existing ESSID network identifiers):

auto eth1
mapping eth1
script /root/WIFI/wldetect.sh
map HomeNET HomeNET
map WorkNET WorkNET

iface HomeNET inet static
address 192.168.1.2
netmask 255.255.255.0
gateway 192.168.1.1
wireless-essid HomeNET
wireless-key s:mysecretpass1 open
dns-nameservers 127.0.0.1 192.168.1.1

iface WorkNET inet dhcp
wireless-essid WorkNET
wireless-key s:mysecretpass2 open
dns-nameservers 127.0.0.1 192.168.100.1

I’m using resolvconf to activate the dns-nameservers directive in the interfaces file, and dnsmasq to take care of some dns issues with a couple of VPNs I use. That explains the first 127.0.0.1 entry.

The script /root/WIFI/wldetect.sh lists the available networks and chooses one of them being in a whitelist (HomeNET|WorkNET):

#!/bin/sh

# Config
WL_IFACE=`iwconfig 2>/dev/null | { read A _; echo $A; };`

# Reset the interface
ifconfig $WL_IFACE down
ifconfig $WL_IFACE 0.0.0.0
ifconfig $WL_IFACE up

# Search networks
NETWORKS=`iwlist $WL_IFACE scanning | grep ESSID | sed -e ‘s/.*”(.*)”/1/’`

# Bring down the interface
ifconfig $WL_IFACE down

# Select preferred networks
for NET in $NETWORKS
do
case $NET in
HomeNEt|WorkNET)
echo $NET;
exit 0;
;;
esac
done
exit 1;

And that’s all! I hope that this configuration could be helpful for someone.

I’ve just bought the webcam in the title of this post at Alcampo and had a hard time trying to set it working.

In the pkg-spca5xx-devel Debian mailing list, Dmitry Semyonov says that this camera model is supported by the spca5xx driver, but its ID isn’t added to the driver source code yet. He’s published a patch, and I tried to apply it, but the patch utility complained. That was because the tabs would have been replaced with spaces or something. Then I discovered the “-l” option, which solved my problem and I could finally apply the patch.

At help.ubuntu.com I found detailed instructions about how to build the spca5xx source code, so, finally I had the work done.

I’ve written a recipe to perform all the steps. At first, copy the patch into the file /tmp/qcexpress.patch. Next, execute these instructions:

apt-get install spca5xx-source
cd /usr/src
tar jxvf spca5xx-source.tar.bz2
cd modules
patch -lp1 < /tmp/qcexpress.patch
cd spca5xx
make
make install

That’s all!
And, at the end, a living proof: MSN video chat using Kopete 🙂

Yes, I’ve just discovered this fantastic program on my Ubuntu Dapper.

Podcasts are just like RSS text feeds, but feeding audio content files, and iPodder is just a client for them. With it, you can download the audio files for offline listening.

It’s a good tool to improve my English listening, isn’t it?