rainboxx - Matthias Dietrich

Clamp.js: "text-overflow" for paragraphs

2011-10-10T15:35:43Z

`text-overflow: ellipsis;` is great for text lines that may be longer then the parent element, eg. for shop article names or user names. But there is currently no solution for paragraphs that are longer than the parent. Of course you can set `overflow: hidden;`, but that will just clip the content/text. Today I stumbled upon a [blog entry by David DeSandro](http://dropshado.ws/post/1015351370/webkit-line-clamp) in which he show the use of `-webkit-line-clamp` which "*limits the number of lines of text displayed in a block element*". Sadly this is only possible within WebKit browsers (Safari and Google Chrome). Entering [Clamp.js](https://github.com/josephschmitt/Clamp.js). Clamp.js is a small Javascript that mimics this behavior in other browsers and sets the appropriate CSS for supported browsers. All you need is: $clamp(document.getElementById('clamp-this'), { clamp: 'auto' }); Works. Nice. Thank you, [Joe](http://reusablebits.com/post/2642059628/introducing-clamp-js).

Who is dreaming of such browser statistics?

2011-06-09T09:57:24Z

Let's see how long this will last ;-)

For what we are about to receive, may the Lord make us truely grateful.

2011-03-29T11:53:21Z

Nice grace:

via jitro.org

Auf Jobsuche? Software Engineer gesucht!

2011-03-29T08:24:38Z

(Click here for English version)

Wie man an diversen Stellen mitbekommen konnte, arbeite ich seit November letzten Jahres für ein Berliner Startup mit dem Namen UNUSUALS. UNUSUALS ist ein social network für die weltweite Werbefilmindustrie. Wir bauen momentan die Seite von Grund auf neu mit vielen weiteren und tollen Features.

Hierbei suchen wir als Unterstützung einen motivierten Software Engineer bzw. Web Developer zur Festanstellung, entweder in Ludwigsburg (bevorzugt) oder Berlin. Wenn du die Fähigkeit hast, schnell und mit dem Job zu lernen, mehrere Programmiersprachen beherrschst und deine Arbeit liebst, würden wir gerne mit dir sprechen.

Geforderte Fähigkeiten:

Erfahrung mit einer Scriptsprache wie Perl, Python oder Ruby
Erfahrung mit Datenbanken wie PostgreSQL oder MySQL
Erfahrung mit HTML, CSS und Javascript
Wissen, wie das Web funktioniert
Fähigkeit, ein Linux System zu bedienen
Fließend Deutsch oder Englisch sprechen und schreiben

Wünschenswerte Fähigkeiten:

Erfahrung mit Modern Perl wie z.B. Catalyst, DBIx::Class etc.
jQuery, HTML5, CSS3
Fähirgkeit, ein Linux System zu administrieren
Erfahrung mit Amazon Web Services wie EC2 und S3
Aktivitäten im "social web" (Facebook, Twitter, last.fm, ...)

UNUSUALS bietet dir eine Möglichkeit, in einem großartigen und wachsenden Startup zu arbeiten. Wir geben dir Raum zum Wachsen sowie die Möglichkeit, deine eigenen Ideen und deine Kreativität einzubringen.

Wenn du interessiert bist, sende bitte eine Bewerbung in Deutsch oder Englisch an jobs[at]unusuals.net. Beispiele deiner Arbeit, evtl. Code-Auszüge, sind sehr willkommen!

Need a job? Software Engineer wanted!

2011-03-24T15:40:47Z

(Zur Deutschen Version)

As you may already have noticed I'm working for a Berlin based start-up called UNUSUALS since November last year. It's a social network for the worldwide commercial film industry and we're doing a rewrite from scratch with tons of new features and awesome stuff!

For this to accomplish, we're searching a motivated Software Engineer/Web Developer to join us. The workplace will be in Ludwigsburg (preferably) or Berlin, which are both in Germany. If you're able to learn quickly on the job, pick up multiple programming languages and are passionate about your work, we'd like to talk to you.

Required skills:

Experience with a scripting language like Perl, Python or Ruby
Experience with databases like PostgreSQL or MySQL
Experience with HTML, CSS and Javascript
Knowledge of how web stuff works
Ability to use a Linux system
Fluent in speaking and writing German or English

Optional skills, which are a plus:

Experience with Modern Perl stuff like Catalyst, DBIx::Class and the like
jQuery, HTML5, CSS3
Ability to administrate a Linux system
Experience with Amazon Web Services like EC2 and S3
Active in social web (Facebook, Twitter, last.fm, ...)

Working at UNUSUALS is a big opportunity to work within an awesome and growing start-up. You'll have room for growth as well as the possibility to bring in your ideas and values.

Please send a copy of your resume in German or English to jobs[at]unusuals.net with some examples of work you've done and/or code samples.

Count Lines of Code

2011-03-02T13:32:45Z

Ever wanted to count the lines of code you've written in a project? I guess you have and you may came to a solution that was build by yourself. It worked, but not more. Today I stumbled upon CLOC -- Count Lines of Code -- that is a small Perl programm which does a fair good job at counting, recognizing comments, empty lines and so on. It also finds out in which language each file is written (by file extensions). There are also plenty of options to ignore folders (like .git or externals) or files by regex (like single Javascript libraries). A simple check looks like this: Check out this project on sourceforge and give it a try. Installing is painless: Just download the file and you're done! http://cloc.sourceforge.net.

Installing Perl modules made easy with cpanm

2011-02-27T12:20:51Z

Ever installed a Perl module on a Linux system? This is what a typical installation process looks like using the command line tool cpan:

While I can read and understand these lines, not every programmer -- in particular someone that is not using Perl on a regular basis -- cannot and is overextended by the massiv amount of text.

Since 2010 there is a really neat alternative to cpan called cpanminus. Excerpt from the docs:

cpanminus is a script to get, unpack, build and install modules from CPAN. Why? It's dependency free, requires zero configuration, and stands alone. When running, it requires only 10MB of RAM.

[...] For less experienced users (mostly from outside the Perl community), or even really experienced Perl developers who know how to shoot themselves in their feet, setting up the CPAN toolchain often feels like yak shaving, especially when all they want to do is just install some modules and start writing code.

So what's the difference? Less "line noise", easier, more lightweight!

If the module you're trying to install is already up to date it tells you this with only one line instead of 5 lines (ok, everyone should be able to read the 5 lines, too ;-)). Installing a module (here: Module::Build) gave me 5 lines of output while cpan has written a novel of 105 lines and using much more RAM and CPU.

I say give it a try by installing with the following single line on your terminal:

curl -L http://cpanmin.us | perl - --sudo App::cpanminus

Deleting a remote Git tag

2011-02-23T09:01:41Z

After uloading all my Perl modules to github I switched a local project from svn to git. After that I had an odd tag named trunk that I wanted to get rid of. gittower wasn't able to delete it.

I came to this solution:

git tag -d trunk
git push origin :refs/tags/trunk

The remote tag trunk was successfully deleted.

Mason 2 announced

2011-02-22T12:25:10Z

HTML::Mason, what I'm using since years as my favorite template system got an overhaul. Jonathan Swartz announced Mason 2 a few hours ago. Read the announcment here: http://www.openswartz.com/2011/02/21/announcing-mason-2/. Mason 2 got a top-level namespace on CPAN, is not completely based on Moose and already has a Catalyst view. What a nice way to start, eh?

Because I missed the recent development process I'm late to the party and just switching a new project from HTML::Mason to Mason.

New Perl module released: CatalystX::UriForStatic

2011-02-21T17:32:39Z

From the Readme:

Sometimes you are fetching static content like images, JavaScript code and CSS files from a different server that your Catalyst application is running on, let's say a CDN (content delivery network). For development, this often should not be the case because your local files may differ from the files on your CDN. Changing the URIs to the static content everytime you upload a new version of your application is not a solution and leads to mistakes.

Entering CatalystX::UriForStatic:

CatalystX::UriForStatic takes care of your static host name

CatalystX::UriForStatic either creates a local or production URI to your static files depending on your configuration! If your differences are that simple so you just can switch the host names, CatalystX::UriForStatic is for you.

As there are plenty of different ways how to integrate a CDN into one's own application, this module reflects the way I'm planning to implement the CDN. Suggestions for a broader range of implementations are very welcome, as well as patches and so on. If you like, hop over to github and fork the project. You're welcome :-).

Or get the latest stable fron CPAN at http://search.cpan.org/perldoc?CatalystX::UriForStatic, or with your command line:

sudo cpan CatalystX::UriForStatic

Oh, BTW about a week ago I released a new version of CloudApp::REST, too. Grab it, it fixes some issues.

Small things changed...

2011-02-21T16:54:42Z

Since almost a year I'm thinking of renovating my blog and switching to english blog posts then to have a broader audience I can offer my programming knowledge. I already made some redesign prototypes, from which one is really what I like. I found me thinking about new blog posts very often and then waiting for more time to make the switch and redesign first. It seems the time never came. So instead of still waiting I'm taking some time now to write this down in hope of writing some more posts soon ;-). You know: Hope Springs Eternal.

With this change came a new commenting system. I implemented DISQUS to have more control of the comments and better features (like subscriptions) for the users. Let's try how it works :-).

So far... Thanks for listening!

Announcement: CloudApp::REST version 0.01 published

2010-06-01T20:56:01Z

For a few weeks now this module lies within my workspace as almost finished, all I needed to do is doing some documentation fixes. Finally I got these done :-).

What is CloudApp?

CloudApp is a nice and small application for all Mac OS X computers. Here's how they describe themself:

» CloudApp allows you to share images, links, music, videos and files. Here is how it works: choose a file, drag it to the menubar and let us take care of the rest. We provide you with a short link automatically copied to your clipboard that you can use to share your upload with co-workers and friends. «

This tool is very handy when it comes to sharing screenshots, as they are uploaded instantly. I use it very often. And because I like it so much I thought to write a small Perl API interface so others can build nice tools around this tool!

The Perl interface

The Perl interface to the API, called CloudApp::REST, is completely object-oriented and takes advantage of Moose, a "postmodern object system for Perl 5". It fully supports the CloudApp API, provides proxy functionality and representations for every CloudApp item type.

You can get CloudApp::REST at the nearest CPAN mirror or from CPAN directly. You can find online documentation, smoke testing results, bug tracker and more at http://search.cpan.org/perldoc?CloudApp::REST (once they are available through the automatic processes of PAUSE).

Please test, bend and break!

Yes, please do so! Every bug and spelling error ;-) that is discovered is one less, so please try and test CloudApp::REST. Please report any bugs to the bug tracker mentioned in the docs, so everyone can see and discuss!

Thanks!

P.S.: There are some API wrappers written in other languages, too, and I heard from a Linux client as well. So maybe this will be a tool for all platforms someday :-).

Call for Papers: 12. Deutscher Perl-Workshop 2010: „Modern Perl"

2010-02-17T21:12:39Z

Vom 7. bis 9. Juni 2010 findet der 12. Deutsche Perl-Workshop in Schorndorf bei Stuttgart statt. Zielgruppe des Workshops sind alle Perl-Anwender und die, die es werden wollen.

Der Workshop steht und fällt mit den Vorträgen, die 5, 20 oder 40 Minuten lang sein können.

Alle Themen, die mit Perl oder dem Perl-Umfeld, insbesondere dem Thema „Modern Perl", zu tun haben, können als Vorträge für den Workshop interessant sein. Die Einreichungsfrist für Vorschläge wurde auf Freitag, den 05. März 2010 verlängert.

http://conferences.yapceurope.org/gpw2010

Der Deutsche Perl-Workshop ist die jährliche Konferenz meist deutschsprachiger Anwender und Entwickler der Programmiersprache Perl. Der Schwerpunkt des von der Deutscher Perl-Workshop GbR und der Wirtschaftsförderung Region Stuttgart (WRS) verstaltelten Workshops ist „Modern Perl". Dabei spielen zum Beispiel die Module Catalyst, Moose und DBIx::Class eine wichtige Rolle. Catalyst als sehr flexibles Webframework, Moose mit einer postmodernen Objektorientierung für Perl und DBIx::Class als Schicht zwischen Anwendung und Datenbanken.

Call for Papers: 12th German Perl Workshop 2010: "Modern Perl"

The 12th German Perl Workshop will be hosted from June 7th to 9th 2010 in Schorndorf near Stuttgart. The workshop is targeted towards all experienced and beginning Perl developers.

Obviously, the workshop is only as good as its talks -- your talks. We are interested in all talks about Perl itself or about Perl related topics, and especially talks that would apply to this year's motto "Modern Perl". The submission deadline for your 5, 20 or 40 minute talks has been extended until March 5th, 2010.

http://conferences.yapceurope.org/gpw2010

The German Perl Workshop is a yearly held conference with mostly german-speaking users and developers of the Perl programming language. The main focus of the workshop held by the Deutscher Perl-Workshop GbR and the Wirtschaftsförderung Region Stuttgart (WRS) is "Modern Perl". The CPAN modules Catalyst, Moose and DBIx::Class are playing major roles: Catalyst as a flexible web framework, Moose with its postmodern object system for Perl and DBIx::Class as a layer between applications and databases.

Tipps & Tricks: perl^M: bad interpreter

2010-01-20T17:40:46Z

Manche Probleme begegnen einem mehrmals im Leben. Meistens erkennt man erst beim zweiten Mal, dass man sich die Lösung besser hätte aufschreiben sollen. Mit den Beiträgen unter "Tipps & Tricks" möchte ich meine Lösungen festhalten und sie gleichzeitig verbreiten.

Windows macht manchmal schon einen Mist mit den Zeilenumbrüchen. Statt ein \n-Zeilenumbruch, wie es z.B. Linux und Mac OS X machen, verwendet Windows einen Zeilenumbruch, bestehend aus zwei Zeichen: \r\n. Ist wohl vielen bekannt.

Nun, als Perl-Entwickler entwickelt man manchmal auch Perl-Skripte, die in der ersten Zeile den Interpreter angegeben ist. Dummerweise erkennen manche Linux-Distributionen ein \r als Teil des Interpreterpfades an, und versuchen anstatt /usr/bin/perl den Aufruf von /usr/bin/perl\r - auf der Konsole steht dann immer /usr/bin/perl^M.

Die Lösung besteht darin, alle in der Datei enthaltenen \r\n durch jeweils \n zu ersetzen. Doch, wie macht man das am Besten ohne viel Aufwand? Der folgende Befehl zeigt, wie es sehr einfach geht, eine einzelne Datei entsprechend umzuwandeln:

perl -pi -e 's/\r\n?/\n/' foo.pl

Tipps & Tricks: MySQL meldet "Thread stack overrun"

2010-01-13T12:00:00Z

Aufgrund des Upgrades von Leopard (Mac OS X 10.5) auf Snow Leopard (10.6) liefen einige über Macports installierte Applikationen und Dienste nicht mehr korrekt. Unter diesen war auch MySQL. Nach einem Update dieser Applikationen lieferte mir MySQL bei einem normalen Insert-Befehl folgende Meldung:

Thread stack overrun: 4992 bytes used of a 131072 stack, and 128000 bytes needed. Use 'mysqld -O thread_stack=#' to specify a bigger stack.

Die Lösung hierzu war, wie ich bereits vermutet hatte, eine (wohl fehlerhaft gesetzte) MySQL-Konfigurationsvariable in der Datei my.cnf zu ändern. Die Variable thread_stack muss auf einen höheren Wert, z.B. "256K" gesetzt und MySQL neugestartet werden.

Übrigens, falls man MySQL unter Mac OS X nicht als LaunchDaemon laufen hat, kann man über die Kommandos mysqladmin shutdown und anschließend mysqld_safe & (als root ausführen, z.B. mit sudo) einen Neustart von MySQL hinlegen.