ananasblau.com

spring games

2011-05-04T17:48:00Z

Bringing sirtet into the Mac App Store was harder than I thought, in the end I had to change the name to Unstack but it’s in the store so go and buy it!

Last weekend I spent working on a new game for ludum dare #20 and the result is called Trollbridge Armours written in Javascript and using RaphaelJS for all the user interface. I couldn’t get the tech tree finished and also should have tested it in Firefox, Chrome or at least Safari (I’m using the nightly Webkit) but that’s also coming in the next days. I learned a lot about RaphaelJS and where it has it’s difficulties with the browsers, good to know. With a better set of graphics the game will be ready to be simply wrapped into a WebView and pushed into the App Store.

Sirtet - a MacRuby game

2011-04-02T05:24:00Z

I’ve spent the better part of yesterday creating a new game. Mind you it’s been a mere 8 hours to create this game. The goal of the game is to remove certain blocks from a large field, basically Tetris in reverse for all those whose stack grew too big in those old days.

As the title says, the game is in MacRuby, actually my third game in this lingo after The Flying Camera in December and the failed attempt (well, only the physics are missing) of Springz last week. I’ve learned a lot about XCode (4), again. Also a lot about the inner workings of a desktop application, compared to the usual web apps.

Check out the code a the github repository there’s also a download for the lazy. The game requires a 64bit Snow Leopard system, sorry about that. I’m planning on putting it into the Mac App Store which is why the game is not open source, you are only allowed the pleasure of looking at it. Kim Jong Il already did and he was moved by it.

macruby-statusmenu example

2011-03-26T07:42:00Z

I pushed a little example application written in MacRuby that shows you how to write an application for your system’s status menu. It also displays a custom view that you can style with images and stuff in this status menu. The app is set to run as agent meaning it won’t show in the Dock Menu when running.

Clone it from my github repository

Naiku in practice

2011-02-19T10:20:00Z

It’s fantastic to see the applications that I create in practice. Naiku is a startup I work for since August and the company has made a video of students using the app (both the web and mobile interface). Kudos to John and Peter who joined my developing efforts in the last months.

Run validations early on

2011-02-05T07:43:00Z

I just had the case in a Rails app where the user’s email doesn’t have to be valid. The validations in this case are done by Devise. If the email isn’t valid I want to remove it. The way to do this is to run all validators on email before the actual validation and then decide wether to set email to nil.

1234	before_validation do User.validators_on(:email).map{\|v\| v.validate(self) } self.email = nil if errors[:email]end

Super Gamedev Weekend 2010

2011-01-14T17:16:00Z

I’m a bit late with writing this post but I’m happy to announce that with the help of my friend “Ben:”https://abwesend.com/ I have won the last game programming competition held at the metalab in Vienna. The game is called The Flying Camera and is written using MacRuby and OpenGL. It is released as open source.

This is very fantastic news as it was in fact the first time I ever wrote a game and it sure will be the base for more games to come.

Prevent double clicks on ajax links (rails.js)

2010-12-09T10:19:00Z

A simple problem: In the new Schoolcamp app I’m working on this month I have some inline editing functionality using jQuery and UJS. In cases where the user clicked twice on the “edit” link he’d get two forms, not quite what a good UI would do. My solution was as simple and unobstrusive as can be and I’d like to share the code with you. I’m using the rails.js for jQuery and the only thing you have to add to your html links (or forms) is data-once="true"

The HTML

   <a href="/tasks/1/edit.js" data-method="get" data-once="true" data-remote="true">Edit</a>

The javascript event listeners

// To prevent double clicks on a remote link or form you can
// simple add data-once=true to your html tag and we
// stop the ajax request here.
$('[data-once=true]').live('ajax:before', function(request, e) {
  var el = request.target;
  if($(el).attr('data-clicked')) {
    return false;
  } else {
    $(el).attr('data-clicked', true);
    return true;
  }
  return true;
});

// after failure we want to click again, hoping for succes on a second try
$('[data-once=true]').live('ajax:failure', function(request,e) {
  $(request.target).removeAttr('data-clicked');
});

Rails 3 upgrade: two tips

2010-10-13T18:34:00Z

I think I’ve started porting photostre.am to Rails3 sometime in January, before the betas I think. Sometimes I had to stick to a certain commit of Rails as later ones were broken for me but the whole thing went quite well and over the last ten months I barely had to change anything. It’s a different story for those among you who excessively used rails’ javascript helpers, but I’ve been on UJS for long time now.

But today I did another upgrade, this from rails commit 9e081caee7 to 3.0.0. Finally. I ran into two problems and would like to share the solution with you.

First of all, there’s a deprecation notice about the routes. I have updated to the new DSL but the devil’s hiding in details:

--- a/config/routes.rb
+++ b/config/routes.rb
@@ -1,4 +1,4 @@
-DasPhotowall::Application.routes.draw do |map|
+DasPhotowall::Application.routes.draw do

Second, I’ve updated my environment.rb from Rails::Application.initialize! to DasPhotowall::Application.initialize! and soon later ran into the nasty problem of not being able to login. Sessions didn’t work!. I should have grepped over the whole repo earlier because in the session_store initializer there was still a Rails::Application hiding. So, update your initialisers.

A custom padrino FormBuilder

2010-07-28T06:07:00Z

More of a noobie question that came up yesterday at the #padrino irc channel:

What is the right place for a custom form builder?

About the location, the manual doesn’t really say it but it says that helpers should be put into app/helpers and as the FormBuilder is a helper it should go there too. From there on the noob ran into the next problem, how to get it work, because the example from the docs didn’t really work:

12	class CustomFormBuilder < StandardFormBuilderend

It could be he didn’t load the padrino-helpers, as his error was something like No constant StandardFormBuilder found and even Padrino::Helpers::FormBuilder::StandardFormBuilder couldn’t be found.

I’ve helped him by looking up in the padrino-code (luckily it’s not that huge) and basically what you do in such situations is to rebuild what you find in that code. In this case we’ve placed the CustomFormBuilder into Padrino::Helpers::FormBuilder and it worked.

# put this into app/helpers/custom_form_builder.rb
module Padrino
  module Helpers
    module FormBuilder #:nodoc:
      class CustomFormBuilder < StandardFormBuilder
      end
    end
  end
end

MongoDB in my photostre.am

2010-06-11T06:21:00Z

Time to blog about the reasons why I’ve migrated my photostre.am application from MySQL to MongoDB. I hope by this I can give a few people arguments to do the same.

Document based

photostre.am is powered by Ruby, a dynamic and fully object oriented programming language (well, except for blocks). MySQL isn’t that far off when using ruby, you can’t stack data rows into other rows but any decent database adapter will do those relations for you. But if you heart dynamic functionality as much as I do there’s no future for you in a schema database and the schema-less ones your best option.

Arrays

Arrays are a wonderful data structure to work with, e.g. for tags, but if you want to access single elements from an array in a mySQL database you most likely end up creating at least one more table, where you store each single element of your array, and query on that. I know, there are workarounds but MongoDB stores it’s data in BSON format which offers you exactly that in the straightmost way. I couldn’t be more happy to store my tags right inside a photo and retrieving photos that match a tag is only one query. For photostre.am it also allowed me to add new features, for example the Website model stored a domainhost and in the MySQL database I was too lazy for this second table as it wasn’t a much important feature but in MongoDB it has become very quick and easy to add with no loss in performance (and for an object that’s queried with every request that is important.

Embedded objects

This is one of the most amazing features and completes my object oriented programming experience. With MongoDB you can store whole objects inside a parent objects. For example the Website in photostream stores albums and pages which are of course models in my rails application and have all the functionality a regular model would have except the don’t need another collection (table in sql speak). The old database had more than 10 tables, the new one only three collections. It helped me to reduce the number of queries and it just feels natural.

No more SQL

I know SQL quite well. Learned it the hard way in the early 2000s and I always prefer the console over phpMyAdmin and similar. With MongoDB I have a great console, with all modern features and as a bonus it’s all javascript which is a great language and much undervalued by most people but as powerful as ruby.

Sharding

Not yet a feature that I need but photostre.am is close to 200 users now and they all come with around 1000 photos (respectively just the data about those photos) and the database must grow faster in the future if I want to make any money with it. MongoDB with it’s sharding will allow me to split the database horizontal over several small instances and I won’t have to upgrade a single DB server with more RAM and CPUs like you would need with a mySQL database.

Mongoid: Permalinks on embedded documents

2010-05-17T16:55:00Z

The target is to create a permalink from the title of the object and append a number if the permalink is already taken by a different object. The code uses Mongoid but I think MonogMapper shouldn’t work that much different nowadays (I yet didn’t have access to the root object of an embedded object in January).

class Photo
  include Mongoid::Document
  embedded_in :parent, :polymorphic => true, :inverse_of => :photos
  before_validate :set_permalink

  def set_permalink(permalink = nil)
    title = self.title.strip.blank? ? 'image' : self.title
    if (title and self.permalink.blank?) || permalink
      permalink ||= title
      permalink = permalink.to_permalink.strip

      permalink_index = nil
      # get all permalinks of the photo's siblings
      permalinks = parent.photos.collect(&:permalink)
      while permalinks.include?([permalink, permalink_index].compact.join('-'))
        permalink_index ||= 0
        permalink_index += 1
      end
      self.permalink = [permalink, permalink_index].compact.join('-')
    end
  end

# Taken from has_permalink plugin by Nando Vieira released under the MIT license
class String
  def to_permalink
    str = ActiveSupport::Multibyte::Chars.new(self)
    str = str.normalize(:kd).gsub(/[^\x00-\x7F]/,'').to_s
    str.gsub!(/[^-\w\d]+/sim, "-")
    str.gsub!(/-+/sm, "-")
    str.gsub!(/^-?(.*?)-?$/, '\1')
    str.downcase!
    str
  end
end

I admit, quite a trivial issue but maybe my code examples saved someone precious time.

MongoDB Vortrag auf den Linuxwochen 2010

2010-05-06T10:44:00Z

Auf den Linuxwochen 2010 in Wien hab ich einen kleinen Vortrag zu MongoDB gehalten. Hier der Link zum Vortrag:

http://tomk32.github.com/mongodb-presentation-linuxwochen-2010/

reprocessing images in Carrierwave

2010-02-19T16:30:00Z

For garmz I’m using the CarrierWave plugin and when I had to implement cropping for the images that are attached to our designs I faced the problem of having to regenerate the various versions based with different cropping values.

The main principle of my solution is to create new images with new ids, path names etc to prevent any photo not being reloaded on the client side. Otherwise I’d have to invalidate the caches, not my thing.

In order to achieve this we create a new Image, assign my cropping parameters and then assign the file from the old Image. For some reasons it looks as if the images are processed once it’s assigned and not when it’s being saved.

In the Controller:

if images = params[:crop_images]
  images.each do |image_id, crop_data|
    next if crop_data.blank?

    old_file = object.images.find(image_id).asset.file
    new_image = Image.new(:crop => JSON.parse(crop_data).reject{|k,v| !%w(x y h w).include?(k)})
    new_image.asset = old_file

    new_image.write_asset_identifier
    new_image.store_asset!
    object.images << new_image
    object.images.find(image_id).destroy
  end
end

In the model:

class ImageUploader
  version :thumb do
    process! :crop
  end
  def crop
    return if model.crop.blank?
    manipulate! do |img|
      img.crop "%ix%i+%i+%i" % [model.crop["w"], model.crop["h"], model.crop["x"], model.crop["y"]]
    end
  end
end
class Image
  include Mongoid::Document
  field :crop, :type => Hash
  belongs_to :designs
  mount_uploader :asset, ImageUploader
end

MySQL to MongoDB conversion

2010-02-01T10:26:00Z

For photostre.am I wrote an migration script to migrate the MySQL database to MongoMapper.

Maybe it’s good use for someone, this is the actual script at gist.github.com and no abstract something which you give up to extend after a while. The migration actually uses ActiveRecord::Migration, could well be about the last one that I’ll ever write.

Lessons learned:

Keep the old id from MySQL for a while, but rename it
Don’t do too many EmbeddedDocuments

photostre.am

2010-01-25T13:08:00Z

I totally forgot to blog about photostre.am here. It’s a project which I’ve worked on since October 2008, put it aside after the first hundrets lines of code and then picked it up again in Summer. And in October it finally went online. In those three months since I saw lots of improvements but it’s still not finished or making any money.

The problem I’ve faced back then was that I have loads of photos on flickr, most shot with my favourite analogue cameras, but no proper portfolio website. Of course you can point people to your flickr website, or even the sideshow there, but it’s not the same. There are a few tool around that show your photos on black background, or javascript widgets that you put into your website, but not a single solution solved the problem from front to end. From photo import, rearranging, grouping, albums, multiple domains, static content pages, no tool does that, except for photostre.am. Just after releasing it I actually found a tool which solved the problem almost, but still my webapplication has a brighter future to come.

Checkout the blog, the Twitter account and did I mention the app is white-labeled? No, then go for 365shots.net