Experimenting with pry and JRubyArt

If we replace ruby-processings 'live.rb' with the following version, we can easily use pry with ruby-processing (but I am not sure what for, because I don't use pry).

# An PRY shell for live coding. 
# This flavor will either load up empty JRubyArt sketch, 
# or will start with your sketch.

require "#{File.dirname(__FILE__)}/base.rb"
Processing.load_and_run_sketch

ARGV.clear # So that IRB doesn't try to load them as files.

require 'pry'

binding.pry

You still need to access sketch using $app, eg $app.background 0 sets background to 0 (but I expect that could be changed?). Ideally you will also install pry with rgem, and use the --gem flag to be sure that jruby can use the pry gem. Actually turn out this is still somewhat flakey, certainly on linux.

Rakefile to run JRubyArt sketches in a directory from netbeans (tested with samples/contributed)

NB need to adjust path to "k9" executable...., since we are running with development version in netbeans (not an externally installed version)

# Simple demo Rakefile to autorun samples in current directory
# from netbeans, adjust path to k9 executable as required

SAMPLES_DIR="./"

desc 'run demo'
task :default => [:demo]

desc 'demo'
task :demo do
  samples_list.shuffle.each{|sample| run_sample sample}
end

def samples_list
  files = []
  Dir.chdir(SAMPLES_DIR)
  Dir.glob("*.rb").each do |file|
    files << File.join(SAMPLES_DIR, file)
  end
  return files
end

def run_sample(sample_name)
  puts "Running #{sample_name}...quit to run next sample"
  open("|../../bin/k9 run #{sample_name}", "r") do |io|
    while l = io.gets
      puts(l.chop)
    end
  end
end

Experimental handler for overloaded "background" in JRubyArt

Previously for ruby-processing, we allowed jruby to make sensible choices for methods with overloaded signatures (but to encourage more efficient operation, jruby is being very noisy about it) and we told you how you can use specific method signatures using java_alias to shut up jruby (and be more efficient). However background it could be argued (because of its ubiquity in sketches and the unique method signatures) might warrant special treatment. Here is an experiment to add such features via the helper_methods module.

# The vanilla processing background method is highly overloaded, a single
# argument could be an instance of PImage, int or float (multiple arguments are
# usually taken to be floats).
# @param *args arguments for background as an Array
#

def background *args
  if (args.length == 1)
    if args[0].kind_of? Integer
      self.java_send :background, [Java::int], args[0]
    elsif args[0].kind_of? Float
      self.java_send :background, [Java::float], args[0]
    elsif args[0].kind_of? Java::ProcessingCore::PImage
      self.java_send :background, [Java::ProcessingCore::PImage], args[0]
    end
  elsif (args.length == 3)
    self.java_send :background, [Java::float, Java::float, Java::float], args[0], args[1], args[2]
  elsif (args.length == 2)  # greyscale with transparency
    self.java_send :background, [Java::float, Java::float], args[0], args[1]
  elsif (args.length == 4) # color with transparency
    self.java_send :background, [Java::float, Java::float, Java::float, Java::float], args[0], args[1], args[2], args[3]
  end
end

JRubyArt is the development branch of ruby-processing.

Developing JRubyArt with Netbeans

Thanks largely to Thomas Enebo, I expect to be doing future development of JRubyArt all under Netbeans see screenshot click on image to see full size:-

Screenshot of me developing JRubyArt  with Netbeans

It is just bloody brilliant, brings all the code together, sensible refactoring hints, easy renaming etc.etc.

Time to give up on the Mac

I have started a new development fork of ruby-processing, I am calling it JRubyArt, there will be no support for the Mac (in the foreseeable future, possibly never, depends on active co-operation Mac users). Calling it JRubyArt may not be so daft because the end point need not only be a processing sketch (PovRAY or Sunflow).
It is an absolute requirement of the development branch to be using at least java-7 (this is requirement for processing-2.1+, but I am really targeting java-8). The idea is that JRubyArt keeps step with jruby-9000 development (which will likely require at least java-7, but likely to work best with java-8). On this basis, there is absolutely no point in supporting the Mac, which post Maverick seems to be designed not to work with java (having said that many jruby users have Macs go figure).

Revised advice for Mac users of ruby-processing

Update 7 December 2012

It appears that vanilla processing-2.1.0 may not be working on Mavericks, however processing-2.0.3 does work, especially using Apple jvm (but this is unsustainable in longer term, and I am sure you should be working towards using the latest Oracle jvm). It seems that you should really install the Oracle JDK rather than just the runtime (to get links updated etc) but if you have previously installed the Apple jvm make sure you do a thorough job of uninstalling it first.
I have just updated the main branch of ruby-processing (version 2.4.0) to make use of an installed version of vanilla processing. Until Ben Fry endorses processing-2.1+ on Mavericks, I recommend installing processing-2.0.3 on the Mac (and processing 2.1+ on linux and Microsoft windows).

Ruby-processing on MacOSX, Apple or Oracle JVM

UPDATED 1 February 2014 ruby-processing-2.4.2 (is now available as a rubygem)

Now this is entirely theoretical as I've exceptionally limited experience with a MAC (I once loaded some photos on my Artist friends MAC). However, from what I understand if you update to Mavericks, you won't have java installed.  The ideal solution from my point of view, is that this forces you to install Oracle jdk1.7.0_51 (64 bit). However apparently if you are conservative, or otherwise need it, apparently you can get Apple jvm here.


For OSX users incapable of using google, for some reason following link may be helpful stackoverflow it may even encourage you to do the sensible thing and install the Oracle jvm?

But what does this mean for ruby-processing on MacOSX:

Use processing-2.0.3 especially if you are sticking with Apple jvm (Mavericks or otherwise)

Step by step instruction to install of the latest version of ruby-processing:-

1. Install jruby (at least version 1.7.3 but preferably 1.7.10)
2. Install processing-2.0.3, or if not on Mavericks you could try processing-2.1.1 & jdk7
3. Run the processing sketch here to set the path to the processing root
4. gem install ruby-processing
5. install_jruby_complete
6. cd your work directory
7. rp5 unpack samples
8. cd samples/contributed
9. rp5 run grapher.rb (or jwishy.rb or whatever)

Please report how you get on since if I can establish that this is working OK we will be able to return to ruby-gems distribution. If you don't have a system installed jruby, all sketches you would need to use the --nojruby flag to use the vendored jruby-complete eg

rp5 --nojruby run grapher.rb

Here is another link that might be useful re installing jdk7.

Running Ruby Processing from Netbeans-7.4

Hey this is a bit exciting I've managed to get ruby processing to run under Netbeans just checkout this screen shot:-

A Rakefile to demo ruby-processing samples

Loosely based on a Shoes4 Rakefile, here is a Rakefile to demo ruby-processing files in a folder:-

# Simple demo Rakefile to autorun samples

SAMPLES_DIR="/home/tux/samples/contributed"

desc 'run demo'
task :default => [:demo]

desc 'demo'
task :demo do
  samples_list.shuffle.each{|sample| run_sample sample}
end

def samples_list
  files = []
  Dir.chdir(SAMPLES_DIR)
  Dir.glob("*.rb").each do |file|
    files << File.join(SAMPLES_DIR, file)
  end
  return files
end

def run_sample(sample_name)
  puts "Running #{sample_name}...quit to run next sample"
  system "rp5 run #{sample_name}"
end

Bleeding edge development of ruby-processing

I have just set myself up to do bleeding edge testing of ruby processing.

1. To use the latest development version of jruby-complete I created this rake file.
2. To use the latest development version of processing I cloned it from github.
3. Did a linux build.
4. Modified my .rp5rc to point PROCESSING_ROOT to the work folder (3)

To keep it separate I created a copy of my monkstone-branch to develop (taking care to remove the .git archive).
Update December 10th see a more recent blog entry for latest developments.

Testing the GLW renderer with ruby-processing

Andreas Colubri is working on an alternative renderer for processing (which seems to be stabilising as the GLW renderer). Here is an example sketch translated to ruby-processing requires processing-2.1.1 to run (OK development version as that is not currently available).

load_library :glw

include_package 'processing.glw'

def setup
  size(2560, 1440, GLW::P2D)
  frame_rate(180) # NB: here we are using the "processing" set frame rate method
end

def draw
  background(255, 0, 0)
  fill(255)
  # NB: here we access "processing" frame rate variable 
  text("FPS: #{frame_rate}", mouse_x, mouse_y)
end

Working towards returning ruby-processing to gem distribution by rubygems.org

I have made the first steps toward returning ruby-processing to rubygems distribution. See my fork on github. This is work in progress, however it does work on Archlinux, and also should also work on the MAC, where the path to the processing jars should be stable?

1. Install processing-2.1.0 (for your OS)
2. Create or edit the YAML .rp5rc file to point to processing-root possibly using the sketch below (in the processing ide) config.pde
3. Clone my fork
4. cd ruby-processing (or whatever you call it)
5. rake (MAC and linux) to see if tests work
6. gem install ruby-processing-{version}.gem for more comprehensive testing

Currently for ease of testing we still include jruby-complete download as option, but it is not required for many sketches, intention is it will be available as a post install option.

String processingRoot = "enter your processing root here"; // edit this line in the sketch
String home;
PFont font;

void setup() {
  size(400, 200);
  font = createFont("Helvetica", 18);
  home = System.getProperty("user.home");
}

void draw() {
  background(255);
  fill(255, 0, 0);
  textFont(font, 18);
  // this adds a blinking cursor after your text, at the expense of redrawing everything every frame
  text(processingRoot+(frameCount/10 % 2 == 0 ? "_" : ""), 35, 45);
}

void keyReleased() {
  if (key != CODED) {
    switch(key) {
    case BACKSPACE:
      processingRoot = processingRoot.substring(0, max(0, processingRoot.length()-1));
      break;
    case ENTER:   // save the processing root to the config file
    case RETURN:
      PrintWriter pw = createWriter(home + "/.rp5rc");
      pw.write("PROCESSING_ROOT: \"" + processingRoot + "\"");
      pw.close();
      break;
    case ESC:
    case DELETE:
      break;
    default:
      processingRoot += key;
    }
  }
}

Promise for the future of jruby, ruby-processing

It was probably a bit too early when I last looked at tuning ruby processing, because it preceded the processing-2.1 release (now compiled for java 7, which means I can run ruby-processing with java 8). Also since I am looking to the future I will use jruby-9000.dev for future performance tests. The esfera test is good because it is dynamic (you can actually see both compiler and garbage collector kick-in, well thats unless I'm very much mistaken). Here are past results combined with current results using jruby-complete-9000.dev (2 November 2013 snapshot). On continued running current test stabilised at ca 6 fps. What the test doesn't show is elapsed time, because there is actual pause seen when I presume compiler kicks in (and the later lesser dips in performance must be the garbage collector). These are my java args for the test (jdk 7 unless stated)
 java_args.txt

• None 
• Args 1.  -XX:+TieredCompilation XX:CompileCommand=dontinline,org.jruby.runtime.invokedynamic.InvokeDynamicSupport::invocationFallback
• Args 2. -XX:+TieredCompilation -XX:TieredStopAtLevel=1 XX:CompileCommand=dontinline,org.jruby.runtime.invokedynamic.InvokeDynamicSupport::invocationFallback
•  Args Java 8 -Xms2048M -Xmx2048M -XX:+UnlockExperimentalVMOptions -XX:G1NewSizePercent=20 -XX:G1MaxNewSizePercent=80.

F.P.S esfera

No Args	Args 1	Args 2	Java 8
5.4547	5.17617	4.70084	5.27
4.26695	3.71662	2.90269	4.24
3.85179	3.22197	2.30053	3.889
3.769622	3.105720	2.077115	3.70
3.769622	3.122121	2.022429	4.15
3.756052	3.340643	1.940190	4.98
4.849481	3.122121	1.953019	5.03
5.173814	3.34064		5.09

PS: linux users it may also be worth doing the following, one-time exercise, to improve java startup

sudo java -Xshare:dump

For Rails Hackers

I'm a bit disappointed so far with the response to my updating of ruby-processing to run with processing-2.1. Perhaps interest in the ruby language is on the decline, or could it be that rails hackers have become so dependent on their frameworks/tools they find it a challenge to work outside them (I mean just look at Shoes4 lots of meaningless tests many contributors, and it is travelling at snails pace to where? _why is surely less than impressed)!!! For inspiration just check out the Karsten Schmidt story.

Eyeo 2013 - Karsten Schmidt from Eyeo Festival on Vimeo.

An opportunity to restore ruby-processing as a rubygem distributable gem?

It occurs to me that at least on ArchLinux (since processing-2.0.1) and the Mac (since processing-2.1) the location (path) of processing jars could be relied on (albeit OS specific). Further the majority of sketches can now be run with an installed jruby (which is now the default) thus we could exclude jruby-complete and the processing jars from the gem, and we would easily return to rubygems distribution. It would be sensible to retain the the jruby-complete as an optional post install to enable access to shader sketches and possibly still support application export, although that would not be a priority. The folks at processing are still inviting integration with the processing ide (see below), but I would prefer ruby-processing to stick with the ruby ecology.

Our sister project Ruby-Processing version 2.2.0 is now available from github: https://t.co/rC2Quwz3hz

— Processing (@ProcessingOrg) October 10, 2013

Who wants to make a Ruby-Processing Mode for the PDE?

— Processing (@ProcessingOrg) October 10, 2013

Using shaders in ruby-processing (tutorial example translated)

Anyone interested in using shaders in ruby-processing should read the vanilla processing tutorial here. All the code for the tutorial is available here, including the necessary glsl shader (and image) files used in this sketch, which is translated from one of the included examples.You might also like to read the codeanticode blog (by Andrés Colubri) to keep up to date with latest developments.

# Texture from Jason Liebig's FLICKR collection of vintage labels and wrappers:
# http://www.flickr.com/photos/jasonliebigstuff/3739263136/in/photostream/

CAN_SIZE = 60

attr_reader :label, :can, :cap, :angle, :color_shader, :light_shader, :texlight_shader
attr_reader :sel_shader, :pixlight_shader, :tex_shader, :texlightx_shader, :bw_shader
attr_reader :edges_shader, :use_texture, :use_light, :emboss_shader

def setup
  size(480, 480, P3D)
  @label = load_image("lachoy.jpg")
  @can = create_can(CAN_SIZE, 2 * CAN_SIZE, 32, label)
  @cap = create_cap(CAN_SIZE, 32)
  @angle = 0
  @color_shader = load_shader("colorfrag.glsl", "colorvert.glsl")
  @light_shader = load_shader("lightfrag.glsl", "lightvert.glsl")
  @tex_shader = load_shader("texfrag.glsl", "texvert.glsl")

  @texlight_shader = load_shader("texlightfrag.glsl", "texlightvert.glsl")
  @pixlight_shader = load_shader("pixlightfrag.glsl", "pixlightvert.glsl")
  @texlightx_shader = load_shader("pixlightxfrag.glsl", "pixlightxvert.glsl")

  @bw_shader = load_shader("bwfrag.glsl")
  @edges_shader = load_shader("edgesfrag.glsl")
  @emboss_shader = load_shader("embossfrag.glsl")

  @sel_shader = texlight_shader
  @use_light = true
  @use_texture = true
  puts "Vertex lights, texture shading"
end

def draw
  background(0)

  x = 1.88 * CAN_SIZE
  y = 2 * CAN_SIZE
  3.times do
    3.times do
      draw_can(x, y, angle)
      x += (2 * CAN_SIZE + 8)
    end
    x = 1.88 * CAN_SIZE
    y += (2 * CAN_SIZE + 5)
  end

  @angle += 0.01
end

def draw_can( centerx, centery, rot_angle)
  push_matrix

  if (use_light)
    point_light(255, 255, 255, centerx, centery, 200)
  end
  shader(sel_shader)

  translate(centerx, centery, 65)
  rotate_y(rot_angle)
  if (use_texture)
    can.set_texture(label)
  else
    can.set_texture(nil)
  end
  shape(can)
  no_lights

  reset_shader

  push_matrix
  translate(0, CAN_SIZE - 5, 0)
  shape(cap)
  pop_matrix

  push_matrix
  translate(0, -CAN_SIZE + 5, 0)
  shape(cap)
  pop_matrix

  pop_matrix
end

def create_can(r, h, detail, tex)
  texture_mode(NORMAL)
  sh = create_shape
  sh.begin_shape(QUAD_STRIP)
  sh.no_stroke
  sh.texture(tex)
  (0 .. detail).each do |i|
    angle = TAU / detail
    x = sin(i * angle)
    z = cos(i * angle)
    u = i.to_f / detail
    sh.normal(x, 0, z)
    sh.vertex(x * r, -h/2, z * r, u, 0)
    sh.vertex(x * r, +h/2, z * r, u, 1)
  end
  sh.end_shape
  return sh
end

def create_cap(r, detail)
  sh = create_shape
  sh.begin_shape(TRIANGLE_FAN)
  sh.no_stroke
  sh.fill(128)
  sh.vertex(0, 0, 0)
  (0 ... detail).each do |i|
    angle = TAU / detail
    x = sin(i * angle)
    z = cos(i * angle)
    sh.vertex(x * r, 0, z * r)
  end
  sh.end_shape
  return sh
end

def key_pressed
  case key
  when '1'
    puts "No lights, no texture shading"
    @sel_shader = color_shader
    @use_light = false
    @use_texture = false
  when '2'
    puts "Vertex lights, no texture shading"
    @sel_shader = light_shader
    @use_light = true
    @use_texture = false
  when '3'
    puts "No lights, texture shading"
    @sel_shader = tex_shader
    @use_light = false
    @use_texture = true
  when '4'
    puts "Vertex lights, texture shading"
    @sel_shader = texlight_shader
    @use_light = true
    @use_texture = true
  when '5'
    puts "Pixel lights, no texture shading"
    @sel_shader = pixlight_shader
    @use_light = true
    @use_texture = false
  when '6'
    puts "Pixel lights, texture shading"
    @sel_shader = texlightx_shader
    @use_light = true
    @use_texture = true
  when '7'
    puts "Black&white texture filtering"
    @sel_shader = bw_shader
    @use_light = false
    @use_texture = true
  when '8'
    puts "Edge detection filtering"
    @sel_shader = edges_shader
    @use_light = false
    @use_texture = true
  when '9'
    puts "Emboss filtering"
    @sel_shader = emboss_shader
    @use_light = false
    @use_texture = true
  end
end

Emboss Filtering

Low Level GL Shader Sketch in Ruby-Processing

Here is a sketch translated from the vanilla processing examples (where to get frag.gsl and vert.gsl), to run on the latest version of ruby-processing.

include_package 'java.nio'
java_import 'processing.opengl.PGL'


attr_reader :sh, :vert_loc, :color_loc, :vert_data, :color_data

def setup
  size(640, 360, P3D)

  # Loads a shader to render geometry w/out
  # textures and lights.
  @sh = loadShader("frag.glsl", "vert.glsl")
  @vert_data = allocate_direct_float_buffer(12)
  @color_data = allocate_direct_float_buffer(12)
end

def draw
  background(0)

  # The geometric transformations will be automatically passed 
  # to the shader.
  rotate(frame_count * 0.01, width, height, 0)

  update_geometry

  pgl = beginPGL
  sh.bind

  @vert_loc = pgl.getAttribLocation(sh.glProgram, "vertex")
  @color_loc = pgl.getAttribLocation(sh.glProgram, "color")

  pgl.enableVertexAttribArray(vert_loc)
  pgl.enableVertexAttribArray(color_loc)

  pgl.vertexAttribPointer(vert_loc, 4, PGL.FLOAT, false, 0, vert_data)
  pgl.vertexAttribPointer(color_loc, 4, PGL.FLOAT, false, 0, color_data)

  pgl.drawArrays(PGL.TRIANGLES, 0, 3)

  pgl.disableVertexAttribArray(vert_loc)
  pgl.disableVertexAttribArray(color_loc)

  sh.unbind

  endPGL
end

def update_geometry
  # Vertex 1
  vertices = [0, 0, 0, 1]
  colors = [1, 0, 0, 1]

  # Corner 2
  vertices += [width/2, height, 0, 1]
  colors += [0, 1, 0, 1]

  # Corner 3
  vertices += [width, 0, 0, 1]
  colors += [0, 0, 1, 1]

  vert_data.rewind
  vert_data.put(vertices.to_java :float)
  vert_data.position(0)

  color_data.rewind
  color_data.put(colors.to_java :float)
  color_data.position(0)
end

def allocate_direct_float_buffer(n)
  ByteBuffer.allocateDirect(n * java.lang.Float::SIZE/8).order(ByteOrder.nativeOrder).asFloatBuffer
end

Penrose Snowflake Revisited

The 1st time I revisited this sketch was in response to some refactoring by Dan Myer see http://learning-ruby-processing.blogspot.co.uk/2010/01/refactoring-my-penrose-snowflake.html. Since the advent of OPENGL for two dimensional sketches, that is all old hat, since we no longer need to keep generating the geometry, we can create a shape and reuse it (multiple times in the same sketch if need be). So here we use a generic grammar library for simple lsystems:-

############################
# Simple lsystem grammar
############################
class Grammar

  attr_reader :axiom, :rules
  def initialize(axiom, rules)
    @axiom = axiom
    @rules = rules
  end

  def expand(production, iterations, &block)
    production.each_char do |token|
      if rules.has_key?(token) && iterations > 0
        expand(rules[token], iterations - 1, &block)
      else
        yield token
      end
    end
  end

  def each gen
    expand(axiom, gen) {|token| yield token }
  end

  def generate gen
    output = []
    each(gen){ |token| output << token }
    return output
  end

end

Here is the animation code adapted from a processing example:-

# A class to describe a Polygon (with a PShape)

class Polygon
  include Processing::Proxy
  # The PShape object
  attr_reader :s, :x, :y, :speed, :height

  def initialize(s_, width, height)
    @x = rand(width)
    @y = rand(height * -1.5 .. -height / 3)
    @s = s_
    @speed = rand(2 .. 6)
    @height = height
  end

  # Simple motion
  def move
    @y +=speed
    @y = -100 if (y > height + 100)
  end

  # Draw the object
  def display
    push_matrix
    translate(x, y)
    shape(s)
    pop_matrix
  end

  def run
    display
    move
  end
end

Here is the penrose snowflake

load_libraries :grammar, :polygon

class PenroseSnowflake
  include Processing::Proxy
  import 'grammar'

  attr_accessor :axiom, :grammar, :start_length, :theta, :production, :draw_length,
  :repeats, :xpos, :ypos
  DELTA = Math::PI / 10 # 18 degrees as radians

  def initialize width
    reset width
  end

  def reset width
    @axiom = "F3-F3-F3-F3-F"
    @grammar = Grammar.new( axiom,
      {"F" => "F3-F3-F45-F++F3-F"}
      )
    @start_length = width / 40
    @theta = 0
    @xpos = width / 80
    @ypos = 0
    @draw_length = start_length
  end

  ##############################################################################
  # Parses the production string, and draws a line when 'F' is found, uses
  # trignometry to calculate dx and dy rather than processing transforms
  ##############################################################################

  def create_flake production
    flake = create_shape
    fil = [200, 220, 255].sample
    flake.begin_shape
    flake.fill fil
    flake.stroke fil
    flake.vertex(xpos, ypos)
    repeats = 1
    production.each do |element|
      case element
      when 'F'
        flake.vertex(@xpos -= multiplier(repeats, :cos), @ypos += multiplier(repeats, :sin))
        repeats = 1
      when '+'
        @theta += DELTA * repeats
        repeats = 1
      when '-'
        @theta -= DELTA * repeats
        repeats = 1
      when '3', '4', '5'
        repeats += element.to_i
      else
        puts "Character '#{element}' is not in grammar"
      end
    end
    flake.end_shape CLOSE
    return flake
  end

  ##########################################
  # adjust draw length with number of repeats
  # uses grammar to set production string 
  # see 'grammar.rb'
  ##########################################

  def create_grammar(gen)
    @draw_length *= 0.4**gen
    @production = grammar.generate gen
  end

  ###########################################
  # a helper method that returns dx or dy with type & repeat
  # multiplier after Dan Mayer
  ###########################################

  def multiplier(repeats, type)
    value = draw_length * repeats
    # using equal? for identity comparison
    (type.equal? :cos)?  value * cos(theta) : value *  sin(theta)
  end
end

##
# Lindenmayer System in ruby-processing by Martin Prout
# Very loosely based on a processing Penrose L-System
# by Geraldine Sarmiento
###

attr_reader :polygons

def setup
  size displayWidth, displayHeight, P2D
  #stroke 255
  flakes = []
  @polygons = []
  penrose = PenroseSnowflake.new width
  (2 .. 4).each do |i|
    production = penrose.create_grammar(i)
    flakes << penrose.create_flake(production)
    penrose.reset width
  end
  40.times do
    polygons << Polygon.new(flakes.sample, width, height)
  end
end

def draw
  background 0
  # Display and move them all
  polygons.each do |poly|
    poly.run
  end
end

Graphic Exploration of Jruby tuning in ruby-processing

The esfera sketch in vanilla processing is in the demo (performance category) and has proved to an interesting test case for various perfomance tune-ups in jruby. For this sketch the rolling fps is a good measure of performance.

 java_args.txt

None
Args 1. -XX:+TieredCompilation XX:CompileCommand=dontinline,org.jruby.runtime.invokedynamic.InvokeDynamicSupport::invocationFallback
Args 2. -XX:+TieredCompilation -XX:TieredStopAtLevel=1 XX:CompileCommand=dontinline,org.jruby.runtime.invokedynamic.InvokeDynamicSupport::invocationFallback

F.P.S esfera

No Args	Args 1	Args 2
5.4547	5.17617	4.70084
4.26695	3.71662	2.90269
3.85179	3.22197	2.30053
3.769622	3.105720	2.077115
3.769622	3.122121	2.022429
3.756052	3.340643	1.940190
4.849481	3.122121	1.953019
5.173814	3.34064

Thereafter no args stabilized at ca 5.4 fps whereas "Args 1" stabilized at ca 5.7 fps. So compilation Tiered compilation beyond level 1 gives best performance, eventually.

Multiple panels example grafica library

Sometime using array as method args is not such a good idea, here we need to explicitly cast our ruby arrays to Java::float arrays.

load_library :grafica
include_package 'grafica'

N_POINTS = 21

def setup
  size(500, 500)
  background(255)

  first_plot_pos = [0, 0]
  panel_dim = [200, 200]
  margins = [60, 70, 40, 30]

  # Create four plots to represent the 4 panels
  plot1 = GPlot.new(self)
  plot1.set_pos(first_plot_pos.to_java(Java::float))
  plot1.set_mar([0, margins[1], margins[2], 0])
  plot1.set_dim(panel_dim.to_java(Java::float))
  plot1.set_axes_offset(0)
  plot1.set_ticks_length(-4)
  plot1.getXAxis.set_draw_tick_labels(false)

  plot2 = GPlot.new(self)
  plot2.set_pos(first_plot_pos[0] + margins[1] + panel_dim[0], first_plot_pos[1])
  plot2.set_mar([0, 0, margins[2], margins[3]])
  plot2.set_dim(panel_dim.to_java(Java::float))
  plot2.set_axes_offset(0)
  plot2.set_ticks_length(-4)
  plot2.getXAxis.set_draw_tick_labels(false)
  plot2.getYAxis.set_draw_tick_labels(false)
  plot3 = GPlot.new(self)
  plot3.set_pos(first_plot_pos[0], first_plot_pos[1] + margins[2] + panel_dim[1])
  plot3.set_mar([margins[0], margins[1], 0, 0])
  plot3.set_dim(panel_dim.to_java(Java::float))
  plot3.set_axes_offset(0)
  plot3.set_ticks_length(-4)
  plot4 = GPlot.new(self)
  plot4.set_pos(first_plot_pos[0] + margins[1] + panel_dim[0], first_plot_pos[1] + margins[2] + panel_dim[1])
  plot4.set_mar([margins[0], 0, 0, margins[3]])
  plot4.set_dim(panel_dim.to_java(Java::float))
  plot4.set_axes_offset(0)
  plot4.set_ticks_length(-4)
  plot4.getYAxis.set_draw_tick_labels(false)

  # Prepare the points for the four plots
  points1 = GPointsArray.new(N_POINTS)
  points2 = GPointsArray.new(N_POINTS)
  points3 = GPointsArray.new(N_POINTS)
  points4 = GPointsArray.new(N_POINTS)

 N_POINTS.times do |i|
    points1.add(sin(TAU*i/(N_POINTS-1)), cos(TAU*i/(N_POINTS-1)))
    points2.add(i, cos(TAU*i/(N_POINTS-1)))
    points3.add(sin(TAU*i/(N_POINTS-1)), i)
    points4.add(i, i)
 end

  # Set the points, the title and the axis labels
  plot1.set_points(points1)
  plot1.getYAxis.set_axis_label_text("cos(i)")
  plot1.set_title_text("Plot with multiple panels")
  plot1.get_title.set_relative_pos(1)
  plot1.get_title.set_text_alignment(CENTER)

  plot2.set_points(points2)

  plot3.set_points(points3)
  plot3.getXAxis.set_axis_label_text("sin(i)")
  plot3.getYAxis.set_axis_label_text("i")
  plot3.setInvertedYScale(true)

  plot4.set_points(points4)
  plot4.getXAxis.set_axis_label_text("i")
  plot4.setInvertedYScale(true)

  # Draw the plots
  plot1.begin_draw
    plot1.draw_box
    plot1.drawXAxis
    plot1.drawYAxis
    plot1.draw_top_axis
    plot1.draw_right_axis
    plot1.draw_title
    plot1.draw_points
    plot1.draw_lines
  plot1.end_draw

  plot2.begin_draw
    plot2.draw_box
    plot2.drawXAxis
    plot2.drawYAxis
    plot2.draw_top_axis
    plot2.draw_right_axis
    plot2.draw_points
    plot2.draw_lines
  plot2.end_draw

  plot3.begin_draw
    plot3.draw_box
    plot3.drawXAxis
    plot3.drawYAxis
    plot3.draw_top_axis
    plot3.draw_right_axis
    plot3.draw_points
    plot3.draw_lines
  plot3.end_draw

  plot4.begin_draw
    plot4.draw_box
    plot4.drawXAxis
    plot4.drawYAxis
    plot4.draw_top_axis
    plot4.draw_right_axis
    plot4.draw_points
    plot4.draw_lines
  plot4.end_draw
end

Since updated for JRubyArt

Reading csv data into a grafica sketch in ruby-processing

The clear superiority of ruby-processing in dealing with csv data is self evident in this sketch. Get the grafica library by jagracar here.

require 'csv'
load_library :grafica
include_package 'grafica'

MONTH_NAMES = %w(January February March April May June July August September October November December)
DAYS_PER_MONTH = [31,28,31,30,31,30,31,31,30,31,30,31]
DAYS_PER_MONTH_LEAP_YEAR = [31,29,31,30,31,30,31,31,30,31,30,31]

attr_reader :plot, :points_oktoberfest, :points_elections

def setup
  size(800, 400)

  # Load the Oktoberfest vs. Bundestagswahl (German elections day) Google 
  # search history file (obtained from the Google trends page). 
  # The csv file has the following format: 
  # year,month,day,oktoberfest,bundestagswahl
  # 2004,0,1,5,1
  # ...
  @points_oktoberfest = GPointsArray.new
  @points_elections = GPointsArray.new


  data = CSV.read("data/OktoberfestVSGermanElections.csv", :headers => true).map{|row| row.to_hash}
  data.each do |row|
    # You access the values via their column name (set by using headers option above)
    year = row["year"].to_i
    month = row["month"].to_i
    day = row["day"].to_i
    date = get_exact_date(year, month, day)
    oktoberfest_count = row["oktoberfest"].to_i
    elections_count = row["bundestagswahl"].to_i
    points_oktoberfest.add(date, oktoberfest_count, MONTH_NAMES[month])
    points_elections.add(date, elections_count, MONTH_NAMES[month])
  end
  # Create the plot
  @plot = GPlot.new(self)
  plot.set_dim(700, 300)
  plot.set_title_text("Oktoberfest vs. Bundestagwahl Google search history")
  plot.getXAxis.set_axis_label_text("Year")
  plot.getYAxis.set_axis_label_text("Google normalized searches")
  plot.getXAxis.setNTicks(10)
  plot.set_points(points_oktoberfest)
  plot.set_line_color(color(100, 100, 100))
  plot.add_layer("German elections day", points_elections)
  plot.get_layer("German elections day").set_line_color(color(255, 100, 255))
  plot.activate_point_labels
end

def draw
  background(255)

  # Draw the plot  
  plot.begin_draw
  plot.draw_box
  plot.drawXAxis
  plot.drawYAxis
  plot.draw_title
  plot.draw_grid_lines(GPlot::VERTICAL)
  plot.draw_filled_contours(GPlot::HORIZONTAL, 0)
  plot.draw_legend(%w(Oktoberfest Bundestagswahl),
    [0.07,0.22],
    [0.92, 0.92])
  plot.draw_labels
  plot.end_draw
end

def get_exact_date(year, month, day)
 leap_year = false
 if (year % 400 == 0)
  leap_year = true
 elsif (year % 100 == 0)
  leap_year = false
 elsif (year % 4 == 0)
  leap_year = true
 end
 if leap_year
  return year + (month + (day - 1) / DAYS_PER_MONTH_LEAP_YEAR[month]) / 12.0
  else
    return year + (month + (day - 1)/ DAYS_PER_MONTH[month]) / 12.0
  end
end

Since updated for JRubyArt

Using the grafica library in ruby-processing

Get the grafica library by jagracar here.

load_library :grafica

include_package 'grafica'

POINTS = 100

def setup
  size(500, 350)
  background(150)

  # Prepare the points for the plot
  points = GPointsArray.new(POINTS)

  POINTS.times do |i|
    points.add(i, 10 * noise(0.1 * i))
  end

  # Create a new plot and set its position on the screen
  plot = GPlot.new(self)
  plot.set_pos(25, 25)

  # Set the plot title and the axis labels
  plot.set_points(points)
  plot.getXAxis.setAxisLabelText("x axis")
  plot.getYAxis.setAxisLabelText("y axis")
  plot.setTitleText("A very simple example")

  # Draw it!
  plot.default_draw
end

The output

Since updated for JRubyArt

A custom Vector library for ruby processing

I recently got my copy of Practical Object Oriented Design in Ruby by Sandi Metz book, this got me thinking what could I could apply the ideas to in ruby-processing, and here is an early crack at it. Creating a custom (pure-ruby) vector library which can replace the hybrid RPVector (that extends PVector from processing) class of a previous post. This is very much a first crack (but it works, only difference needed is load_library :vec and use normalize! instead of normalize, much more ruby like I think) because I have ideas for extending the functionality along some of these lines, however my sentiment is with toxi re simple made easy. Further if I use vanilla processing logic the cross_product method (modulus, dist etc) could all live in Vec, but I can easily defer that decision. I haven't at this stage made use distance_squared externally, however this will be more efficient for testing boundary conditions of say a bouncing ball than regular dist. Not shown here are the rspec tests that I've been using to ensure the code behaves, I am tempted to bundle this and a Quaternion class and possibly some others as a core ruby-processing library, since the operations of PVector are not at all ruby like, and many more people seem to come to ruby-processing from ruby, not the other way round sadly.

class Vec
  attr_accessor :x, :y, :z
  EPSILON = 9.999999747378752e-05     # a value used by processing.org
  def initialize(x = 0 ,y = 0, z = 0)
    @x, @y, @z = x, y, z
    post_initialize
  end

  def post_initialize
    nil
  end

  def ==(vec)
    (x - vec.x).abs < EPSILON && (y - vec.y).abs < EPSILON && (z - vec.z).abs < EPSILON
  end
end


class Vec2D < Vec

  # Modulus of vec. Also known as length, size or norm
  def modulus
    Math.hypot(x, y)
  end

  def self.dist_squared(vec_a, vec_b)
    (vec_a.x - vec_b.x)**2 + (vec_a.y - vec_b.y)**2
  end

  def self.dist(vec_a, vec_b)
    Math.hypot(vec_a.x - vec_b.x, vec_a.y - vec_b.y)
  end

  # vanilla processing returns a Vector, rather than Scalar (defaults to 3D result when z = 0)
  def cross_product(vec)
    x * vec.y - y * vec.x
  end

  # Scalar product, also known as inner product or dot product
  def dot(vec)
    x * vec.x + y * vec.y
  end

  def collinear_with?(vec)
    cross_product(vec).abs < EPSILON
  end

  def +(vec)
    Vec2D.new(x + vec.x, y + vec.y)
  end

  def -(vec)
    Vec2D.new(x - vec.x, y - vec.y)
  end

  def *(scalar)
    Vec2D.new(x * scalar, y * scalar)
  end

  def / (scalar)
    Vec2D.new(x / scalar, y / scalar) unless scalar == 0
  end

  def normalize!
    @x, @y = x / modulus, y / modulus
    return self
  end

  alias :mag :modulus

end

class Vec3D < Vec

  def modulus
    Math.sqrt(x**2 + y**2 + z**2)
  end

  def self.dist_squared(vec_a, vec_b)
    (vec_a.x - vec_b.x)**2 + (vec_a.y - vec_b.y)**2 + (vec_a.z - vec_b.z)**2
  end

  def self.dist(vec_a, vec_b)
    Math.sqrt(self.dist_squared(vec_a, vec_b))
  end


  def cross_product(vec)
    xc = y * vec.z - z * vec.y
    yc = z * vec.x - x * vec.z
    zc = x * vec.y - y * vec.x
    Vec3D.new(xc, yc, zc)
  end

  # Scalar product, also known as inner product or dot product
  def dot(vec)
    x * vec.x + y * vec.y + z * vec.z
  end

  def collinear_with?(vec)
    cross_product(vec) == Vec3D.new
  end

  def +(vec)
    Vec3D.new(x + vec.x, y + vec.y, z + vec.z)
  end

  def -(vec)
    Vec3D.new(x - vec.x, y - vec.y, z - vec.z)
  end

  def * (scalar)
    Vec3D.new(x * scalar, y * scalar, z * scalar)
  end

  def / (scalar)
    Vec3D.new(x / scalar, y / scalar, z / scalar) unless scalar.abs < EPSILON
  end

  def normalize!
    @x, @y, @z = x / modulus, y / modulus, z / modulus
    return self
  end

  alias :mag :modulus
end

A Little Test courtesy of Daniel Shiffman

#
# Vector 
# by Daniel Shiffman.  
# 
# Demonstration some basic vector math: subtraction, normalization, scaling
# Normalizing a vector sets its length to 1.
#
load_library :vec

def setup
  size(640,360)
end

def draw
  background(0)

  # A vector that points to the mouse location
  mouse = Vec2D.new(mouse_x, mouse_y)
  # A vector that points to the center of the window
  center = Vec2D.new(width/2,height/2)
  # Subtract center from mouse which results in a vector that points from center to mouse
  mouse = mouse - center # note need assign result to mouse

  # Normalize the vector the ! means we are changing the value of mouse
  mouse.normalize!

  # Multiply its length by 150 (Scaling its length)
  mouse = mouse * 150 # note need assign the result to mouse

  translate(width/2,height/2)
  # Draw the resulting vector
  stroke(255)
  stroke_weight(4)
  line(0, 0, mouse.x, mouse.y)
end

Now interestingly the following also works so you can use the

+=, -=, /=, and *= assignment methods ( but you cannot override += etc as a methods unlike C++ )

So what this means is that in ruby += etc are just syntactic shortcuts, and not an operator in its own right (which is probably a good thing).

#
# Vector 
# by Daniel Shiffman.  
# 
# Demonstration some basic vector math: subtraction, normalization, scaling
# Normalizing a vector sets its length to 1.
#
load_library :vec

def setup
  size(640,360)
end

def draw
  background(0)

  # A vector that points to the mouse location
  mouse = Vec2D.new(mouse_x, mouse_y)
  # A vector that points to the center of the window
  center = Vec2D.new(width/2,height/2)
  # Subtract center from mouse which results in a vector that points from center to mouse
  mouse -= center  # note we can assign result to mouse using -=

  # Normalize the vector the ! means we are changing the value of mouse
  mouse.normalize!

  # Multiply its length by 150 (Scaling its length)
  mouse *= 150 # note we can assign result to mouse using *=

  translate(width/2,height/2)
  # Draw the resulting vector
  stroke(255)
  stroke_weight(4)
  line(0, 0, mouse.x, mouse.y)

end

Checking out jruby-complete-9000.dev.jar (waiting for jruby-1.7.5)

I just tested the 2 October 2013 snapshot of jruby-complete-9000 with ruby-processing, so far so good (still issue with fisica library that I had thought had gone away with jruby-1.7.5.dev, but I might have been using my "corrected" version of the fisica library, anyway I will be interested to see the effect of Charlies pruning basically removing ruby-1.8.7 compatibility). I've also been reviewing "app.rb", with a view to being a bit more specific about which processing classes to import. Having failed abysmally to make any use of processing event classes directly, they are in firing line. Also I will only import opengl classes for P2D and P3D sketches (work in progress, will be ready for next ruby-processing commit to coincide with jruby-1.7.5 release). On the processing front a new release of vanilla processing may not be too far away either (at long last, and then probably only because of the Mac Retina display issue, java version is going to be upgraded to version 1.7.0_40).