We closed this forum 18 June 2010. It has served us well since 2005 as the ALPHA forum did before it from 2002 to 2005. New discussions are ongoing at the new URL http://forum.processing.org. You'll need to sign up and get a new user account. We're sorry about that inconvenience, but we think it's better in the long run. The content on this forum will remain online.
IndexDiscussionGeneral Discussion,  Status › Processing 0149 is the phoenix
Page Index Toggle Pages: 1
Processing 0149 is the phoenix (Read 1004 times)
Processing 0149 is the phoenix
Oct 15th, 2008, 2:41pm
 
Time for some burning and rebirthing...


ABOUT REV 0149 - 15 October 2008

Very large number of bug fixes and changes. All renderers that use
PGraphics will need to be updated for this release. That means you'll
need a new GLGraphics, P5Sunflow, etc. Lots of libraries will need to
be updated for 0149.

Important! library and tool folders have changed--see notes below.

This release also adds command-line support, which means that Processing
can be integrated into all manner of environments. More info below.

Also many, many updates to the examples and reference in this release.

* RELEASES SINCE 0135 CONTAIN MANY SIGNIFICANT CHANGES, PLEASE READ *
*  THE NOTES IN THIS FILE FOR ANY RELEASE THAT YOU'VE NOT YET USED. *

[ libraries & tools ]

+ Beginning with this release, all libraries should be placed in the sketchbook
 folder, inside a folder called "libraries". It's a bad idea to put them in
 the Processing libraries folder, because 1) users may not have access to that
 folder in lab environments, and 2) it's now hidden on OS X (see below).

+ Also in this release, all tools should be placed in a folder named "tools"
 inside your sketchbook folder. Same reasons as above.

[ launchers ]

+ The Mac OS X release is now a single .app file, simplifying the install and
 bringing us into closer Steve compliance.

+ The Windows release has a new launcher based on a hacked/tweaked version
 of launch4j-3.0.1, an excellent app bundling tool for Windows. Many thanks
 to its developer(s) for their hard work. http://launch4j.sourceforge.net/

+ Fixed a Windows problem with the JDK not being found, even when installed.
 http://dev.processing.org/bugs/show_bug.cgi?id=878

+ Fixed a problem with older/badder JDKs being used on Windows.
 http://dev.processing.org/bugs/show_bug.cgi?id=545

+ Made PDE files double-clickable on Windows.
 http://dev.processing.org/bugs/show_bug.cgi?id=683
 By default we now set the registry associations to launch the most recently
 run version of processing.exe (it must be called that, at least for now)
 when a .pde file is double-clicked. The behavior can be disabled in the
 Preferences dialog.

+ With the new launcher, we even have a splash screen (the about box) that
 shows up while Java is roused from bed.
Re: Processing 0149 is the phoenix
Reply #1 - Oct 15th, 2008, 2:42pm
 
[ changes ]

+ Changed how PGraphics objects are created for better consistency. Sadly,
 this will require any libraries that subclass PGraphics to be rebuilt
 for this release. (However, internally it's a great improvement.)
 Renderer library authors please read the reference for PGraphics:
 http://dev.processing.org/reference/core/javadoc/processing/core/PGraphics.html
 For library developers, a gritty list of the internal function changes
 can be found in processing/core/done.txt in the revision 0149 section under
 the heading "PGraphics API changes". (It's not poetic, but it's real.)

+ We've decided that vector geometry is as important as image data, so the
 Candy library has been removed, and replaced by a built-in loadShape()
 method in PApplet, which returns a PShape object. (Candy lives on as the
 PShapeSVG class, which is the PShape subclass created by loadShape.)
 A shape() command has been added (like the image() command for images).
 See the new SVG examples for more information. In the future, we'd like
 to support other vector formats (such as OBJ) with loadShape(), or
 provide a means for libraries to register themselves to load shape data.

+ The PShape object is just a container for now, but will someday provide
 methods for accessing actual vertex information.

+ Because of the SVG changes, the XML library is now built into the core.
 This means that processing.xml will be imported by default (no need for
 "import processing.xml" in your sketch). This also means that "XML"
 will no longer be seen in the "Import Library" menu.

+ Altered how createInput(), save(), and saveFrame() work so that they
 are more friendly to signed applets.

+ Implemented new error/warning system inside PGraphics so that errors
 are only shown once, and non-fatal errors do not shut down the sketch.

+ The unhint() method has been removed, see the hint() reference for changes.

+ Changed how match() works to bring back in line with other regexp
 implementations. Instead of only returning matching groups, match 0
 is the matching sequence, and array elements [1] and higher represent
 each of the groups (sequences found in parens). This brings us into
 compliance with how "everyone else" does it. Please direct complaints,
 jeers, or accolades to Dan Shiffman, who advocated the change.

+ Also starting in this release, the match() commands use multiline
 matching (Pattern.MULTILINE) and dotall mode (Pattern.DOTALL) by default
 so that ^ and $ match the beginning and end of any lines found in the
 source, and the . operator will also pick up newline characters.

+ The arraycopy() method is now arrayCopy(), for better consistency with
 the rest of the API (think framerate vs. frameRate). Because books have
 been printed with the lowercase version, we're keeping both, but in
 Java parlance, we've deprecated the lowercase version.

+ Removed MACOS9 constant from processing.core.PConstants. Bye bye 2001.

+ Deprecated the NORMALIZED constant, use NORMAL instead.

+ Removed some items from the default imports because they're 1) likely
 to cause naming collisions and 2) more likely to be used by advanced
 users (who can add 'em themselves anyway). The removed items include:
 javax.sound.midi,javax.sound.midi.spi,
 javax.sound.sampled,javax.sound.sampled.spi,
 javax.xml.parsers,javax.xml.transform,javax.xml.transform.dom,
 javax.xml.transform.sax,javax.xml.transform.stream,
 org.xml.sax,org.xml.sax.ext,org.xml.sax.helpers
 You can add default imports to your preferences.txt file by adding to
 the preproc.imports line. But keep in mind that this will make your
 code incompatible with Processing on any machine that doesn't have
 its preferences.txt mangled in the same manner.

+ Update Quaqua to 4.4.7 on Mac OS X (http://www.randelshofer.ch/quaqua/)

+ Disable abyssmally slow Sun renderer on OS X 10.5, by setting
 apple.awt.graphics.UseQuartz to true inside PApplet.main().
Re: Processing 0149 is the phoenix
Reply #2 - Oct 15th, 2008, 2:42pm
 
[ bug fixes ]

+ Prevent caps/joins from throwing an error and halting the render.
 This was a problem when drawing SVG files with OpenGL:
 http://dev.processing.org/bugs/show_bug.cgi?id=628
 However, complex shapes or shapes with holes will produce an error in OpenGL:
 http://dev.processing.org/bugs/show_bug.cgi?id=947

+ The point() command should now work always and consistently
 in OpenGL as well as Java 2D.
 http://dev.processing.org/bugs/show_bug.cgi?id=121

+ Fix GLDrawableFactory.chooseGraphicsConfiguration() error with OpenGL
 http://dev.processing.org/bugs/show_bug.cgi?id=891
 http://dev.processing.org/bugs/show_bug.cgi?id=908

+ Fix problem with error line highlighting (wrong line being highlighted)
 http://dev.processing.org/bugs/show_bug.cgi?id=888

+ Make error messages clear the message area.
 http://dev.processing.org/bugs/show_bug.cgi?id=912

+ "IllegalMonitorStateException: current thread not owner" with delay()
 http://dev.processing.org/bugs/show_bug.cgi?id=894

+ Fix "Show Sketch Folder" on Linux, and prevent launcher error whenever
 exporting a sketch on Linux. (Regression since ~0140)

+ Make parseFloat() with String[] array return NaN for missing values
 http://processing.org/discourse/yabb_beta/YaBB.cgi?board=Syntax;action=display;num=1220880375

+ Renaming the main tab was adding .pde to the sketch folder name
 http://dev.processing.org/bugs/show_bug.cgi?id=922

+ Fix problems with text in a rectangle
 http://dev.processing.org/bugs/show_bug.cgi?id=893
 http://dev.processing.org/bugs/show_bug.cgi?id=899

+ Remove extra files that were accidentally included inside core.jar.

+ Tool menu not showing until the example (Mangler) was compiled.
 http://dev.processing.org/bugs/show_bug.cgi?id=892

+ Hopefully fixed a "transport error 202" error.
 http://dev.processing.org/bugs/show_bug.cgi?id=895

+ hint(DISABLE_DEPTH_TEST) now clears the depth buffer so that there's no
 longer a need for a kludgey workaround (messing with g.zbuffer or dropping
 into OpenGL code to call gl.glClear(GL.GL_DEPTH_BUFFER_BIT))
 http://dev.processing.org/bugs/show_bug.cgi?id=827

+ Header not always updating on rename... this was an old bug that hasn't
 been seen or confirmed recently, closing it.
 http://dev.processing.org/bugs/show_bug.cgi?id=56

+ Add additional newline hack so that Auto Format has less errors.

+ Fix video capture problems with OpenGL on Mac OS X.
 http://dev.processing.org/bugs/show_bug.cgi?id=882

+ Sketch export results in 100x100 default size, regardless of size() setting
 http://dev.processing.org/bugs/show_bug.cgi?id=945

+ Fix case where creating a new tab throws a NullPointerException.
 http://dev.processing.org/bugs/show_bug.cgi?id=940

+ Disable image smoothing with default renderer (JAVA2D) until smooth() called
 http://dev.processing.org/bugs/show_bug.cgi?id=685

+ Call flush() to emit any geometry before addFrame() in MovieMaker
 http://dev.processing.org/bugs/show_bug.cgi?id=692
Re: Processing 0149 is the phoenix
Reply #3 - Oct 15th, 2008, 2:42pm
 


[ additions ]

+ Command-line support is now implemented. See the reference for the new
 Commander class for details on how to use it:
 http://dev.processing.org/reference/everything/javadoc/processing/app/Commander.html
 http://dev.processing.org/bugs/show_bug.cgi?id=219
 http://dev.processing.org/bugs/show_bug.cgi?id=889

+ PVector is a simple 3-dimensional vector (x, y, z) or point class,
 based on code from Dan Shiffman. No more roll-your-own vectorishness.

+ PMatrix interface, along with PMatrix2D (3x2 affine) and PMatrix3D (4x4)
 implementations for use with the renderers (and your own code).

+ Added matchAll() method, which is identical to match(), except that
 it returns an array of all matches in the specified String, rather
 than just the first. (Thanks to Dan Shiffman for the implementation.)

+ Add new "Getting Started" to the help menu
 http://processing.org/learning/gettingstarted/
 The old Getting Started item has been renamed to "Environment".

+ Added getImage() method to PImage, which returns it as a java.awt.Image.
 (Actually it's a BufferedImage, but who's keeping track?)

+ Added a note to prefs dialog that multiple jar export only works w/o libs
 http://dev.processing.org/bugs/show_bug.cgi?id=907

[ known problems ]

+ Slow response on file/sketchbook/examples menus on OS X 10.5. This is
 an Apple bug in Java 1.5 and 1.6 on Mac OS X 10.5 causing these menus
 to be extremely slow. With Apple's Java for Mac OS X 10.4 Release 7,
 they appear to have successfully backported the bug to OS X 10.4.
 Hooray! Now it affects both 10.4 and 10.5. Much success to them for
 their help. Some workarounds are described in the bug report:
 http://dev.processing.org/bugs/show_bug.cgi?id=786
 Please file a bug report with Apple if you want this fixed.

+ See dev.processing.org/bugs for more.
Page Index Toggle Pages: 1