Commit Graph

  • 07830b303f Nixed a bunch of redundant casts. Michael Bayne 2007-10-17 21:52:40 +00:00
  • e0a22877e2 This seems like its been pretty happy on ice, so take off the training wheels Charlie Groves 2007-10-12 20:11:00 +00:00
  • 8bc738c6d8 If a scene block resolution somehow gets abandoned, remove it from the list of blocks we're awaiting. I'd like to believe this is related to the mysterious YPP missing scene block bugs we've seen forever, but since I can't reproduce it - no idea if it actually is. Mike Thomas 2007-10-10 18:15:15 +00:00
  • 59186e034f Added onStart function; cleaned up start(). Michael Bayne 2007-10-09 07:30:35 +00:00
  • 6f67eac059 Brought my path stuff into general purpose land and improved it in the process. Michael Bayne 2007-10-09 07:17:01 +00:00
  • 094e6b1730 Dirty SceneObjectTips as they're laid out. This was relying on tips being added and laid out before the initial paint. Charlie Groves 2007-10-08 23:00:20 +00:00
  • b4417cc330 Add a switch to disable dirtying the whole screen on an invalid component on the Mac again. I can't reproduce the problem now, and I might have just been seeing some sea monster badness before. Leave the current behavior on by default though. Charlie Groves 2007-10-03 22:10:28 +00:00
  • 1423c03890 Incidental genericity Charlie Groves 2007-10-03 21:26:25 +00:00
  • ddb3c92fa1 Let's not make things that actually compose stuff hop about the screen by using their VolatileMirage offset in the place of their origin offset. Charlie Groves 2007-09-26 22:18:27 +00:00
  • c45878681f If we're creating a CompositedMultiFrameImage with only a single source image and it's a TileSetFrameImage, just use the Mirages in that TileSet rather than creating a whole new set of BufferedImages to composite them into. This halves the image memory used by anything that's just colorizing and transforming its images and doesn't really need composition. Charlie Groves 2007-09-26 21:49:11 +00:00
  • 8d5fc6c6ea Rolling back r292 with great sadness in my heart. The non-repainting doesn't happen very often, but sure enough it does and sure enough it's awful. Charlie Groves 2007-09-03 22:45:29 +00:00
  • 68bdd9402c Crank the CharacterManager cache up too so we're not compositing our sea monsters over and over again. Charlie Groves 2007-09-01 00:45:40 +00:00
  • 1e55b4ed7e Haven't been able to see any non-repainting without this on my mac, and this was causing the whole screen to be dirtied once a second when the countdown timer in a seamonster battle ticked. Charlie Groves 2007-08-31 17:18:35 +00:00
  • 65b8cdbc3a Factor in the percentage of the databuffer a BufferedMirage is using to estimate its memory if its BufferedImage is a subimage of another BufferedImage. Charlie Groves 2007-08-30 22:50:24 +00:00
  • 37b5e58961 If a MultiTileSprite overlaps with another item, take their render priorities into account. Mike Thomas 2007-08-30 18:50:16 +00:00
  • a1332d45ff Use junit4.jar. Michael Bayne 2007-08-28 20:26:11 +00:00
  • 11322269c2 Redraw seems to work correctly on the Mac now, so only invalidate the shifted region like on other platforms. Charlie Groves 2007-08-28 17:07:39 +00:00
  • 95c36c22a2 Explanatory comment. Mike Thomas 2007-08-22 22:26:36 +00:00
  • b560a045cd Add ability for a sprite to specify that it resides in a rectangular region of tiles rather than assuming all sprites only occupy one tile. Mike Thomas 2007-08-22 22:24:54 +00:00
  • 9d73fe0587 This needs to respect sprite offset. Mike Thomas 2007-08-22 22:22:59 +00:00
  • e003d97535 Break the code for completing a fade into a method so subclasses can override it and do what they will when it finishes. Charlie Groves 2007-08-10 22:00:50 +00:00
  • 069fe85c70 A whole lot of widening, refactoring, thinking, rethinking and fiddling around for what ultimately boils down adding this method to BundledComponentRepository: Michael Bayne 2007-08-03 22:04:04 +00:00
  • bbd5dd284e Added version of getTile() that takes a rectangle into which to write the results. Michael Bayne 2007-08-03 21:10:48 +00:00
  • 205d57a7fe Allow TileSetRuleSets to specify a custom tileset tag name, and use that to allow UniformTileSets as <uniformTileset> in addition to the existing SwissArmyTileSets as <tileset> in the bundle tasks. Charlie Groves 2007-08-02 23:18:16 +00:00
  • 1f435d582d If our build process fails to properly trim a tile set, it's kinda a big deal, so throw a BuildException so ant hears about it and stops. Mike Thomas 2007-07-30 20:59:32 +00:00
  • 9ba5a61ab0 Have distclean just call what it needs in common-clean Jonathan Le Plastrier 2007-07-28 04:15:13 +00:00
  • 72d93f6bd9 One more for the Gipper. Michael Bayne 2007-07-25 21:47:15 +00:00
  • 4ea0d15236 Infoization. Michael Bayne 2007-07-25 21:46:42 +00:00
  • 81dcd90c1a Simplify and combine fading out for both moving and stationary sprites. Dave Hoover 2007-07-18 18:47:37 +00:00
  • 0670fe26e7 Checking the value of a Loader's contentLoaderInfo.childAllowsParent itself issues a security warning. This is the value to check to SEE if a security exception would be thrown. So things worked, but now that I'm testing security locally I cannot tolerate 10,000 warnings about security filling my trace log, obliterating any other debug logging. Ray Greenwell 2007-07-17 00:45:19 +00:00
  • b5b3a5316e Christmas comes but once a year, but svn keyword substitution day comes whenever the urge strikes! Dave Hoover 2007-07-14 00:27:00 +00:00
  • f57c408d7f If something goes wrong adding our tileset to the bundle, throw an exception so that things fail & attract attentino, rather than trying to work around it and printing a stack trace that will likely go unnoticed amidst a slew of other build messages. Dave Hoover 2007-07-13 18:07:29 +00:00
  • 7fdd8fee17 Add the ability to fade the sprite out without a path and fix fading out with a path so that it's actually fading out instead of fading in Charlie Groves 2007-07-11 23:49:04 +00:00
  • 7efd53a76c Use _minAlpha to figure out when to stop moving alpha down, don't readd the sprite in tick since shutdown is taking care of it also Charlie Groves 2007-07-09 22:49:48 +00:00
  • 603d6bb0d2 It's a security error to specify a non-null SecurityDomain to Loader when in a local sandbox. Annoying. Ray Greenwell 2007-07-05 02:39:10 +00:00
  • 640b6fe1f6 Reset the up and down timers in reset and allow the maximum level of gleam and update frequency to be specified Charlie Groves 2007-06-29 22:41:52 +00:00
  • 717dc79c02 Tabs -> spaces Dave Hoover 2007-06-26 00:07:47 +00:00
  • 0250be64ac Don't turn on variableRowHeight for every menu, but do have submenus inherit their parent's setting. Ray Greenwell 2007-06-16 00:33:48 +00:00
  • 7a6d334a99 Allow an already-instantiated IFlexDisplayObject to be used as the icon for a menu item. JFC, this is pretty straightforward but was a fiasco to get right because I was using a MediaWrapper as the icon, and something wasn't playing nicely (not yet debugged), but the error looked like it was with the Menu superclass. Ray Greenwell 2007-06-16 00:09:23 +00:00
  • 25d48696ee Update masking behavior: configure the mask to be the lesser of the maximum size and the actual SWF frame size in each dimensions. Ray Greenwell 2007-06-14 22:23:43 +00:00
  • 300d14c438 Some layout fixes. These buttons are still really ugly, alas. Michael Bayne 2007-06-14 21:45:42 +00:00
  • 767a82128e Actually do keyword substitution on those freshly headered files. Dave Hoover 2007-06-14 18:20:16 +00:00
  • 7a65a62927 Widening. Michael Bayne 2007-06-14 17:42:52 +00:00
  • 46248c0354 All of our code should have headers, but our redistributable code should definitely have headers. Michael Bayne 2007-06-14 17:40:30 +00:00
  • 247bf757eb Scratch that, reverse it. Ray Greenwell 2007-06-14 00:50:37 +00:00
  • 6b8ab48d17 Pass the url to configureMask(). Ray Greenwell 2007-06-14 00:49:12 +00:00
  • dcc950e9d5 Added missing boilerplate. Robert Zubeck 2007-06-12 01:40:24 +00:00
  • 13f361d8aa Clamping is a useful abstraction, and easy to slip up on. Maybe more math idioms will end up in here as well? Robert Zubeck 2007-06-12 01:39:29 +00:00
  • 7ca448a91e Let us optionally not auto-reposition TransformedMirages; it was being smart for things like scales and rotates so 0,0 still meant the same place, but that meant I couldn't just reposition something, since it would promptly undo the transform I passed in. Dave Hoover 2007-06-01 18:08:34 +00:00
  • 9412b19145 Fix the docs link and add a link to narya. Jonathan Le Plastrier 2007-05-30 22:27:02 +00:00
  • 97f6eb7c85 Switch to Java's logging facilities. Michael Bayne 2007-05-30 01:18:06 +00:00
  • 59c6998c30 Allow subclasses to choose not to call handleFrame() when ADDED_TO_STAGE is received. Ray Greenwell 2007-05-28 01:17:37 +00:00
  • 44d35eb9d3 Moved our custom hit-testing here from MsoySprite. I was hoping to have our nice transparent-pixels-don't-count hitting for every media container, including any game icon decoration over an avatar's head, but after moving this I remembered that it may be pointless, as it *never matters*- flash simply doesn't use this method to test mouse-overness. So it may be moot, unless we do mouse-over polling like we do in the RoomView, which is too expensive to do everywhere.. Ray Greenwell 2007-05-18 20:40:24 +00:00
  • 95f14a3c03 - Planted the TextField constants here. You know: thee ones that Adobe thought it would be fun for everyone to guess. - Added a static method to track TextFields and ensure that only one of those tracked ever has a selection. - On linux, automatically put selected text (in one of these tracked TextFields) into the clipboard. Untested: it at least doesn't work with the standalone player. Ray Greenwell 2007-05-16 22:48:19 +00:00
  • c49e3102e9 Default to unselectable text. Ray Greenwell 2007-05-16 22:45:46 +00:00
  • 91c45db381 Let's do this in a way that allows more than a single value to be used in the natural comparison. Michael Bayne 2007-05-15 17:50:16 +00:00
  • acea29604d Make AbstractMedia implement Comparable instead of using a comparator. Added naturalOrder() which allows media to use a more consistent "same render order" order resolution value than Object.hashCode(). Michael Bayne 2007-05-15 17:45:18 +00:00
  • f7dfe51a9c Oh yeah, we need to call that method. Michael Bayne 2007-05-15 17:05:04 +00:00
  • 9a30debf42 Type safety and widening. Michael Bayne 2007-05-15 16:59:58 +00:00
  • c755fd626a To reduce shader state changes and compilations, just use the four-bone-per-vertex configuration for all skinned meshes. Depending on the models we have, it may be worth compiling a separate one for two bones per vertex. Added a hook to customize the material color (for per-vertex colors, e.g.) Andrzej Kapolka 2007-05-15 00:55:19 +00:00
  • 124af1e0b1 Use a fragment shader as well as a vertex shader for the skinned meshes. I tracked the ATI slowdown down to using a vertex shader in combination with fixed function fog. Also cleaned up ShaderConfig a little. Andrzej Kapolka 2007-05-14 20:13:39 +00:00
  • 6b07087c31 32 matrices is just a feeeeew two many for ATI cards, but of course their drivers can't do anything useful like report an error; they just silently screw up. Limiting it to 31 makes it work, but there's still a strange slowdown issue (and it's not that the shaders are being run in software, but it's worth checking for that anyway). Andrzej Kapolka 2007-05-12 00:57:35 +00:00
  • 12e17a9e33 Remove debug logging. Michael Bayne 2007-05-11 23:28:00 +00:00
  • 82ba950111 Go back to using byte buffers; that wasn't the problem. Andrzej Kapolka 2007-05-11 18:30:51 +00:00
  • 345895c32c Comment fixup. Ray Greenwell 2007-05-10 23:47:29 +00:00
  • ab892bcd9d Handles some boilerplate functionality. Ray Greenwell 2007-05-10 23:46:34 +00:00
  • fd192268c3 Try using shorts rather than bytes for bone indices. Andrzej Kapolka 2007-05-10 23:21:15 +00:00
  • 8a0168760f Much work to reconfigure shaders automatically based on the model's current JME state set. Andrzej Kapolka 2007-05-10 20:35:27 +00:00
  • 1d8f04c43a Prepend the #version directive before compiling the shader. Andrzej Kapolka 2007-05-10 01:31:42 +00:00
  • e5db96b963 Ignore cull mode when we're generating frames to morph between. Andrzej Kapolka 2007-05-10 01:07:54 +00:00
  • 2e0678877b Use the hacky PerfTimer if available and fall back to the NanoTimer if we're running in a sandbox. Michael Bayne 2007-05-09 21:53:42 +00:00
  • 20fe53d07f Did I leave the stove on? Dave Hoover 2007-05-09 20:11:02 +00:00
  • 45df6d814b Use VBOs for vertices, indices when using shaders; multiply vertices by weights even if we only have one weight per vertex. Andrzej Kapolka 2007-05-09 19:51:32 +00:00
  • 0024a02067 Wow, the media framework gets into a bad way if it's forced to use the inaccurate System.currentTimeMillis() timer. Fortuately Sun introduced System.nanoTime() in 1.5 which provides civilized timing information, so we can use that instead of sun.misc.Perf which is unavailable in an applet sandbox. We will need to go in and try to cope as best we can with currentTimeMillis() if we want to support unsigned play in JDK 1.4 or earlier. We'll do that later. Michael Bayne 2007-05-09 19:22:10 +00:00
  • ef35b37648 We need to create a new array when we're *not* using shaders. Andrzej Kapolka 2007-05-08 01:56:53 +00:00
  • 742dc30f9e Basic support for skinning with GLSL shaders. Still need to work out how to handle different lighting arrangements and complex materials (e.g., the metal effect for the Iron Plate). Andrzej Kapolka 2007-05-08 01:50:07 +00:00
  • 8a2e67831e Also need to cope with non-component _root in BackFrameManager Mike Thomas 2007-05-04 23:14:26 +00:00
  • 2c6fdf5d94 Cope if our ManagedRoot is not a component. Michael Bayne 2007-05-04 23:08:08 +00:00
  • 3401bbf3bf Don't skin what you can't see. Andrzej Kapolka 2007-05-04 01:25:40 +00:00
  • d43eabc156 We should be removing ourselves from our parent when we're finished. Ray Greenwell 2007-05-04 00:41:32 +00:00
  • bb54ec1888 Optimize vertex order for cache usage and use a better data structure to (re)index vertices. Andrzej Kapolka 2007-05-04 00:39:54 +00:00
  • 095c718244 Render to the applet instead of its parent window since that doesn't work. Michael Bayne 2007-05-03 22:56:29 +00:00
  • ab6252cee7 Widened. Added getMirage(String). Michael Bayne 2007-05-03 21:13:32 +00:00
  • 277b62bd52 Don't merge meshes where, in any animation, one is visible and the other isn't. Andrzej Kapolka 2007-05-02 18:19:09 +00:00
  • a855fa8340 Comment fixup. Ray Greenwell 2007-05-02 00:13:29 +00:00
  • 330a99f337 Added isContentInitialized(), which should be used to help differentiate between an MsoySprite that does not have any WhirledControl subclass in it and those that do, but are just not yet fully loaded. Ray Greenwell 2007-05-02 00:10:00 +00:00
  • 072926f814 Merge meshes in animated models by finding ones that maintain the same relative position throughout all animations. Also, don't include transforms in the animations for nodes that never move. Andrzej Kapolka 2007-05-01 23:36:23 +00:00
  • 86298ca832 I've decided I like setCallback() better than setFunction(). Ray Greenwell 2007-05-01 16:32:27 +00:00
  • 353d7213d0 Bugfix for animations that complete instantly. Ray Greenwell 2007-04-27 23:06:32 +00:00
  • 250670e699 Default to using S3TC texture compression. Andrzej Kapolka 2007-04-27 20:48:48 +00:00
  • 1d04c51070 Use TextFieldUtil. Ray Greenwell 2007-04-27 01:28:28 +00:00
  • f50613a913 Utility methods to attempt to make initializing a TextField easier. Ray Greenwell 2007-04-27 01:27:54 +00:00
  • fb1eb6026b Call updateAnimation() immediately on any starting animation. Ray Greenwell 2007-04-27 00:51:50 +00:00
  • 5f83a3bbcf Revamping animations to be based off of ENTER_FRAME rather than using a rapidly-expiring Timer. Cleaned up some other bits. Ray Greenwell 2007-04-26 23:52:25 +00:00
  • 79c388a906 Basic 3D vector library. Robert Zubeck 2007-04-25 23:13:16 +00:00
  • c48aa541e8 Added hook for subclasses to initialize things prior to loading a new piece of media. Ray Greenwell 2007-04-24 23:25:03 +00:00
  • 74d0b7286b More text animations, these are capable of animating each character. Ray Greenwell 2007-04-24 22:56:26 +00:00
  • 03e3b1b91c Some fixups. I think I'm going to want to refactor this so that there are util methods for creating nice textfields, and then generic animation classes for floating anything. This can then become a convenience class that marries the two. Ray Greenwell 2007-04-24 22:55:29 +00:00
  • 476334e5ee It's ok to BE null, it just can't be set to null. Ray Greenwell 2007-04-24 21:28:49 +00:00
  • 97ff4caa41 Working towards some easy-to-use score animation type things. Actionscript's lack of method overloading sucks. I can't have more than one constructor, and factory methods will have to have different names. Ray Greenwell 2007-04-24 21:22:54 +00:00