78816 2012-02-16 08:01:04 -0800 [Qt] Previous web page appears outside content rect 2012-03-13 05:37:36 -0700 1 1 1 Unclassified WebKit WebKit2 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 kenneth kenneth hausmann jturcotte noam webkit.review.bot oldest_to_newest 558193 0 kenneth 2012-02-16 08:01:04 -0800 SSIA 558198 1 127382 kenneth 2012-02-16 08:08:59 -0800 Created attachment 127382 Patch 560174 2 127382 jturcotte 2012-02-20 04:32:40 -0800 Comment on attachment 127382 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=127382&action=review > Source/WebKit2/WebProcess/WebPage/qt/LayerTreeHostQt.cpp:450 > + HashSet<WebCore::WebGraphicsLayer*>::iterator end = m_registeredLayers.end(); > + for (HashSet<WebCore::WebGraphicsLayer*>::iterator it = m_registeredLayers.begin(); it != end; ++it) > + (*it)->didRenderFrame(); This assumes that all the tile creation (which happen one tile at a time on a 0-timer) and rendering is going to be completed within one rendering loop to and from the UI process. It might work since this delay can be long, but it might break and cause white-flashing while committing the scale in cases where this delay isn't long enough and the removeTile calls for the previous backing store get to the UI process before the createTile of the new backing store. There would need some weird situation for this to happen though. One way this could work is if the tiledbackingstore became completely synchronous (which might not be a bad thing since we don't need it to be async in WebKit2 AFAIK), or I think we should rely on a trigger to destroy the previous tiles that will work 100% of the time. Maybe for now a reliable way could be to check the main and previous backing store at the end of updateContentBuffers for TiledBackingStore::coverageRatio of the visible rect and decide if it is alright to clear the previous backing store. 560190 3 kenneth 2012-02-20 05:05:31 -0800 How can this coverageRatio work? You want to check whether it is more than 1.0 (100%) meaning that the previous buffer is larger than the new one. But even if we are loading a new page which is smaller than the previous one we might not always get a coverageRatio larger than 1.0 - or am I missing something? 560198 4 jturcotte 2012-02-20 05:33:01 -0800 (In reply to comment #3) In reference to http://trac.webkit.org/changeset/94678 I was using "coverageRatio >= 1.0f" mainly because I've always had this weird fear of floats slipping through my fingers, but it should just be "== 1.0f". The method outputs a float between 0 and 1 being "how much percentage of the rect you gave me is covered by visible tiles in this backing store". The ratio was always calculated on the visible rect since this is where the user would see the empty-content-flashing after committing a scale. So in that case, the moment that "mainBackingStore.coverageRatio(visibleRect) == 1.0" I would clear the previous tiles. There was also a special case to handle where doing two quick scale commit would replace a full previousBackingStore with an almost empty mainBackingStore. 561027 5 127999 noam 2012-02-21 10:54:04 -0800 Created attachment 127999 Patch 561282 6 127999 webkit.review.bot 2012-02-21 15:30:36 -0800 Comment on attachment 127999 Patch Clearing flags on attachment: 127999 Committed r108407: <http://trac.webkit.org/changeset/108407> 561283 7 webkit.review.bot 2012-02-21 15:30:41 -0800 All reviewed patches have been landed. Closing bug. 577207 8 131581 larsgk 2012-03-13 03:20:57 -0700 Created attachment 131581 Test page exposing what happens when clipping is done with no respect for 3D transforms. To me, it seems wrong to do the clipping here - at least when it's done with no respect for 3D transforms. I might be wrong though as I am just getting into this part of the code. 577250 9 jturcotte 2012-03-13 05:37:36 -0700 (In reply to comment #8) Could you add a screenshot or explain the result? It seems to work fine on my build. 127382 2012-02-16 08:08:59 -0800 2012-02-21 10:54:04 -0800 Patch bug-78816-20120216170857.patch text/plain 3091 kenneth U3VidmVyc2lvbiBSZXZpc2lvbjogMTA3OTEzCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggZGVhMjcwYmUzZWJkNjc2 ZDI1NjQwNDIwMjFlYjQxYmNjMWY3ZTFhYi4uY2YyYzBlYTU4MzgxM2E3NTUyYjU4Y2UzZjNhYzFh ZDI5MWZlNmMxYiAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDIxIEBACisyMDEyLTAyLTE2ICBLZW5u ZXRoIFJvaGRlIENocmlzdGlhbnNlbiAgPGtlbm5ldGhAd2Via2l0Lm9yZz4KKworICAgICAgICBb UXRdIFByZXZpb3VzIHdlYiBwYWdlIGFwcGVhcnMgb3V0c2lkZSBjb250ZW50IHJlY3QKKyAgICAg ICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTc4ODE2CisKKyAgICAg ICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgSW50cm9kdWNlIGEgZGlk UmVuZGVyRnJhbWUgbWV0aG9kIGluIFdlYkdyYXBoaWNzTGF5ZXIgd2hpY2gKKyAgICAgICAgaXMg dXNlZCB0byBwdXJnZSB0aGUgcHJldmlvdXMgYmFja2luZyBzdG9yZS4KKworICAgICAgICAqIFdl YlByb2Nlc3MvV2ViQ29yZVN1cHBvcnQvV2ViR3JhcGhpY3NMYXllci5jcHA6CisgICAgICAgIChX ZWJDb3JlOjpXZWJHcmFwaGljc0xheWVyOjpkaWRSZW5kZXJGcmFtZSk6CisgICAgICAgIChXZWJD b3JlKToKKyAgICAgICAgKiBXZWJQcm9jZXNzL1dlYkNvcmVTdXBwb3J0L1dlYkdyYXBoaWNzTGF5 ZXIuaDoKKyAgICAgICAgKFdlYkdyYXBoaWNzTGF5ZXIpOgorICAgICAgICAqIFdlYlByb2Nlc3Mv V2ViUGFnZS9xdC9MYXllclRyZWVIb3N0UXQuY3BwOgorICAgICAgICAoV2ViS2l0OjpMYXllclRy ZWVIb3N0UXQ6OnJlbmRlck5leHRGcmFtZSk6CisKIDIwMTItMDItMTYgIFBhdHJpY2sgR2Fuc3Rl cmVyICA8cGFyb2dhQHdlYmtpdC5vcmc+CiAKICAgICAgICAgW0NNYWtlXSBBZGQgbWlzc2luZyBp bmNsdWRlIGRpcmVjdG9yaWVzLgpkaWZmIC0tZ2l0IGEvU291cmNlL1dlYktpdDIvV2ViUHJvY2Vz cy9XZWJDb3JlU3VwcG9ydC9XZWJHcmFwaGljc0xheWVyLmNwcCBiL1NvdXJjZS9XZWJLaXQyL1dl YlByb2Nlc3MvV2ViQ29yZVN1cHBvcnQvV2ViR3JhcGhpY3NMYXllci5jcHAKaW5kZXggNWU5N2Zk NTQzZWE1MjdhZDI5NjkzZWIzYjE5NTIwY2JjOTYzYjMyZS4uMWZhZjM5ZmI5OTYyNGRhMTViMjgx NWM1OGE0ODY5MDJhOWFlOGQ0YiAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvV2ViUHJvY2Vz cy9XZWJDb3JlU3VwcG9ydC9XZWJHcmFwaGljc0xheWVyLmNwcAorKysgYi9Tb3VyY2UvV2ViS2l0 Mi9XZWJQcm9jZXNzL1dlYkNvcmVTdXBwb3J0L1dlYkdyYXBoaWNzTGF5ZXIuY3BwCkBAIC02MTks NiArNjE5LDExIEBAIHZvaWQgV2ViR3JhcGhpY3NMYXllcjo6c2V0V2ViR3JhcGhpY3NMYXllckNs aWVudChXZWJLaXQ6OldlYkdyYXBoaWNzTGF5ZXJDbGllbnQqCiAgICAgICAgIGNsaWVudC0+YXR0 YWNoTGF5ZXIodGhpcyk7CiB9CiAKK3ZvaWQgV2ViR3JhcGhpY3NMYXllcjo6ZGlkUmVuZGVyRnJh bWUoKQoreworICAgIG1fcHJldmlvdXNCYWNraW5nU3RvcmUuY2xlYXIoKTsKK30KKwogdm9pZCBX ZWJHcmFwaGljc0xheWVyOjphZGp1c3RWaXNpYmxlUmVjdCgpCiB7CiAgICAgaWYgKG1fbWFpbkJh Y2tpbmdTdG9yZSkKZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJLaXQyL1dlYlByb2Nlc3MvV2ViQ29y ZVN1cHBvcnQvV2ViR3JhcGhpY3NMYXllci5oIGIvU291cmNlL1dlYktpdDIvV2ViUHJvY2Vzcy9X ZWJDb3JlU3VwcG9ydC9XZWJHcmFwaGljc0xheWVyLmgKaW5kZXggYTZjYzQzOTA5MGI2NmYxZjg2 NmQ2ZjRkMjFjZWU0OGZhOWIxYmZlYi4uZjIyZTY5NDIyYWRkN2U3YmFjYjJkYzYwOWRmN2E0NDkz Zjk4YjQ1NSAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvV2ViUHJvY2Vzcy9XZWJDb3JlU3Vw cG9ydC9XZWJHcmFwaGljc0xheWVyLmgKKysrIGIvU291cmNlL1dlYktpdDIvV2ViUHJvY2Vzcy9X ZWJDb3JlU3VwcG9ydC9XZWJHcmFwaGljc0xheWVyLmgKQEAgLTEzOCw2ICsxMzgsNyBAQCBwdWJs aWM6CiAKICAgICB2b2lkIHNldFdlYkdyYXBoaWNzTGF5ZXJDbGllbnQoV2ViS2l0OjpXZWJHcmFw aGljc0xheWVyQ2xpZW50Kik7CiAKKyAgICB2b2lkIGRpZFJlbmRlckZyYW1lKCk7CiAgICAgdm9p ZCBhZGp1c3RWaXNpYmxlUmVjdCgpOwogICAgIGJvb2wgaXNSZWFkeUZvclRpbGVCdWZmZXJTd2Fw KCkgY29uc3Q7CiAgICAgdm9pZCB1cGRhdGVDb250ZW50QnVmZmVycygpOwpkaWZmIC0tZ2l0IGEv U291cmNlL1dlYktpdDIvV2ViUHJvY2Vzcy9XZWJQYWdlL3F0L0xheWVyVHJlZUhvc3RRdC5jcHAg Yi9Tb3VyY2UvV2ViS2l0Mi9XZWJQcm9jZXNzL1dlYlBhZ2UvcXQvTGF5ZXJUcmVlSG9zdFF0LmNw cAppbmRleCAwNDEzNzlmZDY4ZDk2ZDBjYThkOTI1MDgwYjk5YzJjMGI2N2UwMzczLi5lNGIyODk4 NWI5ZGIxOGUyYzQ5OTQ4Y2E1OGVhYjc1ZWRlOGVjMGM5IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2Vi S2l0Mi9XZWJQcm9jZXNzL1dlYlBhZ2UvcXQvTGF5ZXJUcmVlSG9zdFF0LmNwcAorKysgYi9Tb3Vy Y2UvV2ViS2l0Mi9XZWJQcm9jZXNzL1dlYlBhZ2UvcXQvTGF5ZXJUcmVlSG9zdFF0LmNwcApAQCAt NDQ0LDYgKzQ0NCwxMCBAQCB2b2lkIExheWVyVHJlZUhvc3RRdDo6cmVuZGVyTmV4dEZyYW1lKCkK IHsKICAgICBtX3dhaXRpbmdGb3JVSVByb2Nlc3MgPSBmYWxzZTsKICAgICBzY2hlZHVsZUxheWVy Rmx1c2goKTsKKworICAgIEhhc2hTZXQ8V2ViQ29yZTo6V2ViR3JhcGhpY3NMYXllcio+OjppdGVy YXRvciBlbmQgPSBtX3JlZ2lzdGVyZWRMYXllcnMuZW5kKCk7CisgICAgZm9yIChIYXNoU2V0PFdl YkNvcmU6OldlYkdyYXBoaWNzTGF5ZXIqPjo6aXRlcmF0b3IgaXQgPSBtX3JlZ2lzdGVyZWRMYXll cnMuYmVnaW4oKTsgaXQgIT0gZW5kOyArK2l0KQorICAgICAgICAoKml0KS0+ZGlkUmVuZGVyRnJh bWUoKTsKIH0KIAogYm9vbCBMYXllclRyZWVIb3N0UXQ6OmxheWVyVHJlZVRpbGVVcGRhdGVzQWxs b3dlZCgpIGNvbnN0Cg== 127999 2012-02-21 10:54:04 -0800 2012-02-21 15:30:36 -0800 Patch patch.diff text/plain 3585 noam ZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZyBiL1NvdXJjZS9XZWJDb3JlL0No YW5nZUxvZwppbmRleCA3ZWNkZDc3Li5kZjFlMjc0IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViQ29y ZS9DaGFuZ2VMb2cKKysrIGIvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsMTgg QEAKKzIwMTItMDItMjEgIE5vJ2FtIFJvc2VudGhhbCAgPG5vYW0ucm9zZW50aGFsQG5va2lhLmNv bT4KKworICAgICAgICBbUXRdIFByZXZpb3VzIHdlYiBwYWdlIGFwcGVhcnMgb3V0c2lkZSBjb250 ZW50IHJlY3QKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lk PTc4ODE2CisKKyAgICAgICAgQWRqdXN0IHRoZSBzY2lzc29yIGNsaXBwaW5nIGJ5IC0xLCBhcyB0 aGUgZ2xTY2lzc29yKCkgZnVuY3Rpb24gcmVxdWlyZXMgY29vcmRpbmF0ZXMgYW5kCisgICAgICAg IG5vdCBzaXplcy4KKworICAgICAgICBSZXZpZXdlZCBieSBLZW5uZXRoIFJvaGRlIENocmlzdGlh bnNlbi4KKworICAgICAgICBObyBuZXcgZnVuY3Rpb25hbGl0eS4KKworICAgICAgICAqIHBsYXRm b3JtL2dyYXBoaWNzL29wZW5nbC9UZXh0dXJlTWFwcGVyR0wuY3BwOgorICAgICAgICAoV2ViQ29y ZTo6c2Npc3NvckNsaXApOgorCiAyMDEyLTAyLTIwICBObydhbSBSb3NlbnRoYWwgIDxub2FtLnJv c2VudGhhbEBub2tpYS5jb20+CiAKICAgICAgICAgW1RleG1hcF0gTGF5ZXJzIGFuZCB0aWxlcyBh cHBlYXIgdG8gaGF2ZSBtaXNzaW5nIHBpeGVscyBpbiB0aGVpciByaWdodC9ib3R0b20gYm9yZGVy cwpkaWZmIC0tZ2l0IGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3Mvb3BlbmdsL1Rl eHR1cmVNYXBwZXJHTC5jcHAgYi9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9vcGVu Z2wvVGV4dHVyZU1hcHBlckdMLmNwcAppbmRleCBjMDM0NjE4Li5mOGVhNDgxIDEwMDY0NAotLS0g YS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9vcGVuZ2wvVGV4dHVyZU1hcHBlckdM LmNwcAorKysgYi9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9vcGVuZ2wvVGV4dHVy ZU1hcHBlckdMLmNwcApAQCAtODE5LDcgKzgxOSw3IEBAIHN0YXRpYyB2b2lkIHNjaXNzb3JDbGlw KGNvbnN0IEludFJlY3QmIHJlY3QpCiB7CiAgICAgR0xpbnQgdmlld3BvcnRbNF07CiAgICAgZ2xH ZXRJbnRlZ2VydihHTF9WSUVXUE9SVCwgdmlld3BvcnQpOwotICAgIGdsU2Npc3NvcihyZWN0Lngo KSwgdmlld3BvcnRbM10gLSByZWN0Lm1heFkoKSwgcmVjdC53aWR0aCgpLCByZWN0LmhlaWdodCgp KTsKKyAgICBnbFNjaXNzb3IocmVjdC54KCksIHZpZXdwb3J0WzNdIC0gcmVjdC5tYXhZKCkgKyAx LCByZWN0LndpZHRoKCkgLSAxLCByZWN0LmhlaWdodCgpIC0gMSk7CiB9CiAKIGJvb2wgVGV4dHVy ZU1hcHBlckdMOjpiZWdpblNjaXNzb3JDbGlwKGNvbnN0IFRyYW5zZm9ybWF0aW9uTWF0cml4JiBt b2RlbFZpZXdNYXRyaXgsIGNvbnN0IEZsb2F0UmVjdCYgdGFyZ2V0UmVjdCkKZGlmZiAtLWdpdCBh L1NvdXJjZS9XZWJLaXQyL0NoYW5nZUxvZyBiL1NvdXJjZS9XZWJLaXQyL0NoYW5nZUxvZwppbmRl eCBkZDY4YTE0Li45ZDA4YmE0IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cK KysrIGIvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsMTUgQEAKKzIwMTItMDIt MjEgIE5vJ2FtIFJvc2VudGhhbCAgPG5vYW0ucm9zZW50aGFsQG5va2lhLmNvbT4KKworICAgICAg ICBbUXRdIFByZXZpb3VzIHdlYiBwYWdlIGFwcGVhcnMgb3V0c2lkZSBjb250ZW50IHJlY3QKKyAg ICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTc4ODE2CisKKyAg ICAgICAgQXBwbHkgYSBjbGlwIG9uIHBhaW50ZWQgdGlsZXMgaWYgc29tZSBvZiB0aGUgdGlsZXMg ZmFsbCBvdXRzaWRlIG9mIHRoZSB0YXJnZXQgcmVjdC4KKworICAgICAgICBSZXZpZXdlZCBieSBL ZW5uZXRoIFJvaGRlIENocmlzdGlhbnNlbi4KKworICAgICAgICAqIFVJUHJvY2Vzcy9xdC9MYXll ckJhY2tpbmdTdG9yZS5jcHA6CisgICAgICAgIChXZWJLaXQ6OkxheWVyQmFja2luZ1N0b3JlOjpw YWludFRvVGV4dHVyZU1hcHBlcik6CisKIDIwMTItMDItMjAgIE5vJ2FtIFJvc2VudGhhbCAgPG5v YW0ucm9zZW50aGFsQG5va2lhLmNvbT4KIAogICAgICAgICBbUXRdW1dLMl0gQWNjZWxlcmF0ZWQg YW5pbWF0aW9ucyBkb24ndCB3b3JrIG9uIE1hYwpkaWZmIC0tZ2l0IGEvU291cmNlL1dlYktpdDIv VUlQcm9jZXNzL3F0L0xheWVyQmFja2luZ1N0b3JlLmNwcCBiL1NvdXJjZS9XZWJLaXQyL1VJUHJv Y2Vzcy9xdC9MYXllckJhY2tpbmdTdG9yZS5jcHAKaW5kZXggOGY1ODMwMy4uYzZkZjU3ZSAxMDA2 NDQKLS0tIGEvU291cmNlL1dlYktpdDIvVUlQcm9jZXNzL3F0L0xheWVyQmFja2luZ1N0b3JlLmNw cAorKysgYi9Tb3VyY2UvV2ViS2l0Mi9VSVByb2Nlc3MvcXQvTGF5ZXJCYWNraW5nU3RvcmUuY3Bw CkBAIC05NCw3ICs5NCw3IEBAIFBhc3NSZWZQdHI8Qml0bWFwVGV4dHVyZT4gTGF5ZXJCYWNraW5n U3RvcmU6OnRleHR1cmUoKSBjb25zdAogICAgIHJldHVybiBQYXNzUmVmUHRyPEJpdG1hcFRleHR1 cmU+KCk7CiB9CiAKLXZvaWQgTGF5ZXJCYWNraW5nU3RvcmU6OnBhaW50VG9UZXh0dXJlTWFwcGVy KFRleHR1cmVNYXBwZXIqIHRleHR1cmVNYXBwZXIsIGNvbnN0IEZsb2F0UmVjdCYsIGNvbnN0IFRy YW5zZm9ybWF0aW9uTWF0cml4JiB0cmFuc2Zvcm0sIGZsb2F0IG9wYWNpdHksIEJpdG1hcFRleHR1 cmUqIG1hc2spCit2b2lkIExheWVyQmFja2luZ1N0b3JlOjpwYWludFRvVGV4dHVyZU1hcHBlcihU ZXh0dXJlTWFwcGVyKiB0ZXh0dXJlTWFwcGVyLCBjb25zdCBGbG9hdFJlY3QmIHRhcmdldFJlY3Qs IGNvbnN0IFRyYW5zZm9ybWF0aW9uTWF0cml4JiB0cmFuc2Zvcm0sIGZsb2F0IG9wYWNpdHksIEJp dG1hcFRleHR1cmUqIG1hc2spCiB7CiAgICAgVmVjdG9yPFRleHR1cmVNYXBwZXJUaWxlKj4gdGls ZXNUb1BhaW50OwogCkBAIC0xMjEsOCArMTIxLDE2IEBAIHZvaWQgTGF5ZXJCYWNraW5nU3RvcmU6 OnBhaW50VG9UZXh0dXJlTWFwcGVyKFRleHR1cmVNYXBwZXIqIHRleHR1cmVNYXBwZXIsIGNvbnN0 CiAgICAgICAgIGNvdmVyZWRSZWN0LnVuaXRlKHRpbGUucmVjdCgpKTsKICAgICB9CiAKKyAgICBi b29sIHNob3VsZENsaXAgPSAhdGFyZ2V0UmVjdC5jb250YWlucyhjb3ZlcmVkUmVjdCk7CisKKyAg ICBpZiAoc2hvdWxkQ2xpcCkKKyAgICAgICAgdGV4dHVyZU1hcHBlci0+YmVnaW5DbGlwKHRyYW5z Zm9ybSwgdGFyZ2V0UmVjdCk7CisKICAgICBmb3IgKHNpemVfdCBpID0gMDsgaSA8IHRpbGVzVG9Q YWludC5zaXplKCk7ICsraSkKICAgICAgICAgdGlsZXNUb1BhaW50W2ldLT5wYWludCh0ZXh0dXJl TWFwcGVyLCB0cmFuc2Zvcm0sIG9wYWNpdHksIG1hc2spOworCisgICAgaWYgKHNob3VsZENsaXAp CisgICAgICAgIHRleHR1cmVNYXBwZXItPmVuZENsaXAoKTsKIH0KIAogdm9pZCBMYXllckJhY2tp bmdTdG9yZTo6c3dhcEJ1ZmZlcnMoVGV4dHVyZU1hcHBlciogdGV4dHVyZU1hcHBlcikK 131581 2012-03-13 03:20:57 -0700 2012-03-13 03:20:57 -0700 Test page exposing what happens when clipping is done with no respect for 3D transforms. index.html text/html 2478 larsgk PCFET0NUWVBFIGh0bWwgUFVCTElDICItLy9XM0MvL0RURCBYSFRNTCAxLjAgVHJhbnNpdGlvbmFs Ly9FTiIKImh0dHA6Ly93d3cudzMub3JnL1RSL3hodG1sMS9EVEQveGh0bWwxLXRyYW5zaXRpb25h bC5kdGQiPgo8aHRtbCB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMTk5OS94aHRtbCI+CiAgICA8 IS0tIERvbmUgYnkgTGFycyBLbnVkc2VuIGluIE1hcmNoLCAyMDEyIC0tPgogICAgPGhlYWQ+CiAg ICAgICAgPHRpdGxlPkNsaXBQcm9ibGVtPC90aXRsZT4KICAgICAgICA8bWV0YSBuYW1lPSJ2aWV3 cG9ydCIgY29udGVudCA9ICJ3aWR0aD05ODAsIHVzZXItc2NhbGFibGUgPSBubyIgLz4KICAgICAg ICA8bWV0YSBodHRwLWVxdWl2PSJDb250ZW50LVR5cGUiIGNvbnRlbnQ9InRleHQvaHRtbDsgY2hh cnNldD1VVEYtOCIgLz4KPHN0eWxlIHR5cGU9InRleHQvY3NzIj4KCmJvZHkgewogIGJhY2tncm91 bmQtY29sb3I6IHdoaXRlOwp9Cgojc3BhY2Vjb250YWluZXIgewogIG1hcmdpbjogMDsKICB3aWR0 aDogOTYwcHg7CiAgaGVpZ2h0OiA3ODBweDsKICBwb3NpdGlvbjogYWJzb2x1dGU7CiAgdG9wOiAx MHB4OwogIGxlZnQ6IDEwcHg7CiAgYmFja2dyb3VuZC1jb2xvcjogYmxhY2s7CiAgLXdlYmtpdC1w ZXJzcGVjdGl2ZTogMzAwMDsKfQoKI3NwYWNlb2Zmc2V0IHsKICAtd2Via2l0LXRyYW5zZm9ybS1z dHlsZTogcHJlc2VydmUtM2Q7CiAgLXdlYmtpdC10cmFuc2Zvcm06IHRyYW5zbGF0ZVgoNDgwcHgp IHRyYW5zbGF0ZVkoMzkwcHgpOwp9Cgojc3BhY2VpdGVtIHsKICAtd2Via2l0LXRyYW5zZm9ybS1z dHlsZTogcHJlc2VydmUtM2Q7CiAgLXdlYmtpdC1hbmltYXRpb24tbmFtZTogcHVsc2U7CiAgLXdl YmtpdC1hbmltYXRpb24tZHVyYXRpb246IDRzOwogIC13ZWJraXQtYW5pbWF0aW9uLWl0ZXJhdGlv bi1jb3VudDogaW5maW5pdGU7CiAgLXdlYmtpdC1hbmltYXRpb24tdGltaW5nLWZ1bmN0aW9uOiBs aW5lYXI7CiAgLy8td2Via2l0LXRyYW5zZm9ybTogdHJhbnNsYXRlWCgtNDAwcHgpIHRyYW5zbGF0 ZVkoLTIwMHB4KSB0cmFuc2xhdGVaKDQwMHB4KTsKICBwb3NpdGlvbjogZml4ZWQ7CiAgd2lkdGg6 IDgwMHB4OwogIGhlaWdodDogNDAwcHg7CiAgYmFja2dyb3VuZC1jb2xvcjogd2hpdGU7Cn0KCQoj c3BhY2VyZWZlcmVuY2V7CiAgLXdlYmtpdC10cmFuc2Zvcm0tc3R5bGU6IHByZXNlcnZlLTNkOwog IC13ZWJraXQtdHJhbnNmb3JtOiB0cmFuc2xhdGVYKC00MDBweCkgdHJhbnNsYXRlWSgtMzAwcHgp OwogIHBvc2l0aW9uOiBmaXhlZDsKICB3aWR0aDogODAwcHg7CiAgaGVpZ2h0OiAyMHB4OwogIGJh Y2tncm91bmQtY29sb3I6IHdoaXRlOwp9CgkKQC13ZWJraXQta2V5ZnJhbWVzIHB1bHNlIHsKICAw JSAgICB7IC13ZWJraXQtdHJhbnNmb3JtOiB0cmFuc2xhdGVYKC00MDBweCkgdHJhbnNsYXRlWSgt MjAwcHgpIHRyYW5zbGF0ZVooLTQwMHB4KX0KICA1MCUgICB7IC13ZWJraXQtdHJhbnNmb3JtOiB0 cmFuc2xhdGVYKC00MDBweCkgdHJhbnNsYXRlWSgtMjAwcHgpIHRyYW5zbGF0ZVooNDAwcHgpfQog IDEwMCUgIHsgLXdlYmtpdC10cmFuc2Zvcm06IHRyYW5zbGF0ZVgoLTQwMHB4KSB0cmFuc2xhdGVZ KC0yMDBweCkgdHJhbnNsYXRlWigtNDAwcHgpfQp9Cgo8L3N0eWxlPgogICAgPC9oZWFkPgogICAg PGJvZHk+CiAgICAgIDxkaXYgaWQ9InNwYWNlY29udGFpbmVyIj4KICAgICAgICA8ZGl2IGlkPSJz cGFjZW9mZnNldCI+CiAgICAgICAgICA8ZGl2IGlkPSJzcGFjZWl0ZW0iPgogICAgICAgICAgICA8 ZGl2IHN0eWxlPSJiYWNrZ3JvdW5kLWNvbG9yOnJlZDsgd2lkdGg6MTAwJTsgaGVpZ2h0OjIwcHg7 Ij5UaGlzIGlzIGxpbmUgMSAtIDIwcHggdGFsbDwvZGl2PgogICAgICAgICAgICA8ZGl2IHN0eWxl PSJiYWNrZ3JvdW5kLWNvbG9yOmJsdWU7IHdpZHRoOjEwMCU7IGhlaWdodDoyMHB4OyI+VGhpcyBp cyBsaW5lIDIgLSAyMHB4IHRhbGw8L2Rpdj4KICAgICAgICAgICAgPGRpdiBzdHlsZT0iYmFja2dy b3VuZC1jb2xvcjpyZWQ7IHdpZHRoOjEwMCU7IGhlaWdodDoyMHB4OyI+VGhpcyBpcyBsaW5lIDMg LSAyMHB4IHRhbGw8L2Rpdj4KICAgICAgICAgICAgPGRpdiBzdHlsZT0iYmFja2dyb3VuZC1jb2xv cjpibHVlOyB3aWR0aDoxMDAlOyBoZWlnaHQ6MjBweDsiPlRoaXMgaXMgbGluZSA0IC0gMjBweCB0 YWxsPC9kaXY+CiAgICAgICAgICAgIDxkaXYgc3R5bGU9IndpZHRoOjEwMCU7IGhlaWdodDo1MHB4 OyI+V2hlbiB6ID4gMCwgdGhlIGNsaXBwaW5nIHNlZW1zIHRvIGN1dCB0b28gbXVjaC48L2Rpdj4K ICAgICAgICAgIDwvZGl2PgogICAgICAgICAgPGRpdiBpZD0ic3BhY2VyZWZlcmVuY2UiPgogICAg ICAgICAgICAgIFRoaXMgYmFyIGhhcyB0aGUgc2FtZSB3aWR0aCBhcyB0aGUgYm94IGJlbG93IC0g YnV0IHN0YXlzIGF0IHogPSAwCiAgICAgICAgICA8L2Rpdj4KICAgICAgICA8L2Rpdj4KICAgICAg PC9kaXY+CiAgICA8L2JvZHk+CjwvaHRtbD4K