88772 2012-06-11 03:03:11 -0700 Viewbox Scaling Exposes Incorrect Transformations on Text Gradient Fills 2023-01-21 06:44:24 -0800 1 1 1 Unclassified WebKit SVG 528+ (Nightly build) Unspecified Unspecified NEW P2 Normal --- 1 d-r webkit-unassigned ahmad.saleem792 eae krit zimmermann oldest_to_newest 645873 0 d-r 2012-06-11 03:03:11 -0700 When using a viewbox attribute that results in upscaling, text gradient fills are rendered incorrectly. 645891 1 146825 d-r 2012-06-11 04:11:02 -0700 Created attachment 146825 RefTests 645892 2 146825 d-r 2012-06-11 04:14:12 -0700 Comment on attachment 146825 RefTests This reftest shows the issue. The gradient should scale from red over green to blue over the size of the text, same in both cases - with viewbox and without. 651386 3 d-r 2012-06-18 07:51:39 -0700 The issue seems to be with identifying the bounding box of small-scale text. Origin of bounding box for the text in LayoutTests/svg/transforms/viewbox-text-fill-gradient.html is 0,0 - extents are 1,1 - while these should be precise float values. 651420 4 d-r 2012-06-18 09:07:07 -0700 In my understanding, the rounding happens in void SVGRootInlineBox::layoutRootBox(). Using LayoutUnits here, would only improve the results but not fix the issue completely - we need higher precision for the text bounding boxes. 651424 5 d-r 2012-06-18 09:11:03 -0700 Probably same root cause as bug 75781. 816515 6 d-r 2013-01-25 05:31:37 -0800 This issue persists even after landing patch from bug 107771. 1927100 7 ahmad.saleem792 2023-01-21 06:44:24 -0800 I took the test case from the attached patch and turned into JSFiddle and also added missing </rect>: JSFiddle - https://jsfiddle.net/9z4Ljr7y/show Safari 16.2, Safari Technology Preview 161, Chrome Canary 111 and Firefox Nightly 111 are matching other, do we need to track anything else? Thanks! 146825 2012-06-11 04:11:02 -0700 2012-06-11 04:14:12 -0700 RefTests bug-88772-20120611141248.patch text/plain 2470 d-r U3VidmVyc2lvbiBSZXZpc2lvbjogMTE5OTY3CmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0cy9DaGFu Z2VMb2cgYi9MYXlvdXRUZXN0cy9DaGFuZ2VMb2cKaW5kZXggZTM5MTYwNjkwODBiY2E3NTZjY2Ix MTgwYjQxODMwMDNmMjc5NTA1Yy4uNzYxYWFhNzM0ZTU2YTYzMDc2NjY3NmE0ZTM4ZDc5Njc5MzFh MTdkZiAxMDA2NDQKLS0tIGEvTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCisrKyBiL0xheW91dFRlc3Rz L0NoYW5nZUxvZwpAQCAtMSwzICsxLDEzIEBACisyMDEyLTA2LTExICBEb21pbmlrIFLDtnR0c2No ZXMgIDxkb21pbmlrLnJvdHRzY2hlc0BpbnRlbC5jb20+CisKKyAgICAgICAgVmlld2JveCBTY2Fs aW5nIEV4cG9zZXMgSW5jb3JyZWN0IFRyYW5zZm9ybWF0aW9ucyBvbiBUZXh0IEdyYWRpZW50IEZp bGxzCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD04ODc3 MgorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogc3Zn L3RyYW5zZm9ybXMvdmlld2JveC10ZXh0LWZpbGwtZ3JhZGllbnQtZXhwZWN0ZWQuaHRtbDogQWRk ZWQuCisgICAgICAgICogc3ZnL3RyYW5zZm9ybXMvdmlld2JveC10ZXh0LWZpbGwtZ3JhZGllbnQu aHRtbDogQWRkZWQuCisKIDIwMTItMDYtMTEgIEFsbGFuIFNhbmRmZWxkIEplbnNlbiAgPGFsbGFu LmplbnNlbkBub2tpYS5jb20+CiAKICAgICAgICAgUmVjdC1iYXNlZCBoaXR0ZXN0aW5nIGRvZXNu J3Qgd29yayBpbiB0YWJsZXMuCmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0cy9zdmcvdHJhbnNmb3Jt cy92aWV3Ym94LXRleHQtZmlsbC1ncmFkaWVudC1leHBlY3RlZC5odG1sIGIvTGF5b3V0VGVzdHMv c3ZnL3RyYW5zZm9ybXMvdmlld2JveC10ZXh0LWZpbGwtZ3JhZGllbnQtZXhwZWN0ZWQuaHRtbApu ZXcgZmlsZSBtb2RlIDEwMDY0NAppbmRleCAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAw MDAwMDAwMDAwLi4xODI0MTBkNzQ4NjE0MjA5ZGU5ZjkyYWJiZWZkZTdmNjgzN2Q0ODE4Ci0tLSAv ZGV2L251bGwKKysrIGIvTGF5b3V0VGVzdHMvc3ZnL3RyYW5zZm9ybXMvdmlld2JveC10ZXh0LWZp bGwtZ3JhZGllbnQtZXhwZWN0ZWQuaHRtbApAQCAtMCwwICsxLDEzIEBACis8aHRtbD48Ym9keT48 c3ZnIHdpZHRoPSI2MDAiIGhlaWdodD0iNjAwIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAw MC9zdmciPgorIDxkZWZzPgorICAgPGxpbmVhcmdyYWRpZW50IHgxPSIwIiB4Mj0iMSIgaWQ9Imdy YWRpZW50Ij4KKyAgICAgPHN0b3Agc3RvcC1jb2xvcj0iI2YwMCIgb2Zmc2V0PSIwIj48L3N0b3A+ CisgICAgIDxzdG9wIHN0b3AtY29sb3I9IiMwZjAiIG9mZnNldD0iMC41Ij48L3N0b3A+CisgICAg IDxzdG9wIHN0b3AtY29sb3I9IiMwMGYiIG9mZnNldD0iMSI+PC9zdG9wPgorICAgPC9saW5lYXJn cmFkaWVudD4KKyA8L2RlZnM+CisgPHRleHQgeT0iMzAiIGZvbnQtc2l6ZT0iMzAiIGZpbGw9InVy bCgjZ3JhZGllbnQpIj5UZXh0IEdyYWRpZW50IFNjYWxpbmcuPC90ZXh0PgorIDxyZWN0IHk9IjM2 IiBoZWlnaHQ9IjMwIiB3aWR0aD0iMjY0IiBmaWxsPSJ1cmwoI2dyYWRpZW50KSI+Cis8L3N2Zz4K KzwvYm9keT4KKzwvaHRtbD4KZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL3N2Zy90cmFuc2Zvcm1z L3ZpZXdib3gtdGV4dC1maWxsLWdyYWRpZW50Lmh0bWwgYi9MYXlvdXRUZXN0cy9zdmcvdHJhbnNm b3Jtcy92aWV3Ym94LXRleHQtZmlsbC1ncmFkaWVudC5odG1sCm5ldyBmaWxlIG1vZGUgMTAwNjQ0 CmluZGV4IDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAuLjZlNDU5Yzk1 Y2NhMmZmNmJkZWUxYzIxZTFkNDY1ODcwMzgwN2I0MWQKLS0tIC9kZXYvbnVsbAorKysgYi9MYXlv dXRUZXN0cy9zdmcvdHJhbnNmb3Jtcy92aWV3Ym94LXRleHQtZmlsbC1ncmFkaWVudC5odG1sCkBA IC0wLDAgKzEsMTQgQEAKKzxodG1sPjxib2R5Pjxzdmcgdmlld0JveD0iMCAwIDEgMSIgd2lkdGg9 IjYwMCIgaGVpZ2h0PSI2MDAiIHhtbG5zPSJodHRwOi8vd3d3LnczLm9yZy8yMDAwL3N2ZyI+Cisg PGRlZnM+CisgICA8bGluZWFyZ3JhZGllbnQgeDE9IjAiIHgyPSIxIiBpZD0iZ3JhZGllbnQiPgor ICAgICA8c3RvcCBzdG9wLWNvbG9yPSIjZjAwIiBvZmZzZXQ9IjAiPjwvc3RvcD4KKyAgICAgPHN0 b3Agc3RvcC1jb2xvcj0iIzBmMCIgb2Zmc2V0PSIwLjUiPjwvc3RvcD4KKyAgICAgPHN0b3Agc3Rv cC1jb2xvcj0iIzAwZiIgb2Zmc2V0PSIxIj48L3N0b3A+CisgICA8L2xpbmVhcmdyYWRpZW50Pgor IDwvZGVmcz4KKyA8dGV4dCB5PSIwLjA1IiBmb250LXNpemU9IjAuMDUiIGZpbGw9InVybCgjZ3Jh ZGllbnQpIj5UZXh0IEdyYWRpZW50IFNjYWxpbmcuPC90ZXh0PgorIDxyZWN0IHk9IjAuMDYiIGhl aWdodD0iMC4wNSIgd2lkdGg9IjAuNDQiIGZpbGw9InVybCgjZ3JhZGllbnQpIj4KKworPC9zdmc+ Cis8L2JvZHk+Cis8L2h0bWw+Cg==