230594 2021-09-21 21:14:28 -0700 Invalid browser selection after splitting a text node 2026-01-05 10:51:43 -0800 1 1 1 Unclassified WebKit DOM Safari 14 Mac (Intel) macOS 10.15 NEW InRadar P2 Normal --- 1 james webkit-unassigned cdumez darin rniwa thespyder webkit-bug-importer wenson_hsieh oldest_to_newest 1795940 0 438917 james 2021-09-21 21:14:28 -0700 Created attachment 438917 Replication HTML file Hi team, As per the spec, when splitting a text node, any live ranges that start or end in the node being split must have their respective nodes and offsets modified. https://dom.spec.whatwg.org/#concept-text-split (Important: Section 7) In Safari 14 (and also in the newly released Safari 15 & Tech Preview), splitting a text node does not correctly adjust the selection to account for this: - The offsets are not adjusted - The nodes are not changed when appropriate Please see the attached HTML file for a replication case. This works correctly in Firefox 92 and Chrome 93. 1797924 1 gsnedders 2021-09-28 09:43:05 -0700 I kinda suspect this might be fixed by bug 220514 (live ranges in selection)? Though at least currently it appears regressed by live ranges in selection (even the initial selection is wrong). Marking this is blocking, then others can debate whether that's fair. ;P 1798175 2 webkit-bug-importer 2021-09-28 21:15:14 -0700 <rdar://problem/83655114> 1920208 3 rniwa 2022-12-17 02:11:30 -0800 Enabling live range selection doesn't fix this bug. 438917 2021-09-21 21:14:28 -0700 2021-09-21 21:14:28 -0700 Replication HTML file selection-after-text-node-split.html text/html 785 james PHA+MS4gT3BlbiB0aGUgZGV2ZWxvcGVyIGNvbnNvbGUuPC9wPgoKPHA+Mi4gU2VsZWN0ICJ3b3Jk MiIgYmVsb3cgYW5kIGNsaWNrICJMb2cgU2VsZWN0aW9uIi48L3A+Cgo8cCBpZD0icDEiPndvcmQx IHdvcmQyIHdvcmQzPC9wPgoKPHA+My4gQ2xpY2sgIlNwbGl0IHRleHQgbm9kZSIgYW5kIHRoZW4g Y2xpY2sgIkxvZyBTZWxlY3Rpb24iIGFnYWluLjwvcD4KCjxwPjQuIE5vdGljZSB0aGF0IHRoZSBz ZWxlY3Rpb24gb2Zmc2V0cyBhbmQgbm9kZXMgYXJlIG5vdyBpbnZhbGlkIGR1ZSB0byB0aGUgdGV4 dCBub2RlIGJlaW5nIHNwbGl0LjwvcD4KCjxwPkFjdHVhbCBhbmNob3Igb2Zmc2V0OiA2PC9wPgo8 cD5FeHBlY3RlZCBhbmNob3Igb2Zmc2V0OiAxPC9wPgoKPHA+QWN0dWFsIGZvY3VzIG9mZnNldDog MTE8L3A+CjxwPkV4cGVjdGVkIGZvY3VzIG9mZnNldDogNjwvcD4KCjxidXR0b24gaWQ9ImxvZ1Nl bCI+TG9nIFNlbGVjdGlvbjwvYnV0dG9uPgoKPGJ1dHRvbiBpZD0ic3BsaXROb2RlIj5TcGxpdCB0 ZXh0IG5vZGU8L2J1dHRvbj4KCjxzY3JpcHQ+CiAgbG9nU2VsLmFkZEV2ZW50TGlzdGVuZXIoJ2Ns aWNrJywgKCkgPT4gewogICAgY29uc29sZS5sb2coZG9jdW1lbnQuZ2V0U2VsZWN0aW9uKCkpOwog IH0pOwoKICBzcGxpdE5vZGUuYWRkRXZlbnRMaXN0ZW5lcignY2xpY2snLCAoKSA9PiB7CiAgICBj b25zdCB0ZXh0Tm9kZSA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCdwMScpLmZpcnN0Q2hpbGQ7 CiAgICB0ZXh0Tm9kZS5zcGxpdFRleHQoNSk7CiAgfSk7Cjwvc2NyaXB0Pgo=