189161 2018-08-30 11:04:24 -0700 Searching for lowercase delta finds capital delta 2018-09-07 23:34:09 -0700 1 1 1 Unclassified WebKit Text Other Unspecified Unspecified NEW InRadar P2 Normal --- 1 mmaxfield webkit-unassigned ap mmaxfield webkit-bug-importer oldest_to_newest 1455290 0 mmaxfield 2018-08-30 11:04:24 -0700 1. Open https://www.khronos.org/registry/vulkan/specs/1.1/html/vkspec.html#_texel_input_operations 2. ⌘F 3. Search for U+03B4 δ Expected: U+0394 Δ is not found Actual: U+0394 Δ is found 1455585 1 webkit-bug-importer 2018-08-30 23:28:26 -0700 <rdar://problem/43925547> 1455718 2 ap 2018-08-31 09:42:39 -0700 > Expected: U+0394 Δ is not found That's not intuitively correct. Why is that the expectation? 1456534 3 mmaxfield 2018-09-04 17:02:19 -0700 (In reply to Alexey Proskuryakov from comment #2) > > Expected: U+0394 Δ is not found > > That's not intuitively correct. Why is that the expectation? These characters are used as variables in math formulas. They represent distinct variables. 1456542 4 ap 2018-09-04 17:09:53 -0700 How will we distinguish that case from when they are used in genuine Greek text? 1456558 5 mmaxfield 2018-09-04 17:41:37 -0700 (In reply to Alexey Proskuryakov from comment #4) > How will we distinguish that case from when they are used in genuine Greek > text? No idea. It's still a bug, though. 1456728 6 ap 2018-09-05 09:20:10 -0700 I don't see anything actionable here. All letters have different meanings in math depending on case, so it doesn't makes sense to nitpick on delta. Consider "o(N)" and "O(N)" - would that be a reason to claim that case insensitive search for "o" shouldn't match "O"? Delta is no different.