From db70f687d5f928987c3912a16beb210e9dc52b62 Mon Sep 17 00:00:00 2001 From: Henri Sivonen Date: Wed, 11 Dec 2024 10:39:36 +0200 Subject: [PATCH] Update components/normalizer/trie-value-format.md Co-authored-by: Elango Cheran --- components/normalizer/trie-value-format.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/components/normalizer/trie-value-format.md b/components/normalizer/trie-value-format.md index e94f7e216ee..d8fa3064ce7 100644 --- a/components/normalizer/trie-value-format.md +++ b/components/normalizer/trie-value-format.md @@ -28,7 +28,7 @@ The decomposition is the NFKD decomposition of U+FDFA: The highest bit is 0, the Hangul syllable: The trie value is 1. (I.e. only the lowest bit is set to 1.) -The decomposition is a singleton decomposition to a single BMP character > 0x1F: The highest bit is 1 iff the decomposition can combining backwards (does not occur as of Unicode 16.0), the second-highest bit is 1, the low half is the decomposition. +The decomposition is a singleton decomposition to a single BMP character > 0x1F: The highest bit is 1 iff the decomposition can combine backwards (does not occur as of Unicode 16.0), the second-highest bit is 1, the low half is the decomposition. The character is not the ANGSTROM SIGN, and the decomposition is to a starter <= U+7FFF and > 0x1F that cannot combine backwards followed by a non-starter <= U+7FFF and > 0x1F: The highest bit is 0. The second-highest bit is set according to its general semantics. The lowest 15 bits are the leading starter. The next 15 bits are the trailing non-starter.