From 4d2807e128b6a934828f0dc44d309e2234f50bce Mon Sep 17 00:00:00 2001 From: ZyX Date: Sun, 26 Jul 2015 16:37:01 +0300 Subject: [PATCH] documentation: Document table meanings in comment to genunicodetables --- scripts/genunicodetables.lua | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/scripts/genunicodetables.lua b/scripts/genunicodetables.lua index e8de5026de..36339e2fc6 100644 --- a/scripts/genunicodetables.lua +++ b/scripts/genunicodetables.lua @@ -1,3 +1,17 @@ +-- Script creates the following tables in unicode_tables.generated.h: +-- +-- 1. doublewidth and ambiguous tables: sorted list of non-overlapping closed +-- intervals. Codepoints in these intervals have double (W or F) or ambiguous +-- (A) east asian width respectively. +-- 2. combining table: same as the above, but characters inside are combining +-- characters (i.e. have general categories equal to Mn, Mc or Me). +-- 3. foldCase, toLower and toUpper tables used to convert characters to +-- folded/lower/upper variants. In these tables first two values are +-- character ranges: like in previous tables they are sorted and must be +-- non-overlapping. Third value means step inside the range: e.g. if it is +-- 2 then interval applies only to first, third, fifth, … character in range. +-- Fourth value is number that should be added to the codepoint to yield +-- folded/lower/upper codepoint. if arg[1] == '--help' then print('Usage:') print(' genunicodetables.lua UnicodeData.txt CaseFolding.txt ' ..