Imperivm Romanvm

Module:Unicode data: Difference between revisions

Module:Unicode data (edit)

Revision as of 15:19, 17 March 2022

4,015 bytes removed ,  2 years ago
move some data to a data submodule
(Update for Unicode 14.0)
 
(move some data to a data submodule)
Line 1: Line 1:
local p = {}
local export = {}
local udata = mw.loadData("Module:Unicode data/data")


local floor = math.floor
local floor = math.floor


local function errorf(level, ...)
local function errorf(first_arg, ...)
if type(level) == "number" then
if type(first_arg) == "number" then
return error(string.format(...), level + 1)
return error(string.format(...), first_arg + 1)
else -- level is actually the format string.
else
return error(string.format(level, ...), 2)
return error(string.format(first_arg, ...), 2)
end
end
end
end
Line 13: Line 14:
local function binary_range_search(codepoint, ranges)
local function binary_range_search(codepoint, ranges)
local low, mid, high
local low, mid, high
low, high = 1, ranges.length or require "Module:TableTools".length(ranges)
low, high = 1, ranges.length or require "Module:table".length(ranges)
while low <= high do
while low <= high do
mid = floor((low + high) / 2)
mid = floor((low + high) / 2)
Line 27: Line 28:
return nil, mid
return nil, mid
end
end
p.binary_range_search = binary_range_search
export.binary_range_search = binary_range_search


--[[
local function linear_range_search(codepoint, ranges)
local function linear_range_search(codepoint, ranges)
for i, range in ipairs(ranges) do
for i, range in ipairs(ranges) do
if range[1] <= codepoint and codepoint <= range[2] then
if codepoint < range[1] then
break
elseif codepoint <= range[2] then
return range
return range
end
end
end
end
end
end
--]]


-- Load a module by indexing "loader" with the name of the module minus the
-- Load a module by indexing "loader" with the name of the module minus the
Line 57: Line 58:
-- see "Hangul Syllable Name Generation" in section 3.12 of the
-- see "Hangul Syllable Name Generation" in section 3.12 of the
-- Unicode Specification:
-- Unicode Specification:
-- https://www.unicode.org/versions/Unicode11.0.0/ch03.pdf
-- https://www.unicode.org/versions/latest/ch03.pdf
-- For most of the name rules given here, see the subsection
-- "Unicode Name Property" in section 4.8 (Name) and the table 4-8
-- (Name Derivation Rule Prefix Strings):
-- https://www.unicode.org/versions/latest/ch04.pdf
local name_hooks = {
local name_hooks = {
{    0x00,    0x1F, "<control-%04X>" }, -- C0 control characters
{    0x00,    0x1F, "<control-%04X>" }, -- C0 control characters
Line 80: Line 85:
{  0xF900,  0xFA6D, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0xF900,  0xFA6D, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0xFA70,  0xFAD9, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0xFA70,  0xFAD9, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0x17000,  0x187F7, "TANGUT IDEOGRAPH-%04X" }, -- Tangut Ideograph
{  0x17000,  0x187F7, "TANGUT IDEOGRAPH-%04X" }, -- Tangut
{  0x18800,  0x18AFF, function (codepoint)
{  0x18800,  0x18AFF, function (codepoint)
return ("TANGUT COMPONENT-%03d"):format(codepoint - 0x187FF)
return ("TANGUT COMPONENT-%03d"):format(codepoint - 0x187FF)
end },
end },
{  0x18D00,  0x18D08, "TANGUT IDEOGRAPH-%04X" }, -- Tangut Ideograph Supplement
{  0x18D00,  0x18D08, "TANGUT IDEOGRAPH-%04X" }, -- Tangut
{  0x18B00,  0x18CD5, "KHITAN SMALL SCRIPT CHARACTER-%04X" },
{  0x1B170,  0x1B2FB, "NUSHU CHARACTER-%04X" }, -- Nushu
{  0x1B170,  0x1B2FB, "NUSHU CHARACTER-%04X" }, -- Nushu
{  0x20000,  0x2A6DF, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension B
{  0x20000,  0x2A6DF, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension B
{  0x2A700,  0x2B738, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension C
{  0x2A700,  0x2B738, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension C
{  0x2B740,  0x2B81D, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension D
{  0x2A740,  0x2B81D, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension D
{  0x2B820,  0x2CEA1, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension E
{  0x2B820,  0x2CEA1, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension E
{  0x2CEB0,  0x2EBE0, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension F
{  0x2CEB0,  0x2EBE0, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension F
-- CJK Compatibility Ideographs Supplement (Supplementary Ideographic Plane)
-- CJK Compatibility Ideographs Supplement (Supplementary Ideographic Plane)
{  0x2F800,  0x2FA1D, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0x2F800,  0x2FA1D, "CJK COMPATIBILITY IDEOGRAPH-%04X" },
{  0x30000,  0x3134A, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension G
{  0xE0100,  0xE01EF, function (codepoint) -- Variation Selectors Supplement
{  0xE0100,  0xE01EF, function (codepoint) -- Variation Selectors Supplement
return ("VARIATION SELECTOR-%d"):format(codepoint - 0xE0100 + 17)
return ("VARIATION SELECTOR-%d"):format(codepoint - 0xE0100 + 17)
end},
end},
{  0x30000,  0x3134A, "CJK UNIFIED IDEOGRAPH-%04X" }, -- CJK Ideograph Extension G
{  0xF0000,  0xFFFFD, "<private-use-%04X>" }, -- Plane 15 Private Use
{  0xF0000,  0xFFFFD, "<private-use-%04X>" }, -- Plane 15 Private Use
{ 0x100000, 0x10FFFD, "<private-use-%04X>" }  -- Plane 16 Private Use
{ 0x100000, 0x10FFFD, "<private-use-%04X>" }  -- Plane 16 Private Use
Line 124: Line 130:
--]]
--]]


-- https://www.unicode.org/versions/Unicode11.0.0/ch04.pdf, section 4.8
-- https://www.unicode.org/versions/latest/ch04.pdf, section 4.8
function p.lookup_name(codepoint)
function export.lookup_name(codepoint)
-- U+FDD0-U+FDEF and all code points ending in FFFE or FFFF are Unassigned
-- U+FDD0-U+FDEF and all code points ending in FFFE or FFFF are Unassigned
-- (Cn) and specifically noncharacters:
-- (Cn) and specifically noncharacters:
Line 160: Line 166:
end
end


--[[
function export.lookup_image(codepoint)
-- No image data modules on Wikipedia yet.
function p.lookup_image(codepoint)
local data = loader[('images/%03X'):format(codepoint / 0x1000)]
local data = loader[('images/%03X'):format(codepoint / 0x1000)]
Line 169: Line 173:
end
end
end
end
--]]
local planes = {
[ 0] = "Basic Multilingual Plane";
[ 1] = "Supplementary Multilingual Plane";
[ 2] = "Supplementary Ideographic Plane";
[ 3] = "Tertiary Ideographic Plane";
[14] = "Supplementary Special-purpose Plane";
[15] = "Supplementary Private Use Area-A";
[16] = "Supplementary Private Use Area-B";
}


-- Load [[Module:Unicode data/blocks]] if needed and assign it to this variable.
-- Load [[Module:Unicode data/blocks]] if needed and assign it to this variable.
Line 189: Line 182:
if data then
if data then
-- Unpack doesn't work on tables loaded with mw.loadData.
-- Unpack doesn't work on tables loaded with mw.loadData.
return i, data[1], data[2], data[3]
return i, data[3], data[1], data[2]
end
end
end
end


-- An ipairs-type iterator generator for the list of blocks.
-- An ipairs-type iterator generator for the list of blocks.
function p.enum_blocks()
function export.enum_blocks()
local blocks = loader.blocks
local blocks = loader.blocks
return block_iter, blocks, 0
return block_iter, blocks, 0
end
end


function p.lookup_plane(codepoint)
function export.get_block_range(name)
local range
for i, block in ipairs(loader.blocks) do
if block[3] == name then
range = block
end
end
if range then
return range[1], range[2]
end
end
 
function export.lookup_plane(codepoint)
local i = floor(codepoint / 0x10000)
local i = floor(codepoint / 0x10000)
return planes[i] or ("Plane %u"):format(i)
return udata.planes[i] or ("Plane %u"):format(i)
end
end


function p.lookup_block(codepoint)
function export.lookup_block(codepoint)
local blocks = loader.blocks
local blocks = loader.blocks
local range = binary_range_search(codepoint, blocks)
local range = binary_range_search(codepoint, blocks)
Line 214: Line 221:
end
end


function p.get_block_info(name)
function export.get_block_info(name)
for i, block in ipairs(loader.blocks) do
for i, block in ipairs(loader.blocks) do
if block[3] == name then
if block[3] == name then
Line 222: Line 229:
end
end


function p.is_valid_pagename(pagename)
function export.is_valid_pagename(pagename)
local has_nonws = false
local has_nonws = false


Line 239: Line 246:
end
end


local printable, result = p.is_printable(cp)
local printable, result = export.is_printable(cp)
if not printable then
if not printable then
return false
return false
Line 328: Line 335:
end
end
return match_func(codepoint)
return match_func(codepoint, unpack(dots))
end
end
end
end
Line 338: Line 345:
-- See https://www.unicode.org/reports/tr44/#Canonical_Combining_Class_Values for
-- See https://www.unicode.org/reports/tr44/#Canonical_Combining_Class_Values for
-- more information.
-- more information.
p.is_combining = memo_lookup(
export.is_combining = memo_lookup(
"combining",
"combining",
function (codepoint, combining_class)
function (codepoint, combining_class)
Line 345: Line 352:
0)
0)


function p.add_dotted_circle(str)
function export.add_dotted_circle(str)
return (mw.ustring.gsub(str, ".",
return (mw.ustring.gsub(str, ".",
function(char)
function(char)
if p.is_combining(mw.ustring.codepoint(char)) then
if export.is_combining(mw.ustring.codepoint(char)) then
return '◌' .. char
return '◌' .. char
end
end
Line 360: Line 367:
end,
end,
"assigned")
"assigned")
p.lookup_control = lookup_control
export.lookup_control = lookup_control


function p.is_assigned(codepoint)
function export.is_assigned(codepoint)
return lookup_control(codepoint) ~= "unassigned"
return lookup_control(codepoint) ~= "unassigned"
end
end


function p.is_printable(codepoint)
function export.is_printable(codepoint)
local result = lookup_control(codepoint)
local result = lookup_control(codepoint)
return (result == "assigned") or (result == "space-separator"), result
return (result == "assigned") or (result == "space-separator"), result
end
end


function p.is_whitespace(codepoint)
function export.is_whitespace(codepoint)
local result = lookup_control(codepoint)
local result = lookup_control(codepoint)
return (result == "space-separator"), result
return (result == "space-separator"), result
end
end


p.lookup_category = memo_lookup(
export.lookup_category = memo_lookup(
"category",
"category",
function (codepoint, category)
function (codepoint, category)
Line 383: Line 390:
"Cn")
"Cn")


local lookup_script = memo_lookup(
export.lookup_script = memo_lookup(
"scripts",
"scripts",
function (codepoint, script_code)
function (codepoint, script)
return script_code or 'Zzzz'
return script
end,
end,
"Zzzz")
"Zzzz")
p.lookup_script = lookup_script
function p.get_best_script(str)
-- Check type of argument, because mw.text.decode coerces numbers to strings!
require "libraryUtil".checkType("get_best_script", 1, str, "string")
-- Convert HTML character references (including named character references,
-- or character entities) to characters.
str = mw.text.decode(str, true)
local scripts = {}
for codepoint in mw.ustring.gcodepoint(str) do
local script = lookup_script(codepoint)
-- Ignore "Inherited", "Undetermined", or "Uncoded" scripts.
if not (script == "Zyyy" or script == "Zinh" or script == "Zzzz") then
scripts[script] = true
end
end
-- If scripts does not contain two or more keys,
-- return first and only key (script code) in table.
if not next(scripts, next(scripts)) then
return next(scripts)
end -- else return majority script, or else "Zzzz"?
end
function p.is_Latin(str)
require "libraryUtil".checkType("get_best_script", 1, str, "string")
str = mw.text.decode(str, true)
-- Search for the leading bytes that introduce the UTF-8 encoding of the
-- code points U+0340-U+10FFFF. If they are not found and there is at least
-- one Latin-script character, the string counts as Latin, because the rest
-- of the characters can only be Zyyy, Zinh, and Zzzz.
-- The only scripts found below U+0370 (the first code point of the Greek
-- and Coptic block) are Latn, Zyyy, Zinh, and Zzzz.
-- See the codepage in the [[UTF-8]] article.
if not str:find "[\205-\244]" then
for codepoint in mw.ustring.gcodepoint(str) do
if lookup_script(codepoint) == "Latn" then
return true
end
end
end
local Latn = false
for codepoint in mw.ustring.gcodepoint(str) do
local script = lookup_script(codepoint)
if script == "Latn" then
Latn = true
elseif not (script == "Zyyy" or script == "Zinh"
or script == "Zzzz") then
return false
end
end
return Latn
end
-- Checks that a string contains only characters belonging to right-to-left
-- scripts, or characters of ignorable scripts.
function p.is_rtl(str)
require "libraryUtil".checkType("get_best_script", 1, str, "string")
str = mw.text.decode(str, true)
-- Search for the leading bytes that introduce the UTF-8 encoding of the
-- code points U+0580-U+10FFFF. If they are not found, the string can only
-- have characters from a left-to-right script, because the first code point
-- in a right-to-left script is U+0591, in the Hebrew block.
if not str:find "[\214-\244]" then
return false
end
local result = false
local rtl = loader.scripts.rtl
for codepoint in mw.ustring.gcodepoint(str) do
local script = lookup_script(codepoint)
if rtl[script] then
result = true
elseif not (script == "Zyyy" or script == "Zinh"
or script == "Zzzz") then
return false
end
end
return result
end


local function get_codepoint(args, arg)
function export.get_entry_title(codepoint)
local codepoint_string = args[arg]
if udata.unsupported_title[codepoint] then
or errorf(2, "Parameter %s is required", tostring(arg))
return udata.unsupported_title[codepoint]
local codepoint = tonumber(codepoint_string, 16)
or errorf(2, "Parameter %s is not a code point in hexadecimal base",
tostring(arg))
if not (0 <= codepoint and codepoint <= 0x10FFFF) then
errorf(2, "code point in parameter %s out of range", tostring(arg))
end
end
return codepoint
if lookup_control(codepoint) ~= "assigned" then
end
return nil
 
local function get_func(args, arg, prefix)
local suffix = args[arg]
or errorf(2, "Parameter %s is required", tostring(arg))
suffix = mw.text.trim(suffix)
local func_name = prefix .. suffix
local func = p[func_name]
or errorf(2, "There is no function '%s'", func_name)
return func
end
 
-- This function allows any of the "lookup" functions to be invoked. The first
-- parameter is the word after "lookup_"; the second parameter is the code point
-- in hexadecimal base.
function p.lookup(frame)
local func = get_func(frame.args, 1, "lookup_")
local codepoint = get_codepoint(frame.args, 2)
local result = func(codepoint)
if func == p.lookup_name then
-- Prevent code point labels such as <control-0000> from being
-- interpreted as HTML tags.
result = result:gsub("<", "&lt;")
end
return result
end
 
function p.is(frame)
local func = get_func(frame.args, 1, "is_")
-- is_Latin and is_valid_pagename take strings.
if func == p.is_Latin or func == p.is_valid_pagename or func == p.is_rtl then
return (func(frame.args[2]))
else -- The rest take code points.
local codepoint = get_codepoint(frame.args, 2)
return (func(codepoint)) -- Adjust to one result.
end
end
return mw.ustring.char(codepoint)
end
end


return p
return export
Anonymous user
Retrieved from "https://wiki.imperivm-romanvm.com/wiki/Special:MobileDiff/1478"