Module:Formatnum/sandbox
Jump to navigation
Jump to search
CodeDiscussionEditHistoryLinksLink count Subpages:DocumentationTestsResultsSandboxLive code All modules
Lua
Documentation for this module may be created at Module:Formatnum/sandbox/doc
Code
-- This module is intended to replace the functionality of Template:Formatnum and related templates.
local p = {}
function p.main(frame)
local args = frame:getParent().args
local prec = args.prec or ''
local sep = args.sep or ''
local number = args[1] or args.number or ''
local lang = args[2] or args.lang or ''
-- validate the language parameter within MediaWiki's caller frame
if lang == "arabic-indic" then -- only for back-compatibility ("arabic-indic" is not a SupportedLanguage)
lang = "fa" -- better support than "ks"
elseif lang == '' or not mw.language.isSupportedLanguage(lang) then
-- Note that 'SupportedLanguages' are not necessarily 'BuiltinValidCodes', and so they are not necessarily
-- 'KnownLanguages' (with a language name defined at least in the default localisation of the local wiki).
-- But they all are ValidLanguageCodes (suitable as Wiki subpages or identifiers: no slash, colon, HTML tags, or entities)
-- In addition, they do not contain any capital letter in order to be unique in page titles (restriction inexistant in BCP47),
-- but they may violate the standard format of BCP47 language tags for specific needs in MediaWiki.
-- Empty/unspecified and unsupported languages are treated here in Commons using the user's language,
-- instead of the local 'ContentLanguage' of the Wiki.
lang = frame:callParserFunction("Int", "Lang") -- get user's chosen language
end
return p.formatNum(number, lang, prec, sep ~= '')
end
-- Substitution of decimal digits for languages not supported by mw.language:formatNum() in core Lua libraries for MediaWiki
local localizedDigits = {}; do
local function makeDigits(zero)
local digits = {}
for i = 0, 9 do
digits[string.char(0x30 + i)] = mw.ustring.char(zero + i)
end
return digits
end
for k, v in pairs({
arab = 0x00660, -- Arabic
arabext = 0x006F0, -- Extended Arabic
nkoo = 0x007C0, -- N’ko
deva = 0x00966, -- Devanagari
beng = 0x009E6, -- Bengla
guru = 0x00A66, -- Gurmukhi
gujr = 0x00AE6, -- Gujarati
orya = 0x00B66, -- Oriya (or Odiya)
taml = 0x00BE6, -- Tamil
telu = 0x00C66, -- Telugu
knda = 0x00CE6, -- Kannada (or Kannara)
mlymold = 0x00D66, -- Malayalam (old)
sinhold = 0x00DE6, -- Sinhalese (old)
thai = 0x00E50, -- Thai
laoo = 0x00ED0, -- Lao
tibt = 0x00F20, -- Tibetan
mymr = 0x01040, -- Myanmar
mymrshn = 0x01090, -- Myanmar (Shan variant)
khmr = 0x017E0, -- Khmer
mongold = 0x01810, -- Mongolian (old)
limb = 0x01946, -- Limbu
talu = 0x019D0, -- New Tai Lue
lanahor = 0x01A80, -- Tai Tham Hora
lana = 0x01A90, -- Tai Tham Tham (Lanna)
bali = 0x01B50, -- Balinese
sund = 0x01BB0, -- Sundanese
lepc = 0x01C40, -- Lepcha (Rong)
olck = 0x01C50, -- Ol Chiki
vaii = 0x0A620, -- Vai
saur = 0x0A8D0, -- Saurashtra
kali = 0x0A900, -- Kayah Li
java = 0x0A9D0, -- Javanese
mymrtai = 0x01040, -- Myanmar Tai
cham = 0x0AA50, -- Cham
mtei = 0x0ABF0, -- Meetei Mayek
zyyyfull = 0x0FF10, -- Fullwidth (for CJK)
osma = 0x104A0, -- Osmanya
rohg = 0x10D30, -- Hanifi Royinga
brah = 0x11066, -- Brahmi
sora = 0x110F0, -- Sora Sompeng
cakm = 0x11136, -- Chakma
shrd = 0x111D0, -- Sharada
sind = 0x112F0, -- Khudawadi (Sindhi)
newa = 0x11450, -- Newa
tirh = 0x114D0, -- Tirhuta
modi = 0x11650, -- Modi
takr = 0x116C0, -- Takri
ahom = 0x11730, -- ahom
wara = 0x118E0, -- Warang Citi
diak = 0x11950, -- Dives Akuru
bhks = 0x11C50, -- Bhaiksuki
gonm = 0x11D50, -- Masaram Gondi
gong = 0x11DA0, -- Gunjala Gondi
mroo = 0x16A60, -- Mro
hmng = 0x16B50, -- Pahawh Hmong
zmthb = 0x1D7CE, -- Mathematical Bold
zmthd = 0x1D7D8, -- Mathematical Double-Struck
zmths = 0x1D7E2, -- Mathematical Sans-Serif
zmthsb = 0x1D7EC, -- Mathematical Sans-Serif Bold
zmthm = 0x1D7CE, -- Mathematical Monospace
hmnp = 0x1E140, -- Nyiakeng Puachue Hmong
wcho = 0x1E2F0, -- Wancho
adlm = 0x1E950, -- Adlam
zyyysegm = 0x1FBF0, -- Segmented
}) do
localizedDigits['*-' .. k] = makeDigits(v)
end
for k, v in pairs({
['*-aran'] = '*-arabext', -- Alias for ISO 15924 script variant code.
-- Languages with missing support in mw.language:formatNum
['ml-old'] = '*-mylm-old', -- Legacy alias in Wikimedia Commons.
ta = '*-taml',
te = '*-telu',
th = '*-thai',
}) do
localizedDigits[k] = localizedDigits[v]
end
end
function p.formatNum(number, lang, prec, compact)
-- Do not alter the specified value when it is not a valid number, return it as is.
local tmp = tonumber(number)
if tmp == nil then
return number
end
-- Basic ASCII-only formatting (without paddings).
number = tostring(tmp) -- May return an English number ASCII string with scientific notation.
-- We no longer need the current `tmp` variable beyond this point.
-- Check the presence of an exponent (incorrectly managed in mw.language:FormatNum()
-- and even forgotten due to an internal bug, e.g. in Hindi).
local sep, exponent = string.find(number, '[Ee]')
if sep then
-- Strip the '[Ee]' found in the exponent, keep its sign.
exponent = string.sub(number, sep + 1, string.len(number))
-- Leading part of the number with its sign, and possible dot and decimals
number = string.sub(number, 1, sep - 1)
end -- We no longer need the current `sep` variable beyond this point.
-- Check the minimum precision requested (in the number string now without its exponent part).
prec = tonumber(prec) -- Returns nil if not specified as a true number.
if prec then
prec = math.floor(prec)
if prec < 0 then
prec = nil -- Discard an incorrect precision (must be positive or zero).
-- Maximum number of decimals displayed by tostring(number):
-- tostring(0.00000000000005) displays '5e-14'.
-- tostring(1.00000000000005) displays '1';
-- tostring(1.00000000000006) displays 13 decimals after '.' (all digits but the last '6');
-- tostring(9.99999999999994) displays 13 decimals after '.' (all digits but the last '4');
-- tostring(9.99999999999995) displays '10'.
-- tostring(9.99999999999994e+13) displays '99999999999999' (14 digits, all but the last '4');
-- tostring(9.99999999999995e+13) displays '1e+14'.
-- tostring(99999999999999.4) displays '99999999999999' (14 digits, all but the last '4');
-- tostring(99999999999999.5) displays '1e+14'.
-- Beside the exponent part, it prints at most 14 significant digits, including at least 1 on each side of '.'
-- It also uses the exponential notation for numbers below 1e-4 in magnitude.
-- There may be also up to 3 non-significant zeroes (not counted) after the dot before the first
-- significant digit in decimals, so there may be up to 17 decimals including these 3 zeroes.
-- Below 1e-4 or starting at 1e+14, tostring uses an exponential notation with 1 digit before the '.'
elseif prec > 14 then
tmp = math.abs(tonumber(number))
if tmp >= 1 or tmp < 1e-4 then
prec = 14
elseif prec > 15 and tmp >= 1e-1 then
prec = 15
elseif prec > 16 and tmp >= 1e-2 then
prec = 16
elseif prec > 17 then -- tmp >= 1e-3
prec = 17
end
end
end -- We no longer use the current `tmp` variable beyond this point.
-- Preprocess the precision of number in the ASCII-only string (round it if necessary).
if prec then -- The last decimal of the fraction part may need to be rounded.
sep = string.find(number, '.', 1, true) -- Plain search, no pattern.
if sep then
prec = sep + prec - string.len(number) -- Effective number of trailing decimals to add or remove.
if prec < 0 then -- Removing decimals in excess requires rounding.
-- Get the most significant decimal in excess.
tmp = string.sub(number, string.len(number) + prec + 1, string.len(number) + prec + 1)
if tmp < '5' then -- Rounding down the decimals is just truncating them.
number = string.sub(number, 1, string.len(number) + prec)
else -- Rounding up the decimals (complex case).
-- Get the decimals to preserve (after the dot), then
-- get its incremented value by prepending a '1' and test the carry on it.
tmp = tostring(tonumber('1' .. string.sub(number, sep + 1, string.len(number) + prec)) + 1)
-- The leading (dummy) digit in decimals is either '1', or '2' if there was a carry.
-- With a carry, we must increment the absolute number (after the possible sign) before the dot.
if string.sub(tmp, 1, 1) == '1' then
number = string.sub(number, 1, sep) .. string.sub(tmp, 2)
elseif string.sub(number, 1, 1) == '-' then
number = '-' .. tostring(tonumber(string.sub(number, 2, sep)) + 1) .. string.sub(tmp, 2)
else
number = tostring(tonumber(string.sub(number, 1, sep)) + 1) .. string.sub(tmp, 2)
end
end -- We no longer need the current `tmp` variable beyond this point.
if string.sub(number, -1) == '.' then
-- Special case where rounding up or down leaves a trailing dot
number = string.sub(number, 1, -2)
end
elseif prec > 0 then -- adding missing decimals (trailing zeroes)
number = number .. string.rep('0', prec)
end
elseif prec > 0 then -- adding missing decimals (trailing zeroes)
number = number .. '.' .. string.rep('0', prec)
end
end -- We no longer need the current `prec` variable beyond this point.
-- Language-dependant grouping, and remapping of signs, digits, and separators in number and exponent.
if type(lang) == 'string' then
lang = string.lower(lang)
if mw.language.isKnownLanguageTag(lang) then
-- Convert number to localized digits, decimal separator, and group separators.
tmp = mw.getLanguage(lang) -- Caveat: can load localized resources for up to 20 languages.
-- Insert appropriate group separators (group separators and their positioning and are language-dependant).
number = tmp:formatNum(tonumber(number), { noCommafy = compact })
-- Localize the base-10 exponent (with its sign but without grouping separators).
if exponent then
exponent = tmp:formatNum(tonumber(exponent), { noCommafy = true })
end -- We no longer need the current `tmp` variable beyond this point.
end
end
if exponent then
-- A few Indic languages need bracketing to make sense for the scientific notation with a Latin 'e'. See CLDR:
-- https://unicode-org.github.io/cldr-staging/charts/latest/by_type/numbers.number_formatting_patterns.html#Standard_Patterns_
tmp = ({
gu = 1, hi = 1, mai = 1, mr = 1, pa = 1, sa = 1, html = 2,
})[not compact and lang or '']
if tmp == 1 then
number = '[' .. number .. 'e' .. exponent .. ']'
elseif tmp == 2 then
number = number .. ' × 10<sup>' .. exponent .. '</sup>'
else
number = number .. 'e' .. exponent
end
end
-- Special cases for substitution of ASCII digits (missing support in Lua core libraries for some languages, known or not).
if lang and localizedDigits[lang] then
number = string.gsub(number, '[0-9]', localizedDigits[lang])
end
return number
end
return p