Module:Lang: Difference between revisions

Revision as of 13:24, 5 January 2025

Revision as of 16:45, 5 January 2025

Line 1,055:

local function is_latin (text, script, transl)

text = mw.text.decode (text); -- convert html entities to characters

script = (script and script:lower()) or nil; -- ensure lowercase for comparisons

local whitelist = '[ʻʼʽʾʿΔαβγδθσφχϑъьӾӿᾱῑ῾上入去平]'

local whitelist = '[ʻʼʽʾʿΔαβγδθσφχϑъьӾӿᾱῑ῾ʻʼʽʾʿΔαβγδθλσφχϑъьӾӿᾱῑ῾上入去平]'

if not transl then -- testing {{lang}} or {{langx}} <text>

if '' == mw.ustring.gsub (text, whitelist, '') then -- when <text> uses only whitelisted characters

return ('latn' == script); -- true when latn script; false when other script or no script

end

end

if mw.ustring.find (text, whitelist) then -- does <text> contain characters from the whitelist?

-- text = text:gsub ('θ', ''); -- remove theta from <text>

text = mw.ustring.gsub (text, whitelist, 'x'); -- replace whitelisted characters from <text> with known Latn-script char 'x'

-- if 0 == text:len() then -- will be zero if theta was the only character in <text>

if transl'' == text then -- notwill nilbe forempty {{transliteration}};string assumeif Latinthe thetaonly becausecharacter(s) thisin is<text> 'transliteration'are in template<whitelist>

if transl then -- not nil for {{transliteration}}; assume Latn because this is 'transliteration' template

return true;

end

return (script and ('latn' == script:lower())) or false; -- {{lang}}, {{langx}}, and {{lang-??}}; true when script is Latn; false else

end

return unicode.is_Latin (text); -- return true when all characters in modified <text> are Latn script; false else