local m_languages = mw.loadData("Module:languages")
local export = {}
-- Detect the script based on the first alphabetical characters of a string
function export.detect_script(text, lang)
local m_scripts = mw.loadData("Module:scripts")
local langinfo = m_languages[lang]
local scFix = false
-- Does this language have more than one script?
-- If not, we can bypass the detection for a speed bonus.
-- But always do the detection if the script is "None" or "Zyyy"
if not (langinfo.scripts[2] or langinfo.scripts[1] == "None" or langinfo.scripts[1] == "Zyyy") then
return langinfo.scripts[1], scFix
end
for i, script in ipairs(langinfo.scripts) do
local script2check = script:gsub(".-%-", "") -- removes the language code from script name, e.g. "nv-Latn" > "Latn"
if script2check == "Latf" or script2check == "Latinx" or script2check == "unicode" then
script2check = "Latn"
elseif script2check == "Hans" or script2check == "Hant" then
script2check = "Hani"
end
local scriptinfo = m_scripts[script2check] or error("The script code \"" .. script .. "\" is not valid.")
if scriptinfo.characters and mw.ustring.match(text, scriptinfo.characters) then
return script, scFix
end
end
scFix = (langinfo.scripts[1] ~= "Zyyy" and langinfo.scripts[1] ~= "None")
-- not written in native script(s); check for all scripts
-- TODO: This is slow; we really shouldn't be doing this!
for script, scriptinfo in pairs(m_scripts) do
if scriptinfo.characters and mw.ustring.match(text, "[%[%d%p%s]-" .. scriptinfo.characters) then
return script, scFix
end
end
return langinfo.scripts[1], scFix
end
-- Format the categories with the appropriate sort key
function export.format_categories(categories, lang, sort_key, sort_base)
NAMESPACE = NAMESPACE or mw.title.getCurrentTitle().nsText
if NAMESPACE == "" or NAMESPACE == "Appendix" then
PAGENAME = PAGENAME or mw.title.getCurrentTitle().text
SUBPAGENAME = SUBPAGENAME or mw.title.getCurrentTitle().subpageText
-- Generate a default language-independent sort key
sort_base = mw.ustring.lower(sort_base or SUBPAGENAME)
-- Remove initial hyphens and *
sort_base = mw.ustring.gsub(sort_base, "^[-־ـ*]+(.)",
"%1")
-- Remove anything in parentheses, as long as they are either preceded or followed by something
sort_base = mw.ustring.gsub(sort_base, "(.)%([^()]+%)", "%1")
sort_base = mw.ustring.gsub(sort_base, "%([^()]+%)(.)", "%1")
-- If there are language-specific rules to generate the key, use those
if lang then
local langinfo = m_languages[lang] or error("The language code \"" .. lang .. "\" is not valid.")
if langinfo.sort_key then
for i, from in ipairs(langinfo.sort_key.from) do
local to = langinfo.sort_key.to[i] or ""
sort_base = mw.ustring.gsub(sort_base, from, to)
end
end
end
if sort_key then
-- Gather some statistics regarding sort keys
if mw.ustring.lower(sort_key) == sort_base then
table.insert(categories, "Sort key tracking/redundant")
elseif lang and lang ~= "cmn" and lang ~= "ja" and lang ~= "zu" and lang ~= "nan" and lang ~= "yue" then
if lang == "ga" or lang == "gv" or lang == "nv" or lang == "roa-jer" or lang == "fr" or lang == "rm" or lang == "prg" or lang == "gd" or lang == "twf" or lang == "en" or lang == "ro" or lang == "egl" or lang == "roa-tar" or lang == "gl" or lang == "ast" or lang == "br" then
table.insert(categories, "Sort key tracking/needed/" .. lang)
else
table.insert(categories, "Sort key tracking/needed")
end
end
else
sort_key = sort_base
end
-- If the resulting key is the same as the wiki software's default, remove it
if sort_key == PAGENAME then
sort_key = nil
end
for key, cat in ipairs(categories) do
categories[key] = "[[Category:" .. cat .. (sort_key and "|" .. sort_key or "") .. "]]"
end
return table.concat(categories, "")
else
return ""
end
end
-- Used by {{categorize}}
function export.template_categorize(frame)
NAMESPACE = NAMESPACE or mw.title.getCurrentTitle().nsText
local format = frame.args["format"]
local args = frame:getParent().args
local lang = args[1]
local sort_key = args["sort"]; if sort_key == "" then sort_key = nil end
local categories = {}
if lang == "" or lang == nil then
if NAMESPACE == "Template" then
lang = "und"
else
error("Language code has not been specified. Please pass parameter 1 to the template.")
end
end
local langinfo = m_languages[lang] or error("The language code \"" .. lang .. "\" is not valid.")
local prefix = ""
if format == "pos" then
prefix = langinfo.names[1] .. " "
elseif format == "topic" then
prefix = lang .. ":"
end
local i = 2
local cat = args[i]
while cat do
if cat ~= "" then
table.insert(categories, prefix .. cat)
end
i = i + 1
cat = args[i]
end
return export.format_categories(categories, lang, sort_key)
end
return export
