Module:Language: Difference between revisions
From EverybodyWiki Bios & Wiki
WikiMaster (talk | contribs) m 1 revision imported |
m Add new revision |
||
| Line 1: | Line 1: | ||
require('Module:No globals') | require('Module:No globals') | ||
local m_data = mw.loadData("Module:Language/data") | local m_data = mw.loadData("Module:Language/data") | ||
local langData = m_data.languages or m_data | |||
local p = {} | local p = {} | ||
local function checkForString(variable) | local function checkForString(variable) | ||
| Line 16: | Line 10: | ||
local function makeLinkedName(languageCode) | local function makeLinkedName(languageCode) | ||
local data = | local data = langData[languageCode] | ||
local article = data["article"] | local article = data["article"] | ||
local name = data["Wikipedia_name"] or data["name"] | local name = data["Wikipedia_name"] or data["name"] | ||
| Line 23: | Line 17: | ||
local function makeEntryName(word, languageCode) | local function makeEntryName(word, languageCode) | ||
local data = | local data = langData[languageCode] | ||
local ugsub = mw.ustring.gsub | |||
word = tostring(word) | word = tostring(word) | ||
if word == nil then | if word == nil then | ||
| Line 31: | Line 26: | ||
else | else | ||
-- Remove bold and italics, so that words that contain bolding or emphasis can be linked without piping. | -- Remove bold and italics, so that words that contain bolding or emphasis can be linked without piping. | ||
word = gsub( | word = word:gsub("\'\'\'", "") | ||
word = gsub( | word = word:gsub("\'\'", "") | ||
if data == nil then | if data == nil then | ||
return word | return word | ||
| Line 40: | Line 35: | ||
return word | return word | ||
else | else | ||
for regex, replacement in pairs(replacements) do | -- Decompose so that the diacritics of characters such | ||
-- as á can be removed in one go. | |||
-- No need to compose at the end, because the MediaWiki software | |||
-- will handle that. | |||
if replacements.decompose then | |||
word = mw.ustring.toNFD(word) | |||
for i, from in ipairs(replacements.from) do | |||
word = ugsub( | |||
word, | |||
from, | |||
replacements.to and replacements.to[i] or "") | |||
end | |||
else | |||
for regex, replacement in pairs(replacements) do | |||
word = ugsub(word, regex, replacement) | |||
end | |||
end | end | ||
return word | return word | ||
| Line 47: | Line 56: | ||
end | end | ||
end | end | ||
end | |||
p.makeEntryName = makeEntryName | |||
local function fixScriptCode(firstLetter, threeLetters) | |||
return string.upper(firstLetter) .. string.lower(threeLetters) | |||
end | end | ||
| Line 54: | Line 69: | ||
if codes == nil or codes == "" then | if codes == nil or codes == "" then | ||
errorText = 'no language or script code provided' | errorText = 'no language or script code provided' | ||
elseif find(codes, "^%s*%a%a%a?%s*$") or find( | |||
-- Private-use subtag: x followed by one or more sequences of 1-8 lowercase | |||
-- letters separated by hyphens. This only allows for one sequence, as it is | |||
-- needed for proto-languages such as ine-x-proto (Proto-Indo-European). | |||
elseif codes:find("^%s*%a%a%a?%-x%-%a%a?%a?%a?%a?%a?%a?%a?$") then | |||
languageCode, scriptCode = | |||
codes:match("^%s*(%a%a%a%-x%-%a%a?%a?%a?%a?%a?%a?%a?)%-?(.*)$") | |||
if not languageCode then | |||
errorText = '<code>'..codes..'</code> is not a valid language or script code.' | |||
elseif scriptCode ~= "" and not scriptCode:find("%a%a%a%a") then | |||
errorText = '<code>'..scriptCode..'</code> is not a valid script code.' | |||
else | |||
scriptCode = scriptCode:gsub( | |||
"(%a)(%a%a%a)", | |||
fixScriptCode, | |||
1 | |||
) | |||
end | |||
elseif codes:find("^%s*%a%a%a?%s*$") or codes:find("^%s*%a%a%a?%-%a%a%a%a%s*$") then | |||
-- A three- or two-letter lowercase sequence at beginning of first parameter | -- A three- or two-letter lowercase sequence at beginning of first parameter | ||
languageCode = | languageCode = | ||
find( | codes:find("^%s*%a%a%a?") and ( | ||
match( | codes:match("^%s*(%l%l%l?)") | ||
or | or codes:match("^%s*(%a%a%a?)") | ||
:gsub("(%a%a%a?)", string.lower, 1) | |||
"(%a%a%a?)", | |||
) | ) | ||
-- One uppercase and three lowercase letters at the end of the first parameter | -- One uppercase and three lowercase letters at the end of the first parameter | ||
scriptCode = | scriptCode = | ||
find( | codes:find("%a%a%a%a%s*$") and ( | ||
match( | codes:match("(%u%l%l%l)%s*$") | ||
or gsub( | or gsub( | ||
match( | codes:match("(%a%a%a%a)%s*$"), | ||
"(%a)(%a%a%a)", | "(%a)(%a%a%a)", | ||
fixScriptCode, | |||
1 | 1 | ||
) | ) | ||
) | ) | ||
elseif find( | elseif codes:find("^%s*%a%a%a%-%a%a%a$") then | ||
languageCode = match( | languageCode = codes:match("^%s*%l%l%l%-%l%l%l$") and codes:match("^%s*%l%l%l%-%l%l%l$") | ||
elseif find( | or codes:match("^%s*%a%a%a%-%a%a%a$"):gsub("(%a%a%a?)", string.lower, 1) | ||
languageCode, invalidCode = match( | elseif codes:find("^%s*%a%a%a?") then | ||
languageCode = lower(languageCode) | languageCode, invalidCode = codes:match("^%s*(%a%a%a?)%-?(.*)") | ||
languageCode = string.lower(languageCode) | |||
errorText = '<code>'..invalidCode..'</code> is not a valid script code.' | errorText = '<code>'..invalidCode..'</code> is not a valid script code.' | ||
elseif find( | elseif codes:find("%-?%a%a%a%a%s*$") then | ||
invalidCode, scriptCode = match( | invalidCode, scriptCode = codes:match("(.*)%-?(%a%a%a%a)%s*$") | ||
scriptCode = gsub( | scriptCode = gsub( | ||
scriptCode, | scriptCode, | ||
"(%a)(%a%a%a)", | "(%a)(%a%a%a)", | ||
fixScriptCode | |||
) | ) | ||
errorText = '<code>'..invalidCode..'</code> is not a valid language code.' | errorText = '<code>'..invalidCode..'</code> is not a valid language code.' | ||
| Line 100: | Line 125: | ||
errorText = '<code>'..codes..'</code> is not a valid language or script code.' | errorText = '<code>'..codes..'</code> is not a valid language or script code.' | ||
end | end | ||
if not scriptCode then | if not scriptCode or scriptCode == "" then | ||
scriptCode = require("Module: | scriptCode = require("Module:Unicode data").is_Latin(text) and "Latn" or "unknown" | ||
end | end | ||
if errorText then | if errorText then | ||
| Line 108: | Line 133: | ||
errorText = "" | errorText = "" | ||
end | end | ||
languageCode = m_data.redirects[languageCode] or languageCode | |||
return languageCode, scriptCode, errorText | return languageCode, scriptCode, errorText | ||
end | end | ||
local function tag(text, languageCode, script, italics) | local function tag(text, languageCode, script, italics) | ||
local data = | local data = langData[languageCode] | ||
-- Use Wikipedia code if it has been given: for instance, | |||
-- Proto-Indo-European has the Wiktionary code "ine-pro" but the Wikipedia | |||
-- code "ine-x-proto". | |||
languageCode = data and data.Wikipedia_code or languageCode | |||
local italicize = script == "Latn" and italics | local italicize = script == "Latn" and italics | ||
| Line 133: | Line 163: | ||
return table.concat(out) | return table.concat(out) | ||
end | end | ||
function p.lang(frame) | function p.lang(frame) | ||
| Line 143: | Line 175: | ||
local languageCode, scriptCode, errorText = getCodes(codes, text) | local languageCode, scriptCode, errorText = getCodes(codes, text) | ||
local italics = args.italics or args.i | local italics = args.italics or args.i or args.italic | ||
italics = not (italics == "n" or italics == "-") | italics = not (italics == "n" or italics == "-" or italics == "no") | ||
return tag(text, languageCode, scriptCode, italics) .. errorText | return tag(text, languageCode, scriptCode, italics) .. errorText | ||
| Line 150: | Line 182: | ||
local function linkToWiktionary(entry, linkText, languageCode) | local function linkToWiktionary(entry, linkText, languageCode) | ||
local data = | local data = langData[languageCode] | ||
local name | local name | ||
if languageCode then | if languageCode then | ||
| Line 159: | Line 191: | ||
name = mw.language.fetchLanguageName(languageCode, 'en') | name = mw.language.fetchLanguageName(languageCode, 'en') | ||
else | else | ||
error("No name for the language " .. (languageCode or | error("No name for the language " .. ("%q"):format(languageCode or nil) .. " could be found") | ||
end | end | ||
if sub( | if entry:sub(1, 1) == "*" then | ||
if name ~= "" then | |||
entry = "Reconstruction:" .. name .. "/" .. entry:sub(2) | |||
else | |||
error("Language name is empty") | |||
end | |||
elseif data and data.type == "reconstructed" then | |||
mw.log("Reconstructed language without asterisk:", languageCode, name, entry) | |||
local frame = mw.getCurrentFrame() | |||
-- Track reconstructed entries with no asterisk by transcluding | |||
-- a nonexistent template. This technique is used in Wiktionary: | |||
-- see [[wikt:Module:debug]]. | |||
-- [[Special:WhatLinksHere/tracking/wikt-lang/reconstructed with no asterisk]] | |||
pcall(frame.expandTemplate, frame, | |||
{ title = 'tracking/wikt-lang/reconstructed with no asterisk' }) | |||
if name ~= "" then | |||
entry = "Reconstruction:" .. name .. "/" .. entry | |||
else | |||
error("Language name is empty") | |||
end | |||
elseif data and data.type == "appendix" then | |||
if name ~= "" then | if name ~= "" then | ||
entry = " | entry = "Appendix:" .. name .. "/" .. entry | ||
else | else | ||
error("Language name is empty") | error("Language name is empty") | ||
| Line 185: | Line 237: | ||
local word1 = args[2] or nil | local word1 = args[2] or nil | ||
local word2 = args[3] or nil | local word2 = args[3] or nil | ||
if not args[2] then | |||
error("Parameter 2 is required") | |||
end | |||
local languageCode, scriptCode, errorText = getCodes(codes, word1) | local languageCode, scriptCode, errorText = getCodes(codes, word1) | ||
Revision as of 13:28, 14 March 2019
| This module is rated as beta, and is ready for widespread use. It is still new and should be used with some caution to ensure the results are as expected. |
Usage
This module is inspired by the templates {{m}} and {{l}} and their associated modules on Wiktionary. It has a Wiktionary link function that links to the correct section of the Wiktionary entry, and applies correct language formatting and italics. The language formatting function does most of what {{lang}} does, as well as applying appropriate italicization.
It should eventually be able to generate the text in numerous {{lang-xx}} templates, such as {{lang-fr}}. (Or even more ambitious, it could generate automatic transliteration and IPA for some languages, to be used in templates like {{lang-rus}}.)
Examples
* {{#invoke:language|wikt|en|language}}
* {{#invoke:language|wikt|en|language|languages}}
* {{#invoke:Language|wikt|fr|bourguignon}}
* {{#invoke:Language|wikt|la|homō}}
* {{#invoke:Language|wikt|grc|δημοκρατίᾱ}}
* {{#invoke:Language|wikt|ru|язы́к}}
* {{#invoke:Language|lang|ru|язы́к}}
* {{#invoke:Language|wikt|ar|العَرَبِيَّة}}
* {{#invoke:Language|wikt|fa|فارسی}}
Invalid codes
* {{#invoke:language|wikt|EN|language}}
* {{#invoke:language|wikt|En|language|languages}}
* {{#invoke:Language|wikt|La|homō}}
* {{#invoke:Language|wikt|Grc|δημοκρατίᾱ}}
* {{#invoke:Language|wikt|Ru|язы́к}}
* {{#invoke:Language|lang|Ru|язы́к}}
* {{#invoke:Language|wikt|Ar|العَرَبِيَّة}}
* {{#invoke:Language|wikt|Fa|فارسی}}
Errors
* {{#invoke:Language|wikt|en-Latin|language}}
- language [
Latinis not a valid script code.]
Subpages
In this modulespace:
Module:
| Language |
- Language/data
- Language/data/ISO 639-1
- Language/data/ISO 639-1/make
- Language/data/ISO 639-2
- Language/data/ISO 639-2/make
- Language/data/ISO 639-3
- Language/data/ISO 639-3/doc
- Language/data/ISO 639-3/make
- Language/data/ISO 639-5
- Language/data/ISO 639-5/make
- Language/data/ISO 639 name to code
- Language/data/ISO 639 name to code/make
- Language/data/ISO 639 override
- Language/data/iana
- Language/data/iana/make
- Language/data/iana languages
- Language/data/iana languages/doc
- Language/data/iana regions
- Language/data/iana scripts
- Language/data/iana suppressed scripts
- Language/data/iana variants
- Language/data/iso 15924
- Language/data/iso 3166-1
- Language/data/names
- Language/data/un m.49
- Language/data/wp languages
- Language/data/wp languages/doc
- Language/doc
- Language/external links
- Language/external links/doc
- Language/name
- Language/name/data
- Language/name/data/ISO 639-3 data extraction tool
- Language/name/data/ISO 639-3 data extraction tool/doc
- Language/name/data/ISO 639 synonym extraction tool
- Language/name/data/doc
- Language/name/data/iana data extraction tool
- Language/name/data/iana data extraction tool/doc
- Language/name/doc
- Language/name/testcases
- Language/sandbox
- Language/scripts
- Language/scripts/codepoints
- Language/scripts/codepoints/doc
- Language/scripts/codepoints/make
- Language/scripts/codepoints adjacent merged
- Language/scripts/codepoints verbatim
- Language/scripts/data
- Language/scripts/data/doc
- Language/scripts/data generation
- Language/scripts/data generation/doc
- Language/scripts/doc
- Language/scripts/testcases
- Language/text
- Language/text/doc
Module talk:
require('Module:No globals')
local m_data = mw.loadData("Module:Language/data")
local langData = m_data.languages or m_data
local p = {}
local function checkForString(variable)
return variable ~= "" and variable ~= nil
end
local function makeLinkedName(languageCode)
local data = langData[languageCode]
local article = data["article"]
local name = data["Wikipedia_name"] or data["name"]
return "[[" .. article .. "|" .. name .. "]]: "
end
local function makeEntryName(word, languageCode)
local data = langData[languageCode]
local ugsub = mw.ustring.gsub
word = tostring(word)
if word == nil then
error("The function makeEntryName requires a string argument")
elseif word == "" then
return ""
else
-- Remove bold and italics, so that words that contain bolding or emphasis can be linked without piping.
word = word:gsub("\'\'\'", "")
word = word:gsub("\'\'", "")
if data == nil then
return word
else
local replacements = data and data["replacements"]
if replacements == nil then
return word
else
-- Decompose so that the diacritics of characters such
-- as á can be removed in one go.
-- No need to compose at the end, because the MediaWiki software
-- will handle that.
if replacements.decompose then
word = mw.ustring.toNFD(word)
for i, from in ipairs(replacements.from) do
word = ugsub(
word,
from,
replacements.to and replacements.to[i] or "")
end
else
for regex, replacement in pairs(replacements) do
word = ugsub(word, regex, replacement)
end
end
return word
end
end
end
end
p.makeEntryName = makeEntryName
local function fixScriptCode(firstLetter, threeLetters)
return string.upper(firstLetter) .. string.lower(threeLetters)
end
local function getCodes(codes, text)
local languageCode, scriptCode, invalidCode
local errorText
if codes == nil or codes == "" then
errorText = 'no language or script code provided'
-- Private-use subtag: x followed by one or more sequences of 1-8 lowercase
-- letters separated by hyphens. This only allows for one sequence, as it is
-- needed for proto-languages such as ine-x-proto (Proto-Indo-European).
elseif codes:find("^%s*%a%a%a?%-x%-%a%a?%a?%a?%a?%a?%a?%a?$") then
languageCode, scriptCode =
codes:match("^%s*(%a%a%a%-x%-%a%a?%a?%a?%a?%a?%a?%a?)%-?(.*)$")
if not languageCode then
errorText = '<code>'..codes..'</code> is not a valid language or script code.'
elseif scriptCode ~= "" and not scriptCode:find("%a%a%a%a") then
errorText = '<code>'..scriptCode..'</code> is not a valid script code.'
else
scriptCode = scriptCode:gsub(
"(%a)(%a%a%a)",
fixScriptCode,
1
)
end
elseif codes:find("^%s*%a%a%a?%s*$") or codes:find("^%s*%a%a%a?%-%a%a%a%a%s*$") then
-- A three- or two-letter lowercase sequence at beginning of first parameter
languageCode =
codes:find("^%s*%a%a%a?") and (
codes:match("^%s*(%l%l%l?)")
or codes:match("^%s*(%a%a%a?)")
:gsub("(%a%a%a?)", string.lower, 1)
)
-- One uppercase and three lowercase letters at the end of the first parameter
scriptCode =
codes:find("%a%a%a%a%s*$") and (
codes:match("(%u%l%l%l)%s*$")
or gsub(
codes:match("(%a%a%a%a)%s*$"),
"(%a)(%a%a%a)",
fixScriptCode,
1
)
)
elseif codes:find("^%s*%a%a%a%-%a%a%a$") then
languageCode = codes:match("^%s*%l%l%l%-%l%l%l$") and codes:match("^%s*%l%l%l%-%l%l%l$")
or codes:match("^%s*%a%a%a%-%a%a%a$"):gsub("(%a%a%a?)", string.lower, 1)
elseif codes:find("^%s*%a%a%a?") then
languageCode, invalidCode = codes:match("^%s*(%a%a%a?)%-?(.*)")
languageCode = string.lower(languageCode)
errorText = '<code>'..invalidCode..'</code> is not a valid script code.'
elseif codes:find("%-?%a%a%a%a%s*$") then
invalidCode, scriptCode = codes:match("(.*)%-?(%a%a%a%a)%s*$")
scriptCode = gsub(
scriptCode,
"(%a)(%a%a%a)",
fixScriptCode
)
errorText = '<code>'..invalidCode..'</code> is not a valid language code.'
else
errorText = '<code>'..codes..'</code> is not a valid language or script code.'
end
if not scriptCode or scriptCode == "" then
scriptCode = require("Module:Unicode data").is_Latin(text) and "Latn" or "unknown"
end
if errorText then
errorText = ' <span style="font-size: smaller">[' .. errorText .. ']</span>'
else
errorText = ""
end
languageCode = m_data.redirects[languageCode] or languageCode
return languageCode, scriptCode, errorText
end
local function tag(text, languageCode, script, italics)
local data = langData[languageCode]
-- Use Wikipedia code if it has been given: for instance,
-- Proto-Indo-European has the Wiktionary code "ine-pro" but the Wikipedia
-- code "ine-x-proto".
languageCode = data and data.Wikipedia_code or languageCode
local italicize = script == "Latn" and italics
if not text then text = "[text?]" end
local textDirectionMarkers = { "", "", "" }
if data and data["direction"] == "rtl" then
textDirectionMarkers = { ' dir="rtl"', '‏', '‎' }
end
local out = { textDirectionMarkers[2] }
if italicize then
table.insert(out, "<i lang=\"" .. languageCode .. "\" xml:lang=\"" .. languageCode .. "\"" .. textDirectionMarkers[1] .. ">" .. text .. "</i>")
else
table.insert(out, "<span lang=\"" .. languageCode .. "\" xml:lang=\"" .. languageCode .. "\"" .. textDirectionMarkers[1] .. ">" .. text .. "</span>")
end
table.insert(out, textDirectionMarkers[3])
return table.concat(out)
end
function p.lang(frame)
local parent = frame:getParent()
local args = parent.args[1] and parent.args or frame.args
local codes = args[1]
local text = args[2] or error("Provide text in the second parameter")
local languageCode, scriptCode, errorText = getCodes(codes, text)
local italics = args.italics or args.i or args.italic
italics = not (italics == "n" or italics == "-" or italics == "no")
return tag(text, languageCode, scriptCode, italics) .. errorText
end
local function linkToWiktionary(entry, linkText, languageCode)
local data = langData[languageCode]
local name
if languageCode then
if data and data.name then
name = data.name
elseif mw.language.fetchLanguageName(languageCode, 'en') ~= "" then
-- On other languages' wikis, use mw.getContentLanguage():getCode(), or replace with that wiki's language code.
name = mw.language.fetchLanguageName(languageCode, 'en')
else
error("No name for the language " .. ("%q"):format(languageCode or nil) .. " could be found")
end
if entry:sub(1, 1) == "*" then
if name ~= "" then
entry = "Reconstruction:" .. name .. "/" .. entry:sub(2)
else
error("Language name is empty")
end
elseif data and data.type == "reconstructed" then
mw.log("Reconstructed language without asterisk:", languageCode, name, entry)
local frame = mw.getCurrentFrame()
-- Track reconstructed entries with no asterisk by transcluding
-- a nonexistent template. This technique is used in Wiktionary:
-- see [[wikt:Module:debug]].
-- [[Special:WhatLinksHere/tracking/wikt-lang/reconstructed with no asterisk]]
pcall(frame.expandTemplate, frame,
{ title = 'tracking/wikt-lang/reconstructed with no asterisk' })
if name ~= "" then
entry = "Reconstruction:" .. name .. "/" .. entry
else
error("Language name is empty")
end
elseif data and data.type == "appendix" then
if name ~= "" then
entry = "Appendix:" .. name .. "/" .. entry
else
error("Language name is empty")
end
end
if entry and linkText then
return "[[wikt:" .. entry .. "#" .. name .. "|" .. linkText .. "]]"
else
error("linkToWiktionary needs a Wiktionary entry or link text, or both")
end
else
return "[[wikt:" .. entry .. "|" .. linkText .. "]]"
end
end
function p.wiktlang(frame)
local parent = frame:getParent()
local args = parent.args[1] and parent.args or frame.args
local codes = args[1] or nil
local word1 = args[2] or nil
local word2 = args[3] or nil
if not args[2] then
error("Parameter 2 is required")
end
local languageCode, scriptCode, errorText = getCodes(codes, word1)
local italics = args.italics or args.i
italics = not (italics == "n" or italics == "-")
local entry, linkText
if checkForString(word2) and checkForString(word1) then
entry = makeEntryName(word1, languageCode)
linkText = word2
elseif checkForString(word1) then
entry = makeEntryName(word1, languageCode)
linkText = word1
end
local out
if languageCode and entry and linkText then
out = tag(linkToWiktionary(entry, linkText, languageCode), languageCode, scriptCode, italics)
elseif entry and linkText then
out = linkToWiktionary(entry, linkText)
else
out = '<span style="font-size: smaller;">[text?]</span>'
end
if out and errorText then
return out .. errorText
else
return errorText or error("The function wiktlang generated nothing")
end
end
function p.wikt(frame)
local parent = frame:getParent()
local args = parent.args[1] and parent.args or frame.args
local codes = args[1] or nil
local word1 = args[2] or nil
local word2 = args[3] or nil
if not word1 then
error("Provide a word in parameter 2.")
end
local languageCode, scriptCode, errorText = getCodes(codes, word1)
local entry, linkText
if checkForString(word2) and checkForString(word1) then
entry = makeEntryName(word1, languageCode)
linkText = word2
elseif checkForString(word1) then
entry = makeEntryName(word1, languageCode)
linkText = word1
end
local out
if languageCode and entry and linkText then
out = linkToWiktionary(entry, linkText, languageCode)
elseif entry and linkText then
out = linkToWiktionary(entry, linkText)
else
out = '<span style="font-size: smaller;">[text?]</span>'
end
if out and errorText then
return out and out .. errorText
else
return errorText or error("The function wikt generated nothing")
end
end
return p
