Linguifex

Module:script utilities: Difference between revisions

← Older edit
No edit summary
m 1 revision imported
 
(2 intermediate revisions by 2 users not shown)
Line 2: Line 2:


local anchors_module = "Module:anchors"
local anchors_module = "Module:anchors"
local debug_track_module = "Module:debug/track"
local links_module = "Module:links"
local links_module = "Module:links"
local munge_text_module = "Module:munge text"
local munge_text_module = "Module:munge text"
Line 13: Line 14:
local require = require
local require = require
local toNFD = mw.ustring.toNFD
local toNFD = mw.ustring.toNFD
local dump = mw.dumpObject


--[==[
--[==[
Loaders for functions in other modules, which overwrite themselves with the target function when called. This ensures modules are only loaded when needed, retains the speed/convenience of locally-declared pre-loaded functions, and has no overhead after the first call, since the target functions are called directly in any subsequent calls.]==]
Loaders for functions in other modules, which overwrite themselves with the target function when called. This ensures modules are only loaded when needed, retains the speed/convenience of locally-declared pre-loaded functions, and has no overhead after the first call, since the target functions are called directly in any subsequent calls.]==]
local function embedded_language_links(...)
local function embedded_language_links(...)
embedded_language_links = require(links_module).embedded_language_links
embedded_language_links = require(links_module).embedded_language_links
return embedded_language_links(...)
return embedded_language_links(...)
end
end
 
local function format_categories(...)
local function find_best_script_without_lang(...)
format_categories = require(utilities_module).format_categories
find_best_script_without_lang = require(scripts_module).findBestScriptWithoutLang
return format_categories(...)
return find_best_script_without_lang(...)
end
end
 
local function get_script(...)
local function format_categories(...)
get_script = require(scripts_module).getByCode
format_categories = require(utilities_module).format_categories
return get_script(...)
return format_categories(...)
end
end
 
local function language_anchor(...)
local function get_script(...)
language_anchor = require(anchors_module).language_anchor
get_script = require(scripts_module).getByCode
return language_anchor(...)
return get_script(...)
end
end
 
local function munge_text(...)
local function language_anchor(...)
munge_text = require(munge_text_module)
language_anchor = require(anchors_module).language_anchor
return munge_text(...)
return language_anchor(...)
end
end
 
local function process_params(...)
local function munge_text(...)
process_params = require(parameters_module).process
munge_text = require(munge_text_module)
return process_params(...)
return munge_text(...)
end
end
 
local function u(...)
local function process_params(...)
u = require(string_utilities_module).char
process_params = require(parameters_module).process
return u(...)
return process_params(...)
end
end
 
local function ugsub(...)
local function track(...)
ugsub = require(string_utilities_module).gsub
track = require(debug_track_module)
return ugsub(...)
return track(...)
end
end
 
local function umatch(...)
local function u(...)
umatch = require(string_utilities_module).match
u = require(string_utilities_module).char
return umatch(...)
return u(...)
end
end
 
local function ugsub(...)
ugsub = require(string_utilities_module).gsub
return ugsub(...)
end
 
local function umatch(...)
umatch = require(string_utilities_module).match
return umatch(...)
end


--[==[
--[==[
Loaders for objects, which load data (or some other object) into some variable, which can then be accessed as "foo or get_foo()", where the function get_foo sets the object to "foo" and then returns it. This ensures they are only loaded when needed, and avoids the need to check for the existence of the object each time, since once "foo" has been set, "get_foo" will not be called again.]==]
Loaders for objects, which load data (or some other object) into some variable, which can then be accessed as "foo or get_foo()", where the function get_foo sets the object to "foo" and then returns it. This ensures they are only loaded when needed, and avoids the need to check for the existence of the object each time, since once "foo" has been set, "get_foo" will not be called again.]==]
local m_data
local m_data
local function get_data()
local function get_data()
m_data, get_data = mw.loadData("Module:script utilities/data"), nil
m_data, get_data = mw.loadData("Module:script utilities/data"), nil
return m_data
return m_data
end
end


--[=[
--[=[
Line 78: Line 90:
[[Module:parameters]]
[[Module:parameters]]
[[Module:utilities]]
[[Module:utilities]]
[[Module:debug/track]]
]=]
]=]


Line 107: Line 120:
if sc then
if sc then
-- Track uses of sc parameter.
if sc:getCode() == lang:findBestScript(text):getCode() then
insert(cats, lang:getFullName() .. " terms with redundant script codes")
else
insert(cats, lang:getFullName() .. " terms with non-redundant manual script codes")
end
else
else
sc = lang:findBestScript(text)
sc = lang:findBestScript(text)
Line 124: Line 143:
return export.tag_text(text, lang, sc, face, class) .. cats
return export.tag_text(text, lang, sc, face, class) .. cats
end
end
end
-- Ustring turns on the codepoint-aware string matching. The basic string function
-- should be used for simple sequences of characters, Ustring function for
-- sets – [].
local function trackPattern(text, pattern, tracking)
if pattern and umatch(text, pattern) then
track("script/" .. tracking)
end
end
local function track_text(text, lang, sc)
if lang and text then
local langCode = lang:getFullCode()
-- [[Special:WhatLinksHere/Wiktionary:Tracking/script/ang/acute]]
if langCode == "ang" then
local decomposed = toNFD(text)
local acute = u(0x301)
trackPattern(decomposed, acute, "ang/acute")
--[=[
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Greek/wrong-phi]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Greek/wrong-theta]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Greek/wrong-kappa]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Greek/wrong-rho]]
ϑ, ϰ, ϱ, ϕ should generally be replaced with θ, κ, ρ, φ.
]=]
elseif langCode == "el" or langCode == "grc" then
trackPattern(text, "ϑ", "Greek/wrong-theta")
trackPattern(text, "ϰ", "Greek/wrong-kappa")
trackPattern(text, "ϱ", "Greek/wrong-rho")
trackPattern(text, "ϕ", "Greek/wrong-phi")
--[=[
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Ancient Greek/spacing-coronis]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Ancient Greek/spacing-smooth-breathing]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Ancient Greek/wrong-apostrophe]]
When spacing coronis and spacing smooth breathing are used as apostrophes,
they should be replaced with right single quotation marks (’).
]=]
if langCode == "grc" then
trackPattern(text, u(0x1FBD), "Ancient Greek/spacing-coronis")
trackPattern(text, u(0x1FBF), "Ancient Greek/spacing-smooth-breathing")
trackPattern(text, "[" .. u(0x1FBD) .. u(0x1FBF) .. "]", "Ancient Greek/wrong-apostrophe", true)
end
-- [[Special:WhatLinksHere/Wiktionary:Tracking/script/Russian/grave-accent]]
elseif langCode == "ru" then
local decomposed = toNFD(text)
trackPattern(decomposed, u(0x300), "Russian/grave-accent")
-- [[Special:WhatLinksHere/Wiktionary:Tracking/script/Chuvash/latin-homoglyph]]
elseif langCode == "cv" then
trackPattern(text, "[ĂăĔĕÇçŸÿ]", "Chuvash/latin-homoglyph")
-- [[Special:WhatLinksHere/Wiktionary:Tracking/script/Tibetan/trailing-punctuation]]
elseif langCode == "bo" then
trackPattern(text, "[་།]$", "Tibetan/trailing-punctuation")
trackPattern(text, "[་།]%]%]$", "Tibetan/trailing-punctuation")
--[=[
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Thai/broken-ae]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Thai/broken-am]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Thai/wrong-rue-lue]]
]=]
elseif langCode == "th" then
trackPattern(text, "เ".."เ", "Thai/broken-ae")
trackPattern(text, "ํ[่้๊๋]?า", "Thai/broken-am")
trackPattern(text, "[ฤฦ]า", "Thai/wrong-rue-lue")
--[=[
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lao/broken-ae]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lao/broken-am]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lao/possible-broken-ho-no]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lao/possible-broken-ho-mo]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lao/possible-broken-ho-lo]]
]=]
elseif langCode == "lo" then
trackPattern(text, "ເ".."ເ", "Lao/broken-ae")
trackPattern(text, "ໍ[່້໊໋]?າ", "Lao/broken-am")
trackPattern(text, "ຫນ", "Lao/possible-broken-ho-no")
trackPattern(text, "ຫມ", "Lao/possible-broken-ho-mo")
trackPattern(text, "ຫລ", "Lao/possible-broken-ho-lo")
--[=[
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lü/broken-ae]]
[[Special:WhatLinksHere/Wiktionary:Tracking/script/Lü/possible-wrong-sequence]]
]=]
elseif langCode == "khb" then
trackPattern(text, "ᦵ".."ᦵ", "Lü/broken-ae")
trackPattern(text, "[ᦀ-ᦫ][ᦵᦶᦷᦺ]", "Lü/possible-wrong-sequence")
end
end
end
local function Kore_ruby(...)
-- Cache character sets on the first call.
local Hang_chars = get_script("Hang"):getCharacters()
local Hani_chars = get_script("Hani"):getCharacters()
-- Overwrite with the actual function, which is called directly on subsequent calls.
function Kore_ruby(txt)
return (ugsub(txt, "([%-".. Hani_chars .. "]+)%(([%-" .. Hang_chars .. "]+)%)", "<ruby>%1<rp>(</rp><rt>%2</rt><rp>)</rp></ruby>"))
end
return Kore_ruby(...)
end
end


Line 141: Line 270:
function export.tag_text(text, lang, sc, face, class, id)
function export.tag_text(text, lang, sc, face, class, id)
if not sc then
if not sc then
sc = lang:findBestScript(text)
if lang then
sc = lang:findBestScript(text)
else
sc = find_best_script_without_lang(text)
end
end
end
 
track_text(text, lang, sc)
 
-- Replace space characters with newlines in Mongolian-script text, which is written top-to-bottom.
-- Replace space characters with newlines in Mongolian-script text, which is written top-to-bottom.
if sc:getDirection():match("vertical") and text:find(" ") then
if sc:getDirection():find("vertical", nil, true) and text:find(" ", nil, true) then
text = munge_text(text, function(txt)
text = munge_text(text, function(txt)
-- having extra parentheses makes sure only the first return value gets through
-- having extra parentheses makes sure only the first return value gets through
Line 157: Line 292:
-- language needing such processing.
-- language needing such processing.
-- 20220221: Also convert 漢字(한자) to ruby, instead of needing [[Template:Ruby]].
-- 20220221: Also convert 漢字(한자) to ruby, instead of needing [[Template:Ruby]].
if sc:getCode() == "Kore" and (text:find("-", 1, true) or text:find("[()]")) then
if sc:getCode() == "Kore" and text:match("[%-()g]") then
text = munge_text(text, function(txt)
local title, display = require("Module:links").get_wikilink_parts(text, true)
txt = txt:gsub("%-(%-?)", "%1")
if title ~= nil then -- special case that the text is a single link, do not munge and preserve affix hyphens
txt = ugsub(txt, "([%-".. get_script("Hani"):getCharacters() .. "]+)%(([%-" .. get_script("Hang"):getCharacters() .. "]+)%)", "<ruby>%1<rp>(</rp><rt>%2</rt><rp>)</rp></ruby>")
if lang and lang:getCode() == "okm" then -- Middle Korean code from [[User:Chom.kwoy]]
return txt
-- Comment from [[User:Lunabunn]]:
end)
-- In Middle Korean orthography, syllable formation is phonemic as opposed to morpheme-boundary-based a la
-- modern Korean. As such, for example, if you were to write nam-i, it would be rendered as na.mi so if you
-- then put na-mi to indicate particle boundaries as in modern Korean, the hyphen would be misplaced.
-- Previously, this was alleviated by specialcasing na--mi but [[User:Theknightwho]] made that resolve to -
-- in the Hangul (previously we used to just delete all -s in Hangul processing), so it broke.
-- [[User:Chom.kwoy]] implemented a different solution, which is writing -> instead using however many >s to
-- shift the hyphen by that number of letters in the romanization.
-- By the time we are called, > signs have been converted to &gt; by a call to encode_entities() in
-- make_link() in [[Module:links]] (near the bottom of the function).
-- 'g' in Middle Korean is a special sign to treat the following ㅇ sign as /G/ instead of null.
display = display:gsub("&gt;", ""):gsub("g", "")
end
if display:find("<") then
display = munge_text(display, function(txt)
txt = txt:gsub("(.)%-(%-?)(.)", "%1%2%3")
return Kore_ruby(txt)
end)
else
display = display:gsub("(.)%-(%-?)(.)", "%1%2%3")
display = Kore_ruby(display)
end
text = "[[" .. title .. "|" .. display .. "]]"
else
text = munge_text(text, function(txt)
if lang and lang:getCode() == "okm" then
txt = txt:gsub("&gt;", ""):gsub("g", "")
end
if txt == text then -- special case for the entire text being plain
txt = txt:gsub("(.)%-(%-?)(.)", "%1%2%3")
else
txt = txt:gsub("%-(%-?)", "%1")
end
return Kore_ruby(txt)
end)
end
end
end
 
if sc:getCode() == "Image" then
if sc:getCode() == "Image" then
face = nil
face = nil
end
end


local function class_attr(classes)
if face == "hypothetical" then
-- if the script code is hyphenated (i.e. language code-script code, add the last component as a class as well)
-- [[Special:WhatLinksHere/Wiktionary:Tracking/script-utilities/face/hypothetical]]
-- e.g. ota-Arab adds both Arab and ota-Arab as classes
track("script-utilities/face/hypothetical")
if sc:getCode():find("-", 1, true) then
end
insert(classes, 1, (ugsub(sc:getCode(), ".+%-", "")))
 
insert(classes, 2, sc:getCode())
local data = (m_data or get_data()).faces[face or "plain"]
else
if data == nil then
insert(classes, 1, sc:getCode())
error('Invalid script face "' .. face .. '".')
end
if class and class ~= '' then
insert(classes, class)
end
return 'class="' .. concat(classes, ' ') .. '"'
end
end
 
local function tag_attr(...)
local tag = data.tag
local output = {}
local opening_tag = {tag}
if id then
 
insert(output, 'id="' .. language_anchor(lang, id) .. '"')
if lang and id then
end
insert(opening_tag, 'id="' .. language_anchor(lang, id) .. '"')
insert(output, class_attr({...}) )
if lang then
-- FIXME: Is it OK to insert the etymology-only lang code and have it fall back to the first part of the
-- lang code (by chopping off the '-...' part)? It seems the :lang() selector does this; not sure about
-- [lang=...] attributes.
insert(output, 'lang="' .. lang:getFullCode() .. '"')
end
return concat(output, " ")
end
end
 
local data = (m_data or get_data()).faces[face or "plain"]
local classes = {data.class}
-- if the script code is hyphenated (i.e. language code-script code, add the last component as a class as well)
-- Add a script wrapper
-- e.g. ota-Arab adds both Arab and ota-Arab as classes
if data then
if sc:getCode():find("-", nil, true) then
return ( data.prefix or "" ) .. '<' .. data.tag .. ' ' .. tag_attr(data.class) .. '>' .. text .. '</' .. data.tag .. '>'
insert(classes, 1, (ugsub(sc:getCode(), ".+%-", "")))
insert(classes, 2, sc:getCode())
else
else
error('Invalid script face "' .. face .. '".')
insert(classes, 1, sc:getCode())
end
if class and class ~= '' then
insert(classes, class)
end
insert(opening_tag, 'class="' .. concat(classes, ' ') .. '"')
 
-- FIXME: Is it OK to insert the etymology-only lang code and have it fall back to the first part of the
-- lang code (by chopping off the '-...' part)? It seems the :lang() selector does this; not sure about
-- [lang=...] attributes.
if lang then
insert(opening_tag, 'lang="' .. lang:getFullCode() .. '"')
end
end
-- Add a script wrapper
return (data.prefix or "") .. "<" .. concat(opening_tag, " ") .. ">" .. text .. "</" .. tag .. ">"
end
end


Line 229: Line 398:
or error("Second argument to tag_translit should be a language code or language object.")
or error("Second argument to tag_translit should be a language code or language object.")
end
end
 
local data = (m_data or get_data()).translit[kind or "default"]
local data = (m_data or get_data()).translit[kind or "default"]
 
local opening_tag = {}
local tag = data.tag
local opening_tag = {tag}
insert(opening_tag, data.tag)
 
local class = data.class
if lang == "ja" then
if lang == "ja" then
insert(opening_tag, 'class="' .. (data.classes and data.classes .. " " or "") .. (is_manual and "manual-tr " or "") .. 'tr"')
insert(opening_tag, 'class="' .. (class and (class .. " ") or "") .. (is_manual and "manual-tr " or "") .. 'tr"')
else
else
insert(opening_tag, 'lang="' .. lang .. '-Latn"')
insert(opening_tag, 'lang="' .. lang .. '-Latn"')
insert(opening_tag, 'class="' .. (data.classes and data.classes .. " " or "") .. (is_manual and "manual-tr " or "") .. 'tr Latn"')
insert(opening_tag, 'class="' .. (class and (class .. " ") or "") .. (is_manual and "manual-tr " or "") .. 'tr Latn"')
end
end
 
if data.dir then
local dir = data.dir
insert(opening_tag, 'dir="' .. data.dir .. '"')
if dir then
insert(opening_tag, 'dir="' .. dir .. '"')
end
 
if attributes then
track("tag_translit/attributes")
insert(opening_tag, attributes)
end
end
 
insert(opening_tag, attributes)
return "<" .. concat(opening_tag, " ") .. ">" .. translit .. "</" .. tag .. ">"
return "<" .. concat(opening_tag, " ") .. ">" .. translit .. "</" .. data.tag .. ">"
end
end


Line 257: Line 431:
or error("Second argument to tag_transcription should be a language code or language object.")
or error("Second argument to tag_transcription should be a language code or language object.")
end
end
 
local data = (m_data or get_data()).transcription[kind or "default"]
local data = (m_data or get_data()).transcription[kind or "default"]
 
local opening_tag = {}
local tag = data.tag
local opening_tag = {tag}
insert(opening_tag, data.tag)
 
local class = data.class
if lang == "ja" then
if lang == "ja" then
insert(opening_tag, 'class="' .. (data.classes and data.classes .. " " or "") .. 'ts"')
insert(opening_tag, 'class="' .. (class and (class .. " ") or "") .. 'ts"')
else
else
insert(opening_tag, 'lang="' .. lang .. '-Latn"')
insert(opening_tag, 'lang="' .. lang .. '-Latn"')
insert(opening_tag, 'class="' .. (data.classes and data.classes .. " " or "") .. 'ts Latn"')
insert(opening_tag, 'class="' .. (class and (class .. " ") or "") .. 'ts Latn"')
end
 
local dir = data.dir
if dir then
insert(opening_tag, 'dir="' .. dir .. '"')
end
 
if attributes then
track("tag_transcription/attributes")
insert(opening_tag, attributes)
end
 
return "<" .. concat(opening_tag, " ") .. ">" .. transcription .. "</" .. tag .. ">"
end
 
--[==[Tags {def} as a definition.
The <code>def</code> parameter must be one of the following:
; {{code|lua|"gloss"}}
: The text is wrapped in {{code|html|2=<span class="(mention-gloss">...</span>}}.
; {{code|lua|"non-gloss"}}
: The text is wrapped in {{code|html|2=<span class="use-with-mention">...</span>}}.
The optional <code>attributes</code> parameter is used to specify additional HTML attributes for the tag.]==]
function export.tag_definition(def, kind, attributes)
local data = (m_data or get_data()).definition[kind]
if data == nil then
error("Second argument to tag_definition should specify the kind of definition from the list in [[Module:script utilities/data]].")
end
 
local tag = data.tag
local opening_tag = {tag}
 
local class = data.class
if class then
insert(opening_tag, 'class="' .. class .. '"')
end
end
 
if data.dir then
if attributes then
insert(opening_tag, 'dir="' .. data.dir .. '"')
insert(opening_tag, attributes)
end
end
 
insert(opening_tag, attributes)
return "<" .. concat(opening_tag, " ") .. ">" .. def .. "</" .. tag .. ">"
return "<" .. concat(opening_tag, " ") .. ">" .. transcription .. "</" .. data.tag .. ">"
end
end


Line 323: Line 530:
-- If there are no non-Latin scripts, return nothing.
-- If there are no non-Latin scripts, return nothing.
if not has_nonlatin then
if not has_nonlatin and lang:getCode() ~= "und" then
return ""
return ""
end
end
Line 338: Line 545:


--[==[This is used by {{temp|rfscript}}. See there for more information.]==]
--[==[This is used by {{temp|rfscript}}. See there for more information.]==]
do
function export.template_rfscript(frame)
local function get_args(frame)
local boolean = {type = "boolean"}
local boolean = {type = "boolean"}
local args = process_params(frame:getParent().args, {
return process_params(frame:getParent().args, {
[1] = {required = true, type = "language", default = "und"},
[1] = {required = true, type = "language", default = "und"},
["sc"] = {type = "script"},
["sc"] = {type = "script"},
["usex"] = boolean,
["usex"] = boolean,
["quote"] = boolean,
["quote"] = boolean,
["nocat"] = boolean,
["nocat"] = boolean,
["sort"] = true,
["sort"] = true,
})
})
end
local ret = export.request_script(args[1], args["sc"], args.quote and "quote" or args.usex, args.nocat, args.sort)
function export.template_rfscript(frame)
if ret == "" then
local args = get_args(frame)
error("This language is written in the Latin alphabet. It does not need a native script.")
local ret = export.request_script(args[1], args["sc"], args.quote and "quote" or args.usex, args.nocat, args.sort)
if ret == "" then
error("This language is written in the Latin alphabet. It does not need a native script.")
else
return ret
end
end
end
return ret
end
end


Retrieved from "https://linguifex.com/wiki/Module:script_utilities"