Module:Delink/sandbox - Revision history

Lemondoge: Synced (rip one testcase but whatever), dramatically improved efficiency of one of the regex patterns

2023-04-22T18:29:13Z

Synced (rip one testcase but whatever), dramatically improved efficiency of one of the regex patterns

← Previous revision		Revision as of 18:29, 22 April 2023
Line 29:		Line 29:
	s = s:match("(.-) ?%(.-%)$")		s = s:match("(.-) ?%(.-%)$")
	elseif s:match(",") then -- If there are no brackets, display only the text before the first comma.		elseif s:match(",") then -- If there are no brackets, display only the text before the first comma.
	s = s:match("~~(.-)~~,.*$")		s = s:match("[^,]*")
	end		end
	return s		return s
Line 94:		Line 94:

	-- Remove the colon if the link is using the [[Help:Colon trick]].		-- Remove the colon if the link is using the [[Help:Colon trick]].
	-- if result:match("%[%[:") then		if result:match("%[%[:") then
	-- result = "[[" .. result:match("%[%[:(.*%]%])")		result = "[[" .. result:match("%[%[:(.*%]%])")
⚫	-- end
	if result:sub(3,3) == ':' then
⚫	result = ~~"[["~~ .. ~~result:~~sub(4,-1)
	end		end

Line 159:		Line 156:
	error("Attempt to de-link non-string input.", 2)		error("Attempt to de-link non-string input.", 2)
	end		end
		⚫	if type(pattern) ~= "string" or mw.ustring.sub(pattern, 1, 1) ~= "^" then
	-- unneccessary check
		⚫	error('Invalid pattern detected. Patterns must begin with "^".', 2)
⚫	-- if type(pattern) ~= "string" or ~~string~~.sub(pattern, 1, 1) ~= "^" then
		⚫	end
⚫	-- error('Invalid pattern detected. Patterns must begin with "^".', 2)
⚫	-- end
	-- Iterate over the text string, and replace any matched text. using the		-- Iterate over the text string, and replace any matched text. using the
	-- delink function.		-- delink function. We need to iterate character by character rather
			-- than just use gsub, otherwise nested links aren't detected properly.
	local result = ""		local result = ""
		⚫	while text ~= "" do
	local sb = text:find('%[')
	local n
⚫	while sb do
⚫	~~result~~ = ~~result ..~~ mw.ustring.sub(text, 1, sb - 1)
	text = mw.ustring.sub(text, sb, -1)
	-- Replace text using one iteration of gsub.		-- Replace text using one iteration of gsub.
	text~~, n~~ = mw.ustring.gsub(text, pattern, delinkFunction, 1)		text = mw.ustring.gsub(text, pattern, delinkFunction, 1)
			-- Append the left-most character to the result string.
	sb = text:find('%[', 2)
		⚫	result = result .. mw.ustring.sub(text, 1, 1)
		⚫	text = mw.ustring.sub(text, 2, -1)
	end		end
	result = result .. text
	return result		return result
	end		end
Line 190:		Line 184:
	end		end

			if args.wikilinks ~= "no" and args.wikilinks ~= "target" then
	if text:match("%[") then -- does the text contain a square bracket?
	if ~~args.~~wikilinks ~= ~~"no"~~ ~~and~~ ~~args.wikilinks~~ ~= ~~"target"~~ ~~then~~		-- De-link wikilinks and return the label portion of the wikilink.
		⚫	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel)
	-- De-link wikilinks and return the label portion of the wikilink.
		⚫	elseif args.wikilinks == "target" then
⚫	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel)
	~~elseif~~ ~~args.~~wikilinks == "target" ~~then~~		-- De-link wikilinks and return the target portions of the wikilink.
		⚫	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget)
	-- De-link wikilinks and return the target portions of the wikilink.
		⚫	end
⚫	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget)
			if args.urls ~= "no" then
	end
		⚫	text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.
⚫	if args.~~urls~~ ~= "no" then
⚫	text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.
	end
	end		end

	if args.whitespace ~= "no" then		if args.whitespace ~= "no" then
	-- Replace single new lines with a single space, but leave double new lines		-- Replace single new lines with a single space, but leave double new lines

WOSlinker: use require("strict") instead of require("Module:No globals")

2022-10-21T09:51:51Z

use require("strict") instead of require("Module:No globals")

← Previous revision		Revision as of 09:51, 21 October 2022
Line 1:		Line 1:
	-- This module de-links most wikitext.		-- This module de-links most wikitext.

	require("~~Module:No globals~~")		require("strict")

	local p = {}		local p = {}

Desb42: performance enhancement

2022-04-30T06:37:26Z

performance enhancement

← Previous revision		Revision as of 06:37, 30 April 2022
Line 94:		Line 94:

	-- Remove the colon if the link is using the [[Help:Colon trick]].		-- Remove the colon if the link is using the [[Help:Colon trick]].
	if result:match("%[%[:") then		-- if result:match("%[%[:") then
	result = "[[" .. result:match("%[%[:(.*%]%])")		-- result = "[[" .. result:match("%[%[:(.*%]%])")
		⚫	-- end
			if result:sub(3,3) == ':' then
		⚫	result = "[[" .. result:sub(4,-1)
	end		end

Line 156:		Line 159:
	error("Attempt to de-link non-string input.", 2)		error("Attempt to de-link non-string input.", 2)
	end		end
			-- unneccessary check
	if type(pattern) ~= "string" or ~~mw.ustring~~.sub(pattern, 1, 1) ~= "^" then		-- if type(pattern) ~= "string" or string.sub(pattern, 1, 1) ~= "^" then
	error('Invalid pattern detected. Patterns must begin with "^".', 2)		-- error('Invalid pattern detected. Patterns must begin with "^".', 2)
⚫	end
			-- end
	-- Iterate over the text string, and replace any matched text. using the		-- Iterate over the text string, and replace any matched text. using the
	-- delink function. ~~We need to iterate character by character rather~~		-- delink function.
	-- than just use gsub, otherwise nested links aren't detected properly.
	local result = ""		local result = ""
			local sb = text:find('%[')
⚫	while ~~text ~= ""~~ do
			local n
		⚫	while sb do
		⚫	result = result .. mw.ustring.sub(text, 1, sb - 1)
			text = mw.ustring.sub(text, sb, -1)
	-- Replace text using one iteration of gsub.		-- Replace text using one iteration of gsub.
	text = mw.ustring.gsub(text, pattern, delinkFunction, 1)		text, n = mw.ustring.gsub(text, pattern, delinkFunction, 1)
			sb = text:find('%[', 2)
	-- Append the left-most character to the result string.
⚫	result = ~~result~~ .. ~~mw.ustring.~~sub(~~text~~, 1, 1)
⚫	~~text~~ = mw.ustring.sub(text, 2, -1)
	end		end
			result = result .. text
	return result		return result
	end		end
Line 184:		Line 190:
	end		end

			if text:match("%[") then -- does the text contain a square bracket?
	if args.wikilinks ~= "no" and args.wikilinks ~= "target" then
	~~-- De-link~~ wikilinks ~~and~~ ~~return~~ ~~the~~ ~~label~~ ~~portion~~ of ~~the wikilink.~~		if args.wikilinks ~= "no" and args.wikilinks ~= "target" then
			-- De-link wikilinks and return the label portion of the wikilink.
	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel)		text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel)
	elseif args.wikilinks == "target" then
	-- De-link wikilinks and return the target portions of the wikilink.		elseif args.wikilinks == "target" then
			-- De-link wikilinks and return the target portions of the wikilink.
	text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget)		text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget)
	end		end
	if args.urls ~= "no" then		if args.urls ~= "no" then
	text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.		text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.
			end
	end		end

	if args.whitespace ~= "no" then		if args.whitespace ~= "no" then
	-- Replace single new lines with a single space, but leave double new lines		-- Replace single new lines with a single space, but leave double new lines

Desb42: Undid revision 1085268485 by Desb42 (talk)

2022-04-29T13:52:38Z

Undid revision 1085268485 by Desb42 (talk)

← Previous revision		Revision as of 13:52, 29 April 2022
Line 156:		Line 156:
	error("Attempt to de-link non-string input.", 2)		error("Attempt to de-link non-string input.", 2)
	end		end
		⚫	if type(pattern) ~= "string" or mw.ustring.sub(pattern, 1, 1) ~= "^" then
	-- unneccessary check
		⚫	error('Invalid pattern detected. Patterns must begin with "^".', 2)
⚫	-- if type(pattern) ~= "string" or ~~string~~.sub(pattern, 1, 1) ~= "^" then
		⚫	end
⚫	-- error('Invalid pattern detected. Patterns must begin with "^".', 2)
⚫	-- end
	-- Iterate over the text string, and replace any matched text. using the		-- Iterate over the text string, and replace any matched text. using the
	-- delink function.		-- delink function. We need to iterate character by character rather
			-- than just use gsub, otherwise nested links aren't detected properly.
	local result = ""		local result = ""
		⚫	while text ~= "" do
	local sb = text:find('%[')
⚫	while sb do
⚫	result = result .. mw.ustring.sub(text, 1, ~~sb -~~ 1)
⚫	text = mw.ustring.sub(text, sb, -1)
	-- Replace text using one iteration of gsub.		-- Replace text using one iteration of gsub.
	text = mw.ustring.gsub(text, pattern, delinkFunction, 1)		text = mw.ustring.gsub(text, pattern, delinkFunction, 1)
			-- Append the left-most character to the result string.
	sb = text:find('%[')
		⚫	result = result .. mw.ustring.sub(text, 1, 1)
		⚫	text = mw.ustring.sub(text, 2, -1)
	end		end
	result = result .. text
	return result		return result
	end		end

Desb42: Undid revision 1085270311 by Desb42 (talk)

2022-04-29T13:43:13Z

Undid revision 1085270311 by Desb42 (talk)

← Previous revision		Revision as of 13:43, 29 April 2022
Line 165:		Line 165:
	local sb = text:find('%[')		local sb = text:find('%[')
	while sb do		while sb do
	mw.log(sb)
	result = result .. mw.ustring.sub(text, 1, sb - 1)		result = result .. mw.ustring.sub(text, 1, sb - 1)
	text = mw.ustring.sub(text, sb, -1)		text = mw.ustring.sub(text, sb, -1)

Desb42 at 13:42, 29 April 2022

2022-04-29T13:42:05Z

← Previous revision		Revision as of 13:42, 29 April 2022
Line 165:		Line 165:
	local sb = text:find('%[')		local sb = text:find('%[')
	while sb do		while sb do
			mw.log(sb)
	result = result .. mw.ustring.sub(text, 1, sb - 1)		result = result .. mw.ustring.sub(text, 1, sb - 1)
	text = mw.ustring.sub(text, sb, -1)		text = mw.ustring.sub(text, sb, -1)

Desb42: performance enhancement

2022-04-29T13:25:00Z

performance enhancement

← Previous revision		Revision as of 13:25, 29 April 2022
Line 156:		Line 156:
	error("Attempt to de-link non-string input.", 2)		error("Attempt to de-link non-string input.", 2)
	end		end
			-- unneccessary check
	if type(pattern) ~= "string" or ~~mw.ustring~~.sub(pattern, 1, 1) ~= "^" then		-- if type(pattern) ~= "string" or string.sub(pattern, 1, 1) ~= "^" then
	error('Invalid pattern detected. Patterns must begin with "^".', 2)		-- error('Invalid pattern detected. Patterns must begin with "^".', 2)
	end		-- end
	-- Iterate over the text string, and replace any matched text. using the		-- Iterate over the text string, and replace any matched text. using the
	-- delink function. ~~We need to iterate character by character rather~~		-- delink function.
	-- than just use gsub, otherwise nested links aren't detected properly.
	local result = ""		local result = ""
			local sb = text:find('%[')
	while ~~text ~= ""~~ do		while sb do
		⚫	result = result .. mw.ustring.sub(text, 1, sb - 1)
		⚫	text = mw.ustring.sub(text, sb, -1)
	-- Replace text using one iteration of gsub.		-- Replace text using one iteration of gsub.
	text = mw.ustring.gsub(text, pattern, delinkFunction, 1)		text = mw.ustring.gsub(text, pattern, delinkFunction, 1)
			sb = text:find('%[')
	-- Append the left-most character to the result string.
⚫	result = result .. mw.ustring.sub(text, 1, 1)
⚫	text = mw.ustring.sub(text, 2, -1)
	end		end
			result = result .. text
	return result		return result
	end		end

Gonnym: fix for question mark

2021-09-30T12:32:48Z

fix for question mark

← Previous revision		Revision as of 12:32, 30 September 2021
Line 53:		Line 53:
	target_area = result:match("^%[%[(.-)%]%]")		target_area = result:match("^%[%[(.-)%]%]")
	end		end

	-- Check for bad characters.		-- Check for bad characters.
	if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") then		if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") and mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") ~= "?" then
	return s		return s
	end		end
Line 67:		Line 68:
	return delinkReversePipeTrick(result)		return delinkReversePipeTrick(result)
	end		end


	result = mw.uri.decode(result, "PATH") -- decode percent-encoded entities. Leave underscores and plus signs.		result = mw.uri.decode(result, "PATH") -- decode percent-encoded entities. Leave underscores and plus signs.
	result = mw.text.decode(result, true) -- decode HTML entities.		result = mw.text.decode(result, true) -- decode HTML entities.


	-- Check for bad titles. To do this we need to find the		-- Check for bad titles. To do this we need to find the
	-- title area of the link, i.e. the part before any pipes.		-- title area of the link, i.e. the part before any pipes.
Line 79:		Line 80:
	target_area = result:match("^%[%[(.-)%]%]")		target_area = result:match("^%[%[(.-)%]%]")
	end		end

	-- Check for bad characters.		-- Check for bad characters.
	if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") then		if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") and mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") ~= "?" then
	return s		return s
	end		end
Line 90:		Line 92:
	return ""		return ""
	end		end


	-- Remove the colon if the link is using the [[Help:Colon trick]].		-- Remove the colon if the link is using the [[Help:Colon trick]].
	if result:match("%[%[:") then		if result:match("%[%[:") then
	result = "[[" .. result:match("%[%[:(.*%]%])")		result = "[[" .. result:match("%[%[:(.*%]%])")
	end		end


	-- Deal with links using the [[Help:Pipe trick]].		-- Deal with links using the [[Help:Pipe trick]].
	if mw.ustring.match(result, "^%[%[[^\|]*\|%]%]") then		if mw.ustring.match(result, "^%[%[[^\|]*\|%]%]") then
Line 181:		Line 183:
	text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments.		text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments.
	end		end

	if args.wikilinks ~= "no" and args.wikilinks ~= "target" then		if args.wikilinks ~= "no" and args.wikilinks ~= "target" then
	-- De-link wikilinks and return the label portion of the wikilink.		-- De-link wikilinks and return the label portion of the wikilink.

Gonnym: cleaner code - if not (x == y) -> if x ~= y, etc

2021-09-30T12:00:23Z

cleaner code - if not (x == y) -> if x ~= y, etc

← Previous revision		Revision as of 12:00, 30 September 2021
Line 1:		Line 1:
	-- This module de-links most wikitext.		-- This module de-links most wikitext.

	require('Module:No globals')		require("Module:No globals")

	local p = {}		local p = {}
Line 10:		Line 10:
	if s:match("^%[%[\|.*[\|\n]") then -- Check for newlines or multiple pipes.		if s:match("^%[%[\|.*[\|\n]") then -- Check for newlines or multiple pipes.
	return s		return s
	else
⚫	return s:match("%[%[\|(.*)%]%]")
	end		end
		⚫	return s:match("%[%[\|(.*)%]%]")

	end		end

	local function delinkPipeTrick(s)		local function delinkPipeTrick(s)
	local linkarea, display = "", ""
	-- We need to deal with colons, brackets, and commas, per [[Help:Pipe trick]].		-- We need to deal with colons, brackets, and commas, per [[Help:Pipe trick]].

	-- First, remove the text before the first colon, if any.		-- First, remove the text before the first colon, if any.
	if s:match(":") then		if s:match(":") then
Line 36:		Line 34:
	end		end

	~~local function delinkWikilink_tgt(s)~~ -- ~~return~~ wikilink target \|wikilinks=target		-- Return wikilink target \|wikilinks=target
			local function getDelinkedTarget(s)
	local result = s		local result = s
	-- Deal with the reverse pipe trick.		-- Deal with the reverse pipe trick.
Line 62:		Line 61:
	end		end

	local function ~~delinkWikilink~~(s)		local function getDelinkedLabel(s)
	local result = s		local result = s
	-- Deal with the reverse pipe trick.		-- Deal with the reverse pipe trick.
Line 86:		Line 85:

	-- Check for categories, interwikis, and files.		-- Check for categories, interwikis, and files.
	local ~~colonprefix~~ = result:match("%[%[(.-):.*%]%]") or "" -- Get the text before the first colon.		local colon_prefix = result:match("%[%[(.-):.*%]%]") or "" -- Get the text before the first colon.
	local ns = mw.site.namespaces[~~colonprefix~~] -- see if this is a known namespace		local ns = mw.site.namespaces[colon_prefix] -- see if this is a known namespace
	if mw.language.isKnownLanguageTag(~~colonprefix~~)		if mw.language.isKnownLanguageTag(colon_prefix) or (ns and (ns.canonicalName == "File" or ns.canonicalName == "Category")) then
	or ( ns and ( ns.canonicalName == "File" or ns.canonicalName == "Category" ) ) then
	return ""		return ""
	end		end
Line 128:		Line 126:
	local valid_url_prefixes = {"//", "http://", "https://", "ftp://", "gopher://", "mailto:", "news:", "irc://"}		local valid_url_prefixes = {"//", "http://", "https://", "ftp://", "gopher://", "mailto:", "news:", "irc://"}
	local url_prefix		local url_prefix
	for i,v in ipairs(valid_url_prefixes) do		for _ ,v in ipairs(valid_url_prefixes) do
	if mw.ustring.match(s, '^%[' .. v ..'[^"%s].*%]' ) then		if mw.ustring.match(s, '^%[' .. v ..'[^"%s].*%]' ) then
	url_prefix = v		url_prefix = v
Line 146:		Line 144:
	if mw.ustring.match(s_decoded, "%c") then		if mw.ustring.match(s_decoded, "%c") then
	return s		return s
	else
⚫	return s_decoded
	end		end

		⚫	return s_decoded

	end		end

	local function delinkLinkClass(s, pattern, delinkFunction)		local function delinkLinkClass(text, pattern, delinkFunction)
	if ~~not~~ type(s) == "string" then		if type(text) ~= "string" then
	error("Attempt to de-link non-string input.", 2)		error("Attempt to de-link non-string input.", 2)
	end		end
	if ~~not (~~ type(pattern) == "string" ~~and~~ mw.ustring.sub(pattern, 1, 1) == "^" ) then		if type(pattern) ~= "string" or mw.ustring.sub(pattern, 1, 1) ~= "^" then
	error('Invalid pattern detected. Patterns must begin with "^".', 2)		error('Invalid pattern detected. Patterns must begin with "^".', 2)
	end		end
Line 162:		Line 161:
	-- than just use gsub, otherwise nested links aren't detected properly.		-- than just use gsub, otherwise nested links aren't detected properly.
	local result = ""		local result = ""
	while s ~= '' do		while text ~= "" do
	-- Replace text using one iteration of gsub.		-- Replace text using one iteration of gsub.
	s = mw.ustring.gsub(s, pattern, delinkFunction, 1)		text = mw.ustring.gsub(text, pattern, delinkFunction, 1)
	-- Append the left-most character to the result string.		-- Append the left-most character to the result string.
	result = result .. mw.ustring.sub(s, 1, 1)		result = result .. mw.ustring.sub(text, 1, 1)
	s = mw.ustring.sub(s, 2, -1)		text = mw.ustring.sub(text, 2, -1)
	end		end
	return result		return result
Line 179:		Line 178:
	text = mw.ustring.gsub(text, "UNIQ%w%-ref%-%d%-QINU", "")		text = mw.ustring.gsub(text, "UNIQ%w%-ref%-%d%-QINU", "")
	end		end
	if ~~not (~~args.comments == "no") then		if args.comments ~= "no" then
	text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments.		text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments.
	end		end
	if ~~not (~~args.wikilinks == "no") and ~~'target'~~ ~= ~~args.wikilinks~~ then		if args.wikilinks ~= "no" and args.wikilinks ~= "target" then
	~~text = delinkLinkClass(text, "^%[%[.-%]%]", delinkWikilink)~~ -- De-link wikilinks and return the label portion of the wikilink.		-- De-link wikilinks and return the label portion of the wikilink.
			text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel)
	elseif 'target' == args.wikilinks then
			elseif args.wikilinks == "target" then
	~~text = delinkLinkClass(text, "^%[%[.-%]%]", delinkWikilink_tgt)~~ -- De-link wikilinks and return the target portions of the wikilink		-- De-link wikilinks and return the target portions of the wikilink.
			text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget)
	end		end
	if ~~not (~~args.urls == "no") then		if args.urls ~= "no" then
	text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.		text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs.
	end		end
	if ~~not (~~args.whitespace == "no") then		if args.whitespace ~= "no" then
	-- Replace single new lines with a single space, but leave double new lines		-- Replace single new lines with a single space, but leave double new lines
	-- and new lines only containing spaces or tabs before a second new line.		-- and new lines only containing spaces or tabs before a second new line.

Gonnym: consistent spacing

2021-09-30T11:47:04Z

consistent spacing

Show changes