Modulo:Delink: differenze tra le versioni
synced from sandbox: consistent spacing; cleaner code - if not (x == y) -> if x ~= y, etc; fix for question mark - can now delink What If...? style titles.
(singolo link non necessariamente a inizio stringa come da precedente comportamento del template:Scollega) |
(synced from sandbox: consistent spacing; cleaner code - if not (x == y) -> if x ~= y, etc; fix for question mark - can now delink What If...? style titles.) |
||
Riga 1: | Riga 1: | ||
-- | -- This module de-links most wikitext. | ||
require( | require("Module:No globals") | ||
local p = {} | local p = {} | ||
local | |||
local getArgs | |||
local function delinkReversePipeTrick(s) | local function delinkReversePipeTrick(s) | ||
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 | ||
end | end | ||
return s:match("%[%[|(.*)%]%]") | |||
end | end | ||
local function delinkPipeTrick(s) | local function delinkPipeTrick(s) | ||
-- 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 | ||
Riga 30: | Riga 24: | ||
s = s:match("%[%[(.*)|%]%]") | s = s:match("%[%[(.*)|%]%]") | ||
end | end | ||
-- Next up, brackets and commas. | |||
if s:match("%(.-%)$") then -- Brackets trump commas. | |||
s = s:match("(.-) ?%(.-%)$") | |||
elseif s:match(",") then -- If there are no brackets, display only the text before the first comma. | |||
s = s:match("(.-),.*$") | |||
end | |||
return s | |||
end | |||
if | -- Return wikilink target |wikilinks=target | ||
-- | local function getDelinkedTarget(s) | ||
local result = s | |||
-- Deal with the reverse pipe trick. | |||
if result:match("%[%[|") then | |||
return delinkReversePipeTrick(result) | |||
end | |||
result = mw.uri.decode(result, "PATH") -- decode percent-encoded entities. Leave underscores and plus signs. | |||
result = mw.text.decode(result, true) -- decode HTML entities. | |||
-- Check for bad titles. To do this we need to find the | |||
-- title area of the link, i.e. the part before any pipes. | |||
local target_area | |||
if result:match("|") then -- Find if we're dealing with a piped link. | |||
target_area = result:match("^%[%[(.-)|.*%]%]") | |||
else | |||
target_area = result:match("^%[%[(.-)%]%]") | |||
end | |||
-- Check for bad characters. | |||
if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") and mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") ~= "?" then | |||
return s | |||
end | end | ||
return | |||
return target_area | |||
end | end | ||
local function | local function getDelinkedLabel(s) | ||
local result = s | local result = s | ||
-- Deal with the reverse pipe trick. | -- Deal with the reverse pipe trick. | ||
Riga 54: | Riga 74: | ||
-- 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. | ||
local | local target_area | ||
if result:match("|") then -- Find if we're dealing with a piped link. | if result:match("|") then -- Find if we're dealing with a piped link. | ||
target_area = result:match("^%[%[(.-)|.*%]%]") | |||
else | else | ||
target_area = result:match("^%[%[(.-)%]%]") | |||
end | end | ||
-- Check for bad characters. | -- Check for bad characters. | ||
if mw.ustring.match( | if mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") and mw.ustring.match(target_area, "[%[%]<>{}%%%c\n]") ~= "?" then | ||
return s | return s | ||
end | end | ||
-- Check for categories, interwikis, and files. | -- Check for categories, interwikis, and files. | ||
local | local colon_prefix = result:match("%[%[(.-):.*%]%]") or "" -- Get the text before the first colon. | ||
local ns = mw.site.namespaces[colon_prefix] -- see if this is a known namespace | |||
if mw.language.isKnownLanguageTag(colon_prefix) or (ns and (ns.canonicalName == "File" or ns.canonicalName == "Category")) then | |||
return "" | |||
end | end | ||
Riga 84: | Riga 102: | ||
return delinkPipeTrick(result) | return delinkPipeTrick(result) | ||
end | end | ||
-- Find the display area of the wikilink | -- Find the display area of the wikilink | ||
if result:match("|") then -- Find if we're dealing with a piped link. | if result:match("|") then -- Find if we're dealing with a piped link. | ||
result = result:match("^%[%[.-|(.+)%]%]") | |||
-- Remove new lines from the display of multiline piped links, | -- Remove new lines from the display of multiline piped links, | ||
-- where the pipe is before the first new line. | -- where the pipe is before the first new line. | ||
Riga 105: | Riga 119: | ||
-- Assume we have already delinked internal wikilinks, and that | -- Assume we have already delinked internal wikilinks, and that | ||
-- we have been passed some text between two square brackets [foo]. | -- we have been passed some text between two square brackets [foo]. | ||
-- If the text contains a line break it is not formatted as a URL, regardless of other content. | -- If the text contains a line break it is not formatted as a URL, regardless of other content. | ||
if s:match("\n") then | if s:match("\n") then | ||
return s | return s | ||
end | end | ||
-- Check if the text has a valid URL prefix and at least one valid URL character. | -- Check if the text has a valid URL prefix and at least one valid URL character. | ||
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 | 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 | ||
Riga 120: | Riga 134: | ||
end | end | ||
end | end | ||
-- Get display text | -- Get display text | ||
if not url_prefix then | if not url_prefix then | ||
Riga 132: | Riga 146: | ||
if mw.ustring.match(s_decoded, "%c") then | if mw.ustring.match(s_decoded, "%c") then | ||
return s | return s | ||
end | end | ||
return s_decoded | |||
end | end | ||
local function delinkLinkClass( | local function delinkLinkClass(text, pattern, delinkFunction) | ||
if | if type(text) ~= "string" then | ||
error(" | error("Attempt to de-link non-string input.", 2) | ||
end | end | ||
if | if type(pattern) ~= "string" or mw.ustring.sub(pattern, 1, 1) ~= "^" then | ||
error(' | 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 | ||
Riga 148: | Riga 163: | ||
-- 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 | while text ~= "" do | ||
-- Replace text using one iteration of gsub. | |||
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( | result = result .. mw.ustring.sub(text, 1, 1) | ||
text = mw.ustring.sub(text, 2, -1) | |||
end | end | ||
return result | return result | ||
end | end | ||
function p. | function p._delink(args) | ||
local text = args[1] or "" | local text = args[1] or "" | ||
if args.refs == "yes | if args.refs == "yes" then | ||
-- Remove any [[Help:Strip markers]] representing ref tags. In most situations | -- Remove any [[Help:Strip markers]] representing ref tags. In most situations | ||
-- this is not a good idea - only use it if you know what you are doing! | -- this is not a good idea - only use it if you know what you are doing! | ||
text = mw.ustring.gsub(text, "UNIQ%w*%-ref%-%d*%-QINU", "") | text = mw.ustring.gsub(text, "UNIQ%w*%-ref%-%d*%-QINU", "") | ||
end | end | ||
if | if args.comments ~= "no" then | ||
text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments. | text = text:gsub("<!%-%-.-%-%->", "") -- Remove html comments. | ||
end | end | ||
if | |||
text = delinkLinkClass(text, "^%[%[.-%]%]", | if args.wikilinks ~= "no" and args.wikilinks ~= "target" then | ||
-- De-link wikilinks and return the label portion of the wikilink. | |||
text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedLabel) | |||
elseif args.wikilinks == "target" then | |||
-- De-link wikilinks and return the target portions of the wikilink. | |||
text = delinkLinkClass(text, "^%[%[.-%]%]", getDelinkedTarget) | |||
end | end | ||
if | if args.urls ~= "no" then | ||
text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs. | text = delinkLinkClass(text, "^%[.-%]", delinkURL) -- De-link URLs. | ||
end | end | ||
if | 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. | ||
Riga 195: | Riga 203: | ||
end | end | ||
function p. | function p.delink(frame) | ||
if not getArgs then | |||
return p. | getArgs = require('Module:Arguments').getArgs | ||
end | |||
return p._delink(getArgs(frame, {wrappers = 'Template:Delink'})) | |||
end | end | ||
return p | return p |