Modul:Transcluder

------------------------------------------------------------
-- THIS MODULE IS DEPRECATED AND NO LONGER MAINTAINED
-- PLEASE USE Module:WikitextParser INSTEAD
-- SEE https://en.wikipedia.org/wiki/Module:WikitextParser
------------------------------------------------------------
-- Module:Transcluder is a general-purpose transclusion engine
-- Documentation and master version: https://en.wikipedia.org/wiki/Module:Transcluder
-- Authors: User:Sophivorus, User:Certes & others
-- License: CC-BY-SA-3.0
localp={}
-- Helper function to test for truthy and falsy values
-- @todo Somehow internationalize it
localfunctiontruthy(value)
ifnotvalueorvalue==''orvalue==0orvalue=='0'orvalue=='false'orvalue=='no'orvalue=='non'then
returnfalse
end
returntrue
end
-- Helper function to match from a list of regular expressions
-- Like so: match pre..list[1]..post or pre..list[2]..post or ...
localfunctionmatchAny(text,pre,list,post,init)
localmatch={}
fori=1,#listdo
match={mw.ustring.match(text,pre..list[i]..post,init)}
ifmatch[1]thenreturnunpack(match)end
end
returnnil
end
-- Like matchAny but for Category/File links with less overhead
localfunctionmatchAnyLink(text,list)
localmatch
for_,vinipairs(list)do
match=string.match(text,'%[%[%s*'..v..'%s*:.*%]%]')
ifmatchthenbreakend
end
returnmatch
end
-- Helper function to escape a string for use in regexes
localfunctionescapeString(str)
returnstring.gsub(str,'[%^%$%(%)%.%[%]%*%+%-%?%%]','%%%0')
end
-- Helper function to remove a string from a text
localfunctionremoveString(text,str)
localpattern=escapeString(str)
if#pattern>9999then-- strings longer than 10000 bytes can't be put into regexes
pattern=escapeString(mw.ustring.sub(str,1,999))..'.-'..escapeString(mw.ustring.sub(str,-999))
end
returnstring.gsub(text,pattern,'')
end
-- Helper function to convert a comma-separated list of numbers or min-max ranges into a list of booleans
-- @param flags Comma-separated list of numbers or min-max ranges, for example '1,3-5'
-- @return Map from integers to booleans, for example {1=true,2=false,3=true,4=true,5=true}
-- @return Boolean indicating whether the flags should be treated as a blacklist or not
localfunctionparseFlags(value)
localflags={}
localblacklist=false
ifnotvaluethenreturnnil,falseend
iftype(value)=='number'then
ifvalue<0then
value=-value
blacklist=true
end
flags={[value]=true}
elseiftype(value)=='string'then
ifstring.sub(value,1,1)=='-'then
blacklist=true
value=string.sub(value,2)
end
localranges=mw.text.split(value,',')-- split ranges: '1,3-5' to {'1','3-5'}
for_,rangeinpairs(ranges)do
range=mw.text.trim(range)
localmin,max=mw.ustring.match(range,'^(%d+)%s*[-–—]%s*(%d+)$')-- '3-5' to min=3 max=5
ifnotmaxthenmin,max=string.match(range,'^((%d+))$')end-- '1' to min=1 max=1
ifmaxthen
fori=min,maxdoflags[i]=trueend
else
flags[range]=true-- if we reach this point, the string had the form 'a,b,c' rather than '1,2,3'
end
end
-- List has the form { [1] = false, [2] = true, ['c'] = false }
-- Convert it to { [1] = true, [2] = true, ['c'] = true }
-- But if ANY value is set to false, treat the list as a blacklist
elseiftype(value)=='table'then
fori,vinpairs(value)do
ifv==falsethenblacklist=trueend
flags[i]=true
end
end
returnflags,blacklist
end
-- Helper function to see if a value matches any of the given flags
localfunctionmatchFlag(value,flags)
ifnotvaluethenreturnfalseend
value=tostring(value)
locallang=mw.language.getContentLanguage()
locallcvalue=lang:lcfirst(value)
localucvalue=lang:ucfirst(value)
forflaginpairs(flags)do
ifvalue==tostring(flag)
orlcvalue==flag
orucvalue==flag
or(nottonumber(flag)andmw.ustring.match(value,flag))then
returntrue
end
end
end
-- Helper function to convert template arguments into an array of options fit for get()
localfunctionparseArgs(frame)
localargs={}
forkey,valueinpairs(frame:getParent().args)doargs[key]=valueend
forkey,valueinpairs(frame.args)doargs[key]=valueend-- args from Lua calls have priority over parent args from template
returnargs
end
-- Error handling function
-- Throws a Lua error or returns an empty string if error reporting is disabled
localfunctionthrowError(key,value)
localTNT=require('Module:TNT')
localok,message=pcall(TNT.format,'I18n/Module:Transcluder.tab','error-'..key,value)
ifnotokthenmessage=keyend
error(message,2)
end
-- Error handling function
-- Returns a wiki friendly error or an empty string if error reporting is disabled
localfunctiongetError(key,value)
localTNT=require('Module:TNT')
localok,message=pcall(TNT.format,'I18n/Module:Transcluder.tab','error-'..key,value)
ifnotokthenmessage=keyend
message=mw.html.create('div'):addClass('error'):wikitext(message)
returnmessage
end
-- Helper function to get the local name of a namespace and all its aliases
-- @param name Canonical name of the namespace, for example 'File'
-- @return Local name of the namespace and all aliases, for example {'File','Image','Archivo','Imagen'}
localfunctiongetNamespaces(name)
localnamespaces=mw.clone(mw.site.namespaces[name].aliases)-- Clone because https://en.wikipedia.org/w/index.php?diff=1056921358
table.insert(namespaces,mw.site.namespaces[name].name)
table.insert(namespaces,mw.site.namespaces[name].canonicalName)
returnnamespaces
end
-- Get the page wikitext, following redirects
-- Also returns the page name, or the target page name if a redirect was followed, or false if no page was found
-- For file pages, returns the content of the file description page
localfunctiongetText(page,noFollow)
page=mw.text.decode(page)
localtitle=mw.title.new(page)
ifnottitlethenreturnfalse,falseend
localtarget=title.redirectTarget
iftargetandnotnoFollowthentitle=targetend
localtext=title:getContent()
ifnottextthenreturnfalse,title.prefixedTextend
-- Remove <noinclude> tags
text=string.gsub(text,'<[Nn][Oo][Ii][Nn][Cc][Ll][Uu][Dd][Ee]>.-</[Nn][Oo][Ii][Nn][Cc][Ll][Uu][Dd][Ee]>','')-- remove noinclude bits
-- Keep <onlyinclude> tags
ifstring.find(text,'onlyinclude')then-- avoid expensive search if possible
text=text
:gsub('</onlyinclude>.-<onlyinclude>','')-- remove text between onlyinclude sections
:gsub('^.-<onlyinclude>','')-- remove text before first onlyinclude section
:gsub('</onlyinclude>.*','')-- remove text after last onlyinclude section
end
returntext,title.prefixedText
end
-- Get the requested files from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of files to return, for example 2 or '1,3-5'. Omit to return all files.
-- @return Sequence of strings containing the wikitext of the requested files.
-- @return Original wikitext minus requested files.
localfunctiongetFiles(text,flags)
localfiles={}
localflags,blacklist=parseFlags(flags)
localfileNamespaces=getNamespaces('File')
localname
localcount=0
forfileinstring.gmatch(text,'%b[]')do
ifmatchAnyLink(file,fileNamespaces)then
name=string.match(file,'%[%[[^:]-:([^]|]+)')
count=count+1
ifnotblacklistand(notflagsorflags[count]ormatchFlag(name,flags))
orblacklistandflagsandnotflags[count]andnotmatchFlag(name,flags)then
table.insert(files,file)
else
text=removeString(text,file)
end
end
end
returnfiles,text
end
-- Get the requested tables from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of tables to return, for example 2 or '1,3-5'. Omit to return all tables.
-- @return Sequence of strings containing the wikitext of the requested tables.
-- @return Original wikitext minus requested tables.
localfunctiongetTables(text,flags)
localtables={}
localflags,blacklist=parseFlags(flags)
localid
localcount=0
fortinstring.gmatch('\n'..text,'\n%b{}')do
ifstring.sub(t,1,3)=='\n{|'then
id=string.match(t,'\n{|[^\n]-id%s*=%s*["\']?([^"\'\n]+)["\']?[^\n]*\n')
count=count+1
ifnotblacklistand(notflagsorflags[count]orflags[id])
orblacklistandflagsandnotflags[count]andnotflags[id]then
table.insert(tables,t)
else
text=removeString(text,t)
end
end
end
returntables,text
end
-- Get the requested templates from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of templates to return, for example 2 or '1,3-5'. Omit to return all templates.
-- @return Sequence of strings containing the wikitext of the requested templates.
-- @return Original wikitext minus requested templates.
localfunctiongetTemplates(text,flags)
localtemplates={}
localflags,blacklist=parseFlags(flags)
localname
localcount=0
fortemplateinstring.gmatch(text,'{%b{}}')do
ifstring.sub(template,1,3)~='{{#'then-- skip parser functions like #if
name=mw.text.trim(string.match(template,'{{([^}|\n]+)')or"")-- get the template name
ifname~=""then
count=count+1
ifnotblacklistand(notflagsorflags[count]ormatchFlag(name,flags))
orblacklistandflagsandnotflags[count]andnotmatchFlag(name,flags)then
table.insert(templates,template)
else
text=removeString(text,template)
end
end
end
end
returntemplates,text
end
-- Get the requested template parameters from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of parameters to return, for example 2 or '1,3-5'. Omit to return all parameters.
-- @return Map from parameter name to value, NOT IN THE ORIGINAL ORDER
-- @return Original wikitext minus requested parameters.
-- @return Order in which the parameters were parsed.
localfunctiongetParameters(text,flags)
localparameters,parameterOrder={},{}
localflags,blacklist=parseFlags(flags)
localparams,count,parts,key,value
fortemplateinstring.gmatch(text,'{%b{}}')do
params=string.match(template,'{{[^|}]-|(.*)}}')
ifparamsthen
count=0
-- Temporarily replace pipes in subtemplates and links to avoid chaos
forsubtemplateinstring.gmatch(params,'{%b{}}')do
params=string.gsub(params,escapeString(subtemplate),string.gsub(subtemplate,".",{["%"]="%%",["|"]="@@:@@",["="]="@@_@@"}))
end
forlinkinstring.gmatch(params,'%b[]')do
params=string.gsub(params,escapeString(link),string.gsub(link,".",{["%"]="%%",["|"]="@@:@@",["="]="@@_@@"}))
end
forparameterinmw.text.gsplit(params,'|')do
parts=mw.text.split(parameter,'=')
key=mw.text.trim(parts[1])
if#parts==1then
value=key
count=count+1
key=count
else
value=mw.text.trim(table.concat(parts,'=',2))
end
value=string.gsub(string.gsub(value,'@@:@@','|'),'@@_@@','=')
ifnotblacklistand(notflagsormatchFlag(key,flags))
orblacklistandflagsandnotmatchFlag(key,flags)then
table.insert(parameterOrder,key)
parameters[key]=value
else
text=removeString(text,parameter)
end
end
end
end
returnparameters,text,parameterOrder
end
-- Get the requested lists from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of lists to return, for example 2 or '1,3-5'. Omit to return all lists.
-- @return Sequence of strings containing the wikitext of the requested lists.
-- @return Original wikitext minus requested lists.
localfunctiongetLists(text,flags)
locallists={}
localflags,blacklist=parseFlags(flags)
localcount=0
forlistinstring.gmatch('\n'..text..'\n\n','\n([*#].-)\n[^*#]')do
count=count+1
ifnotblacklistand(notflagsorflags[count])
orblacklistandflagsandnotflags[count]then
table.insert(lists,list)
else
text=removeString(text,list)
end
end
returnlists,text
end
-- Get the requested paragraphs from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of paragraphs to return, for example 2 or '1,3-5'. Omit to return all paragraphs.
-- @return Sequence of strings containing the wikitext of the requested paragraphs.
-- @return Original wikitext minus requested paragraphs.
localfunctiongetParagraphs(text,flags)
localparagraphs={}
localflags,blacklist=parseFlags(flags)
-- Remove non-paragraphs
localelements
localtemp='\n'..text..'\n'
elements,temp=getLists(temp,0)-- remove lists
elements,temp=getFiles(temp,0)-- remove files
temp=mw.text.trim((temp
:gsub('\n%b{} *\n','\n%0\n')-- add spacing between tables and block templates
:gsub('\n%b{} *\n','\n')-- remove tables and block templates
:gsub('\n==+[^=]+==+ *\n','\n')-- remove section titles
))
-- Assume that anything remaining is a paragraph
localcount=0
forparagraphinmw.text.gsplit(temp,'\n\n+')do
ifmw.text.trim(paragraph)~=''then
count=count+1
ifnotblacklistand(notflagsorflags[count])
orblacklistandflagsandnotflags[count]then
table.insert(paragraphs,paragraph)
else
text=removeString(text,paragraph)
end
end
end
returnparagraphs,text
end
-- Get the requested categories from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of categories to return, for example 2 or '1,3-5'. Omit to return all categories.
-- @return Sequence of strings containing the wikitext of the requested categories.
-- @return Original wikitext minus requested categories.
localfunctiongetCategories(text,flags)
localcategories={}
localflags,blacklist=parseFlags(flags)
localcategoryNamespaces=getNamespaces('Category')
localname
localcount=0
forcategoryinstring.gmatch(text,'%b[]')do
ifmatchAnyLink(category,categoryNamespaces)then
name=string.match(category,'%[%[[^:]-:([^]|]+)')
count=count+1
ifnotblacklistand(notflagsorflags[count]ormatchFlag(name,flags))
orblacklistandflagsandnotflags[count]andnotmatchFlag(name,flags)then
table.insert(categories,category)
else
text=removeString(text,category)
end
end
end
returncategories,text
end
-- Get the requested references from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of references to return, for example 2 or '1,3-5'. Omit to return all references.
-- @return Sequence of strings containing the wikitext of the requested references.
-- @return Original wikitext minus requested references.
localfunctiongetReferences(text,flags)
localreferences={}
-- Remove all references, including citations, when 0 references are requested
-- This is kind of hacky but currently necessary because the rest of the code
-- doesn't remove citations like <ref name="Foo" /> if Foo is defined elsewhere
ifflagsandnottruthy(flags)then
text=string.gsub(text,'<%s*[Rr][Ee][Ff][^>/]*>.-<%s*/%s*[Rr][Ee][Ff]%s*>','')
text=string.gsub(text,'<%s*[Rr][Ee][Ff][^>/]*/%s*>','')
returnreferences,text
end
localflags,blacklist=parseFlags(flags)
localname
localcount=0
forreferenceinstring.gmatch(text,'<%s*[Rr][Ee][Ff][^>/]*>.-<%s*/%s*[Rr][Ee][Ff]%s*>')do
name=string.match(reference,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?([^"\'>/]+)["\']?[^>]*%s*>')
count=count+1
ifnotblacklistand(notflagsorflags[count]ormatchFlag(name,flags))
orblacklistandflagsandnotflags[count]andnotmatchFlag(name,flags)then
table.insert(references,reference)
else
text=removeString(text,reference)
ifnamethen
forcitationinstring.gmatch(text,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?'..escapeString(name)..'["\']?[^/>]*/%s*>')do
text=removeString(text,citation)
end
end
end
end
returnreferences,text
end
-- Get the lead section from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @return Wikitext of the lead section.
localfunctiongetLead(text)
text=string.gsub('\n'..text,'\n==.*','')
text=mw.text.trim(text)
ifnottextthenreturnthrowError('lead-empty')end
returntext
end
-- Get the requested sections from the given wikitext.
-- @param text Required. Wikitext to parse.
-- @param flags Range of sections to return, for example 2 or '1,3-5'. Omit to return all sections.
-- @return Sequence of strings containing the wikitext of the requested sections.
-- @return Original wikitext minus requested sections.
localfunctiongetSections(text,flags)
localsections={}
localflags,blacklist=parseFlags(flags)
localcount=0
localprefix,section,suffix
fortitleinstring.gmatch('\n'..text..'\n==','\n==+%s*([^=]+)%s*==+')do
count=count+1
prefix,section,suffix=string.match('\n'..text..'\n==','\n()==+%s*'..escapeString(title)..'%s*==+(.-)()\n==')
ifnotblacklistand(notflagsorflags[count]ormatchFlag(title,flags))
orblacklistandflagsandnotflags[count]andnotmatchFlag(title,flags)then
sections[title]=section
else
text=string.sub(text,1,prefix)..string.sub(text,suffix)
text=string.gsub(text,'\n?==$','')-- remove the trailing \n==
end
end
returnsections,text
end
-- Get the requested section or <section> tag from the given wikitext (including subsections).
-- @param text Required. Wikitext to parse.
-- @param section Required. Title of the section to get (in wikitext), for example 'History' or 'History of [[Athens]]'.
-- @return Wikitext of the requested section.
localfunctiongetSection(text,section)
section=mw.text.trim(section)
localescapedSection=escapeString(section)
-- First check if the section title matches a <section> tag
ifstring.find(text,'<%s*[Ss]ection%s+begin%s*=%s*["\']?%s*'..escapedSection..'%s*["\']?%s*/>')then-- avoid expensive search if possible
text=mw.text.trim((text
:gsub('<%s*[Ss]ection%s+end=%s*["\']?%s*'..escapedSection..'%s*["\']?%s*/>.-<%s*[Ss]ection%s+begin%s*=%s*["\']?%s*'..escapedSection..'%s*["\']?%s*/>','')-- remove text between section tags
:gsub('^.-<%s*[Ss]ection%s+begin%s*=%s*["\']?%s*'..escapedSection..'%s*["\']?%s*/>','')-- remove text before first section tag
:gsub('<%s*[Ss]ection%s+end=%s*["\']?%s*'..escapedSection..'%s*["\']?%s*/>.*','')-- remove text after last section tag
))
iftext==''thenreturnthrowError('section-tag-empty',section)end
returntext
end
locallevel,text=string.match('\n'..text..'\n','\n(==+)%s*'..escapedSection..'%s*==.-\n(.*)')
ifnottextthenreturnthrowError('section-not-found',section)end
localnextSection='\n=='..string.rep('=?',#level-2)..'[^=].*'
text=string.gsub(text,nextSection,'')-- remove later sections with headings at this level or higher
text=mw.text.trim(text)
iftext==''thenreturnthrowError('section-empty',section)end
returntext
end
-- Replace the first call to each reference defined outside of the text for the full reference, to prevent undefined references
-- Then prefix the page title to the reference names to prevent conflicts
-- that is, replace <ref name="Foo"> for <ref name="Title of the article Foo">
-- and also <ref name="Foo" /> for <ref name="Title of the article Foo" />
-- also remove reference groups: <ref name="Foo" group="Bar"> for <ref name="Title of the article Foo">
-- and <ref group="Bar"> for <ref>
-- @todo The current regex may fail in cases with both kinds of quotes, like <ref name="Darwin's book">
localfunctionfixReferences(text,page,full)
ifnotfullthenfull=getText(page)end
localrefNames={}
localrefName
localrefBody
localposition=1
whileposition<mw.ustring.len(text)do
refName,position=mw.ustring.match(text,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?([^"\'>]+)["\']?[^>]*/%s*>()',position)
ifrefNamethen
refName=mw.text.trim(refName)
ifnotrefNames[refName]then-- make sure we process each ref name only once
table.insert(refNames,refName)
refName=escapeString(refName)
refBody=mw.ustring.match(text,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?%s*'..refName..'%s*["\']?[^>/]*>.-<%s*/%s*[Rr][Ee][Ff]%s*>')
ifnotrefBodythen-- the ref body is not in the excerpt
refBody=mw.ustring.match(full,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?%s*'..refName..'%s*["\']?[^/>]*>.-<%s*/%s*[Rr][Ee][Ff]%s*>')
ifrefBodythen-- the ref body was found elsewhere
text=mw.ustring.gsub(text,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?%s*'..refName..'%s*["\']?[^>]*/?%s*>',mw.ustring.gsub(refBody,'%%','%%%%'),1)
end
end
end
else
position=mw.ustring.len(text)
end
end
page=string.gsub(page,'"','')-- remove any quotation marks from the page title
text=mw.ustring.gsub(text,'<%s*[Rr][Ee][Ff][^>]*name%s*=%s*["\']?([^"\'>/]+)["\']?[^>/]*(/?)%s*>','<ref name="'..page..' %1"%2>')
text=mw.ustring.gsub(text,'<%s*[Rr][Ee][Ff]%s*group%s*=%s*["\']?[^"\'>/]+["\']%s*>','<ref>')
returntext
end
-- Replace the bold title or synonym near the start of the page by a link to the page
localfunctionlinkBold(text,page)
locallang=mw.language.getContentLanguage()
localposition=mw.ustring.find(text,"'''"..lang:ucfirst(page).."'''",1,true)-- look for "'''Foo''' is..." (uc) or "A '''foo''' is..." (lc)
ormw.ustring.find(text,"'''"..lang:lcfirst(page).."'''",1,true)-- plain search: special characters in page represent themselves
ifpositionthen
locallength=mw.ustring.len(page)
text=mw.ustring.sub(text,1,position+2).."[["..mw.ustring.sub(text,position+3,position+length+2).."]]"..mw.ustring.sub(text,position+length+3,-1)-- link it
else-- look for anything unlinked in bold, assumed to be a synonym of the title (e.g. a person's birth name)
text=mw.ustring.gsub(text,"()'''(.-'*)'''",function(a,b)
ifnotmw.ustring.find(b,"%[")andnotmw.ustring.find(b,"%{")then-- if not wikilinked or some weird template
return"'''[["..page.."|"..b.."]]'''"-- replace '''Foo''' by '''[[page|Foo]]'''
else
returnnil-- instruct gsub to make no change
end
end,1)-- "end" here terminates the anonymous replacement function(a, b) passed to gsub
end
returntext
end
-- Remove non-free files.
-- @param text Required. Wikitext to clean.
-- @return Clean wikitext.
localfunctionremoveNonFreeFiles(text)
localfileNamespaces=getNamespaces('File')
localfileName
localfileDescription
localframe=mw.getCurrentFrame()
forfileinstring.gmatch(text,'%b[]')do
ifmatchAnyLink(file,fileNamespaces)then
fileName='File:'..string.match(file,'%[%[[^:]-:([^]|]+)')
fileDescription,fileName=getText(fileName)
iffileNamethen
ifnotfileDescriptionorfileDescription==''then
fileDescription=frame:preprocess('{{'..fileName..'}}')-- try Commons
end
iffileDescriptionandstring.match(fileDescription,'[Nn]on%-free')then
text=removeString(text,file)
end
end
end
end
returntext
end
-- Remove any self links
localfunctionremoveSelfLinks(text)
locallang=mw.language.getContentLanguage()
localpage=escapeString(mw.title.getCurrentTitle().prefixedText)
localucpage=lang:ucfirst(page)
locallcpage=lang:lcfirst(page)
text=text
:gsub('%[%[('..ucpage..')%]%]','%1')
:gsub('%[%[('..lcpage..')%]%]','%1')
:gsub('%[%['..ucpage..'|([^]]+)%]%]','%1')
:gsub('%[%['..lcpage..'|([^]]+)%]%]','%1')
returntext
end
-- Remove all wikilinks
localfunctionremoveLinks(text)
text=text
:gsub('%[%[[^%]|]+|([^]]+)%]%]','%1')
:gsub('%[%[([^]]+)%]%]','%1')
:gsub('%[[^ ]+ ([^]]+)%]','%1')
:gsub('%[([^]]+)%]','%1')
returntext
end
-- Remove HTML comments
localfunctionremoveComments(text)
text=string.gsub(text,'<!%-%-.-%-%->','')
returntext
end
-- Remove behavior switches, such as __NOTOC__
localfunctionremoveBehaviorSwitches(text)
text=string.gsub(text,'__[A-Z]+__','')
returntext
end
-- Remove bold text
localfunctionremoveBold(text)
text=string.gsub(text,"'''",'')
returntext
end
-- Main function for modules
localfunctionget(page,options)
ifnotoptionsthenoptions={}end
-- Make sure the page exists
ifnotpagethenreturnthrowError('no-page')end
page=mw.text.trim(page)
page=mw.text.decode(page)
ifpage==''thenreturnthrowError('no-page')end
localpage,hash,section=string.match(page,'([^#]+)(#?)(.*)')
localtext,temp=getText(page,options.noFollow)
ifnottempthenreturnthrowError('invalid-title',page)end
page=temp
ifnottextthenreturnthrowError('page-not-found',page)end
localfull=text-- save the full text for fixReferences below
-- Get the requested section
iftruthy(section)then
text=getSection(text,section)
elseiftruthy(hash)then
text=getLead(text)
end
-- Keep only the requested elements
localelements
ifoptions.onlythen
ifoptions.only=='sections'thenelements=getSections(text,options.sections)end
ifoptions.only=='lists'thenelements=getLists(text,options.lists)end
ifoptions.only=='files'thenelements=getFiles(text,options.files)end
ifoptions.only=='tables'thenelements=getTables(text,options.tables)end
ifoptions.only=='templates'thenelements=getTemplates(text,options.templates)end
ifoptions.only=='parameters'thenelements=getParameters(text,options.parameters)end
ifoptions.only=='paragraphs'thenelements=getParagraphs(text,options.paragraphs)end
ifoptions.only=='categories'thenelements=getCategories(text,options.categories)end
ifoptions.only=='references'thenelements=getReferences(text,options.references)end
text=''
ifelementsthen
forkey,elementinpairs(elements)do
text=text..'\n'..element..'\n'
end
end
end
-- Filter the requested elements
ifoptions.sectionsandoptions.only~='sections'thenelements,text=getSections(text,options.sections)end
ifoptions.listsandoptions.only~='lists'thenelements,text=getLists(text,options.lists)end
ifoptions.filesandoptions.only~='files'thenelements,text=getFiles(text,options.files)end
ifoptions.tablesandoptions.only~='tables'thenelements,text=getTables(text,options.tables)end
ifoptions.templatesandoptions.only~='templates'thenelements,text=getTemplates(text,options.templates)end
ifoptions.parametersandoptions.only~='parameters'thenelements,text=getParameters(text,options.parameters)end
ifoptions.paragraphsandoptions.only~='paragraphs'thenelements,text=getParagraphs(text,options.paragraphs)end
ifoptions.categoriesandoptions.only~='categories'thenelements,text=getCategories(text,options.categories)end
ifoptions.referencesandoptions.only~='references'thenelements,text=getReferences(text,options.references)end
-- Misc options
iftruthy(options.fixReferences)thentext=fixReferences(text,page,full)end
iftruthy(options.linkBold)andnottruthy(section)thentext=linkBold(text,page)end
iftruthy(options.noBold)thentext=removeBold(text)end
iftruthy(options.noLinks)thentext=removeLinks(text)end
iftruthy(options.noSelfLinks)thentext=removeSelfLinks(text)end
iftruthy(options.noNonFreeFiles)thentext=removeNonFreeFiles(text)end
iftruthy(options.noBehaviorSwitches)thentext=removeBehaviorSwitches(text)end
iftruthy(options.noComments)thentext=removeComments(text)end
-- Remove multiple newlines left over from removing elements
text=string.gsub(text,'\n\n\n+','\n\n')
text=mw.text.trim(text)
returntext
end
-- Main invocation function for templates
localfunctionmain(frame)
localargs=parseArgs(frame)
localpage=args[1]
localok,text=pcall(get,page,args)
ifnotokthenreturngetError(text)end
localraw=args['raw']
ifrawthenreturntextend
returnframe:preprocess(text)
end
-- Entry points for templates
functionp.main(frame)returnmain(frame)end
-- Entry points for modules
functionp.get(page,options)returnget(page,options)end
functionp.getText(page,noFollow)returngetText(page,noFollow)end
functionp.getLead(text)returngetLead(text)end
functionp.getSection(text,section)returngetSection(text,section)end
functionp.getSections(text,flags)returngetSections(text,flags)end
functionp.getParagraphs(text,flags)returngetParagraphs(text,flags)end
functionp.getParameters(text,flags)returngetParameters(text,flags)end
functionp.getCategories(text,flags)returngetCategories(text,flags)end
functionp.getReferences(text,flags)returngetReferences(text,flags)end
functionp.getTemplates(text,flags)returngetTemplates(text,flags)end
functionp.getTables(text,flags)returngetTables(text,flags)end
functionp.getLists(text,flags)returngetLists(text,flags)end
functionp.getFiles(text,flags)returngetFiles(text,flags)end
functionp.getError(message,value)returngetError(message,value)end
-- Expose handy methods
functionp.truthy(value)returntruthy(value)end
functionp.parseArgs(frame)returnparseArgs(frame)end
functionp.matchAny(text,pre,list,post,init)returnmatchAny(text,pre,list,post,init)end
functionp.matchFlag(value,flags)returnmatchFlag(value,flags)end
functionp.getNamespaces(name)returngetNamespaces(name)end
functionp.removeBold(text)returnremoveBold(text)end
functionp.removeLinks(text)returnremoveLinks(text)end
functionp.removeSelfLinks(text)returnremoveSelfLinks(text)end
functionp.removeNonFreeFiles(text)returnremoveNonFreeFiles(text)end
functionp.removeBehaviorSwitches(text)returnremoveBehaviorSwitches(text)end
functionp.removeComments(text)returnremoveComments(text)end
returnp