Module:Category handler: Difference between revisions

Jump to navigation Jump to search
(tweak blacklist - on matching subpages, don't match "basepage/", only "basepage/foo")
m (1 revision imported)
 
(13 intermediate revisions by 4 users not shown)
Line 1: Line 1:
----------------------------------------------------------------------
--------------------------------------------------------------------------------
--                                                                 --
--                                                                           --
--                         CATEGORY HANDLER                         --
--                             CATEGORY HANDLER                             --
--                                                                 --
--                                                                           --
--      This module implements the {{category handler}} template   --
--      This module implements the {{category handler}} template in Lua,      --
--      in Lua, with a few improvements: all namespaces and all    --
--      with a few improvements: all namespaces and all namespace aliases     --
--      namespace aliases are supported, and namespace names are   --
--      are supported, and namespace names are detected automatically for    --
--      detected automatically for the local wiki. This module      --
--      the local wiki. This module requires [[Module:Namespace detect]]     --
--      requires [[Module:Namespace detect]] to be available on    --
--      and [[Module:Yesno]] to be available on the local wiki. It can be     --
--      the local wiki. It can be configured for different wikis   --
--      configured for different wikis by altering the values in              --
--      by altering the values in the "cfg" table.                 --
--      [[Module:Category handler/config]], and pages can be blacklisted      --
--                                                                 --
--      from categorisation by using [[Module:Category handler/blacklist]].   --
----------------------------------------------------------------------
--                                                                           --
--------------------------------------------------------------------------------


----------------------------------------------------------------------
-- Load required modules
--                      Configuration data                          --
local yesno = require('Module:Yesno')
--      Language-specific parameter names and values can be set    --
--      here.                                                      --
----------------------------------------------------------------------


local cfg = {}
-- Lazily load things we don't always need
local mShared, mappings


-- cfg.nocat is the parameter name to suppress categorisation.
local p = {}
-- cfg.nocatTrue is the value to suppress categorisation, and
-- cfg.nocatFalse is the value to both categorise and to skip the
-- blacklist check.
cfg.nocat = 'nocat'   
cfg.nocatTrue = 'true'
cfg.nocatFalse = 'false'


-- The parameter name for the legacy "categories" parameter. This
--------------------------------------------------------------------------------
-- skips the blacklist if set to the cfg.category2Yes value, and
-- Helper functions
-- suppresses categorisation if set to the cfg.categoriesNo value.
--------------------------------------------------------------------------------
cfg.categories = 'categories'
cfg.categoriesYes = 'yes'
cfg.categoriesNo = 'no'


-- The parameter name for the legacy "category2" parameter. This
local function trimWhitespace(s, removeBlanks)
-- skips the blacklist if set to the cfg.category2Yes value, and
if type(s) ~= 'string' then
-- suppresses categorisation if present but equal to anything other
return s
-- than cfg.category2Yes or cfg.category2Negative.
end
cfg.category2 = 'category2'
s = s:match('^%s*(.-)%s*$')
cfg.category2Yes = 'yes'
if removeBlanks then
cfg.category2Negative = '¬'
if s ~= '' then
return s
else
return nil
end
else
return s
end
end


-- cfg.subpage is the parameter name to specify how to behave on
--------------------------------------------------------------------------------
-- subpages. cfg.subpageNo is the value to specify to not
-- CategoryHandler class
-- categorise on subpages; cfg.only is the value to specify to only
--------------------------------------------------------------------------------
-- categorise on subpages.
cfg.subpage = 'subpage'
cfg.subpageNo = 'no'
cfg.subpageOnly = 'only'


-- The parameter for data to return in all namespaces.
local CategoryHandler = {}
cfg.all = 'all'
CategoryHandler.__index = CategoryHandler


-- The parameter name for data to return if no data is specified for
function CategoryHandler.new(data, args)
-- the namespace that is detected. This must be the same as the
local obj = setmetatable({ _data = data, _args = args }, CategoryHandler)
-- cfg.other parameter in [[Module:Namespace detect]].
cfg.other = 'other'
-- Set the title object
do
local pagename = obj:parameter('demopage')
local success, titleObj
if pagename then
success, titleObj = pcall(mw.title.new, pagename)
end
if success and titleObj then
obj.title = titleObj
if titleObj == mw.title.getCurrentTitle() then
obj._usesCurrentTitle = true
end
else
obj.title = mw.title.getCurrentTitle()
obj._usesCurrentTitle = true
end
end


-- The parameter name used to specify a page other than the current
-- Set suppression parameter values
-- page; used for testing and demonstration. This must be the same
for _, key in ipairs{'nocat', 'categories'} do
-- as the cfg.page parameter in [[Module:Namespace detect]].
local value = obj:parameter(key)
cfg.page = 'page'
value = trimWhitespace(value, true)
obj['_' .. key] = yesno(value)
end
do
local subpage = obj:parameter('subpage')
local category2 = obj:parameter('category2')
if type(subpage) == 'string' then
subpage = mw.ustring.lower(subpage)
end
if type(category2) == 'string' then
subpage = mw.ustring.lower(category2)
end
obj._subpage = trimWhitespace(subpage, true)
obj._category2 = trimWhitespace(category2) -- don't remove blank values
end
return obj
end


-- The categorisation blacklist. Pages that match Lua patterns in this
function CategoryHandler:parameter(key)
-- list will not be categorised unless any of the following options are
local parameterNames = self._data.parameters[key]
-- set: "nocat=false", "categories=yes", or "category2=yes".
local pntype = type(parameterNames)
-- If the namespace name has a space in, it must be written with an
if pntype == 'string' or pntype == 'number' then
-- underscore, e.g. "Wikipedia_talk". Other parts of the title can have
return self._args[parameterNames]
-- either underscores or spaces.
elseif pntype == 'table' then
cfg.blacklist = {
for _, name in ipairs(parameterNames) do
    '^Main Page$', -- don't categorise the main page.
local value = self._args[name]
   
if value ~= nil then
    -- Don't categorise the following pages or their subpages.
return value
    '^Wikipedia:Cascade%-protected items$',
end
    '^Wikipedia:Cascade%-protected items/.*$',
end
    '^User:UBX$', -- The userbox "template" space.
return nil
    '^User:UBX/.*$',
else
    '^User_talk:UBX$',
error(string.format(
    '^User_talk:UBX/.*$',
'invalid config key "%s"',
   
tostring(key)
    -- Don't categorise subpages of these pages, but allow
), 2)
    -- categorisation of the base page.
end
    '^Wikipedia:Template messages/.+$',
end
   
    '/[aA]rchive' -- Don't categorise archives.
}


-- This is a table of namespaces to categorise by default. They
function CategoryHandler:isSuppressedByArguments()
-- should be in the format of parameter names accepted by
return
-- [[Module:Namespace detect]].
-- See if a category suppression argument has been set.
cfg.defaultNamespaces = {
self._nocat == true
    'main',
or self._categories == false
    'file',
or (
    'help',
self._category2
    'category'
and self._category2 ~= self._data.category2Yes
}
and self._category2 ~= self._data.category2Negative
)


----------------------------------------------------------------------
-- Check whether we are on a subpage, and see if categories are
--                    End configuration data                      --
-- suppressed based on our subpage status.
----------------------------------------------------------------------
or self._subpage == self._data.subpageNo and self.title.isSubpage
or self._subpage == self._data.subpageOnly and not self.title.isSubpage
end


-- Get [[Module:Namespace detect]] and declare the table of functions
function CategoryHandler:shouldSkipBlacklistCheck()
-- that we will return.
-- Check whether the category suppression arguments indicate we
local NamespaceDetect = require('Module:Namespace detect')
-- should skip the blacklist check.
local p = {}
return self._nocat == false
or self._categories == true
or self._category2 == self._data.category2Yes
end


----------------------------------------------------------------------
function CategoryHandler:matchesBlacklist()
--                        Local functions                          --
if self._usesCurrentTitle then
--      The following are internal functions, which we do not want  --
return self._data.currentTitleMatchesBlacklist
--      to be accessible from other modules.                       --
else
----------------------------------------------------------------------
mShared = mShared or require('Module:Category handler/shared')
return mShared.matchesBlacklist(
self.title.prefixedText,
mw.loadData('Module:Category handler/blacklist')
)
end
end


-- Find whether we need to return a category or not.
function CategoryHandler:isSuppressed()
local function needsCategory( pageObject, args )
-- Find if categories are suppressed by either the arguments or by
    -- Don't categorise if the relevant options are set.
-- matching the blacklist.
    if args[cfg.nocat] == cfg.nocatTrue
return self:isSuppressedByArguments()
        or args[cfg.categories] == cfg.categoriesNo
or not self:shouldSkipBlacklistCheck() and self:matchesBlacklist()
        or ( args[cfg.category2]
            and args[cfg.category2] ~= cfg.category2Yes
            and args[cfg.category2] ~= cfg.category2Negative )
        then
        return false
    end
    -- If there is no pageObject available, then that either means that we are over
    -- the expensive function limit or that the title specified was invalid. Invalid
    -- titles will probably only be a problem during testing, so we choose the best
    -- fallback for being over the expensive function limit. The fallback behaviour
    -- of the old template was to assume the page was not a subpage, so we will do
    -- the same here.
    if args[cfg.subpage] == cfg.subpageNo and pageObject and pageObject.isSubpage then
        return false
    end
    if args[cfg.subpage] == cfg.subpageOnly
        and (not pageObject or (pageObject and not pageObject.isSubpage) ) then
        return false
    end
    return true
end
end


-- Find whether we need to check the blacklist or not.
function CategoryHandler:getNamespaceParameters()
local function needsBlacklistCheck( args )
if self._usesCurrentTitle then
    if args[cfg.nocat] == cfg.nocatFalse
return self._data.currentTitleNamespaceParameters
        or args[cfg.categories] == cfg.categoriesYes
else
        or args[cfg.category2] == cfg.category2Yes then
if not mappings then
        return false
mShared = mShared or require('Module:Category handler/shared')
    else
mappings = mShared.getParamMappings(true) -- gets mappings with mw.loadData
        return true
end
    end
return mShared.getNamespaceParameters(
self.title,
mappings
)
end
end
end


-- Searches the blacklist to find a match with the page object. The
function CategoryHandler:namespaceParametersExist()
-- string searched is the namespace plus the title, including subpages.
-- Find whether any namespace parameters have been specified.
-- Returns true if there is a match, otherwise returns false.
-- We use the order "all" --> namespace params --> "other" as this is what
local function findBlacklistMatch( pageObject )
-- the old template did.
    if not pageObject then return end
if self:parameter('all') then
   
return true
    -- Get the title to check.
end
    local title = pageObject.nsText -- Get the namespace.
if not mappings then
    -- Append a colon if the namespace isn't the blank string.
mShared = mShared or require('Module:Category handler/shared')
    if #title > 0 then
mappings = mShared.getParamMappings(true) -- gets mappings with mw.loadData
        title = title .. ':' .. pageObject.text
end
    else
for ns, params in pairs(mappings) do
        title = pageObject.text
for i, param in ipairs(params) do
    end
if self._args[param] then
   
return true
    -- Check the blacklist.
end
    for i, pattern in ipairs( cfg.blacklist ) do
end
        if mw.ustring.match( title, pattern ) then
end
            return true
if self:parameter('other') then
        end
return true
    end
end
    return false
return false
end
end


-- Find whether any namespace parameters have been specified.
function CategoryHandler:getCategories()
-- Mappings is the table of parameter mappings taken from
local params = self:getNamespaceParameters()
-- [[Module:Namespace detect]].
local nsCategory
local function nsParamsExist( mappings, args )
for i, param in ipairs(params) do
    if args[cfg.all] or args[cfg.other] then
local value = self._args[param]
        return true
if value ~= nil then
    end
nsCategory = value
    for ns, params in pairs( mappings ) do
break
        for i, param in ipairs( params ) do
end
            if args[param] then
end
                return true
if nsCategory ~= nil or self:namespaceParametersExist() then
            end
-- Namespace parameters exist - advanced usage.
        end
if nsCategory == nil then
    end
nsCategory = self:parameter('other')
    return false
end
local ret = {self:parameter('all')}
local numParam = tonumber(nsCategory)
if numParam and numParam >= 1 and math.floor(numParam) == numParam then
-- nsCategory is an integer
ret[#ret + 1] = self._args[numParam]
else
ret[#ret + 1] = nsCategory
end
if #ret < 1 then
return nil
else
return table.concat(ret)
end
elseif self._data.defaultNamespaces[self.title.namespace] then
-- Namespace parameters don't exist, simple usage.
return self._args[1]
end
return nil
end
end


-- The main structure of the module. Checks whether we need to categorise,
--------------------------------------------------------------------------------
-- and then passes the relevant arguments to [[Module:Namespace detect]].
-- Exports
local function _main( args )
--------------------------------------------------------------------------------
    -- Get the page object and argument mappings from
 
    -- [[Module:Namespace detect]], to save us from having to rewrite the
local p = {}
    -- code.
 
    local pageObject = NamespaceDetect.getPageObject( args[cfg.page] )
function p._exportClasses()
    local mappings = NamespaceDetect.getParamMappings()
-- Used for testing purposes.
   
return {
    -- Check if we need a category or not, and return nothing if not.
CategoryHandler = CategoryHandler
    if not needsCategory( pageObject, args ) then return end
}
   
    local ret = '' -- The string to return.
    -- Check blacklist if necessary.
    if not needsBlacklistCheck( args )
        or not findBlacklistMatch( pageObject ) then
       
        if not nsParamsExist( mappings, args ) then
            -- No namespace parameters exist; basic usage. Pass args[1] to
            -- [[Module:Namespace detect]] using the default namespace
            -- parameters, and return the result.
            local ndargs = {}
            for _, ndarg in ipairs( cfg.defaultNamespaces ) do
                ndargs[ndarg] = args[1]
            end
            ndargs.page = args.page
            local ndresult = NamespaceDetect.main( ndargs )
            if ndresult then
                ret = ret .. ndresult
            end
        else
            -- Namespace parameters exist; advanced usage.
            -- If the all parameter is specified, return it.
            if args.all then
                ret = ret .. args.all
            end
           
            -- Get the arguments to pass to [[Module:Namespace detect]].
            local ndargs = {}
            for ns, params in pairs( mappings ) do
                for _, param in ipairs( params ) do
                    ndargs[param] = args[param] or args[cfg.other] or nil
                end
            end
            if args.other then
                ndargs.other = args.other
            end
            if args.page then
                ndargs.page = args.page
            end
           
            local data = NamespaceDetect.main( ndargs )
           
            -- Work out what to return based on the result of the namespace
            -- detect call.
            local datanum = tonumber( data )
            if type( datanum ) == 'number' then
                -- "data" is a number, so return that positional parameter.
                -- Remove non-positive integer values, as only positive integers
                -- from 1-10 were used with the old template.
                if datanum > 0
                    and math.floor( datanum ) == datanum
                    and args[datanum] then
                    ret = ret .. args[ datanum ]
                end
            else
                -- "data" is not a number, so return it as it is.
                if type(data) == 'string' then
                    ret = ret .. data
                end
            end
        end
    end
    return ret
end
end


----------------------------------------------------------------------
function p._main(args, data)
--                        Global functions                          --
data = data or mw.loadData('Module:Category handler/data')
--      The following functions are global, because we want them    --
local handler = CategoryHandler.new(data, args)
--      to be accessible from #invoke and from other Lua modules.  --
if handler:isSuppressed() then
--      At the moment only the main function is here. It processes  --
return nil
--      the arguments and passes them to the _main function.        --
end
----------------------------------------------------------------------
return handler:getCategories()
 
end
function p.main( frame )
    -- If called via #invoke, use the args passed into the invoking
    -- template, or the args passed to #invoke if any exist. Otherwise
    -- assume args are being passed directly in.
    local origArgs
    if frame == mw.getCurrentFrame() then
        origArgs = frame:getParent().args
        for k, v in pairs( frame.args ) do
            origArgs = frame.args
            break
        end
    else
        origArgs = frame
    end


    -- Trim whitespace and remove blank arguments for the following args:
function p.main(frame, data)
    -- 1, 2, 3 etc., "nocat", "categories", "subpage", and "page".
data = data or mw.loadData('Module:Category handler/data')
    local args = {}
local args = require('Module:Arguments').getArgs(frame, {
    for k, v in pairs( origArgs ) do
wrappers = data.wrappers,
        v = mw.text.trim(v) -- Trim whitespace.
valueFunc = function (k, v)
        if type(k) == 'number'
v = trimWhitespace(v)
            or k == cfg.nocat
if type(k) == 'number' then
            or k == cfg.categories
if v ~= '' then
            or k == cfg.subpage
return v
            or k == cfg.page then
else
            if v ~= '' then
return nil
                args[k] = v
end
            end
else
        else
return v
            args[k] = v
end
        end
end
    end
})
   
return p._main(args, data)
    -- Lower-case "nocat", "categories", "category2", and "subpage". These
    -- parameters are put in lower case whenever they appear in the old
    -- template, so we can just do it once here and save ourselves some work.
    local lowercase = { cfg.nocat, cfg.categories, cfg.category2, cfg.subpage }
    for _, v in ipairs( lowercase ) do
        if args[v] then
            args[v] = mw.ustring.lower( args[v] )
        end
    end
   
    return _main( args )
end
end


return p
return p

Latest revision as of 07:50, 9 January 2019

Documentation for this module may be created at Module:Category handler/doc

--------------------------------------------------------------------------------
--                                                                            --
--                              CATEGORY HANDLER                              --
--                                                                            --
--      This module implements the {{category handler}} template in Lua,      --
--      with a few improvements: all namespaces and all namespace aliases     --
--      are supported, and namespace names are detected automatically for     --
--      the local wiki. This module requires [[Module:Namespace detect]]      --
--      and [[Module:Yesno]] to be available on the local wiki. It can be     --
--      configured for different wikis by altering the values in              --
--      [[Module:Category handler/config]], and pages can be blacklisted      --
--      from categorisation by using [[Module:Category handler/blacklist]].   --
--                                                                            --
--------------------------------------------------------------------------------

-- Load required modules
local yesno = require('Module:Yesno')

-- Lazily load things we don't always need
local mShared, mappings

local p = {}

--------------------------------------------------------------------------------
-- Helper functions
--------------------------------------------------------------------------------

local function trimWhitespace(s, removeBlanks)
	if type(s) ~= 'string' then
		return s
	end
	s = s:match('^%s*(.-)%s*$')
	if removeBlanks then
		if s ~= '' then
			return s
		else
			return nil
		end
	else
		return s
	end
end

--------------------------------------------------------------------------------
-- CategoryHandler class
--------------------------------------------------------------------------------

local CategoryHandler = {}
CategoryHandler.__index = CategoryHandler

function CategoryHandler.new(data, args)
	local obj = setmetatable({ _data = data, _args = args }, CategoryHandler)
	
	-- Set the title object
	do
		local pagename = obj:parameter('demopage')
		local success, titleObj
		if pagename then
			success, titleObj = pcall(mw.title.new, pagename)
		end
		if success and titleObj then
			obj.title = titleObj
			if titleObj == mw.title.getCurrentTitle() then
				obj._usesCurrentTitle = true
			end
		else
			obj.title = mw.title.getCurrentTitle()
			obj._usesCurrentTitle = true
		end
	end

	-- Set suppression parameter values
	for _, key in ipairs{'nocat', 'categories'} do
		local value = obj:parameter(key)
		value = trimWhitespace(value, true)
		obj['_' .. key] = yesno(value)
	end
	do
		local subpage = obj:parameter('subpage')
		local category2 = obj:parameter('category2')
		if type(subpage) == 'string' then
			subpage = mw.ustring.lower(subpage)
		end
		if type(category2) == 'string' then
			subpage = mw.ustring.lower(category2)
		end
		obj._subpage = trimWhitespace(subpage, true)
		obj._category2 = trimWhitespace(category2) -- don't remove blank values
	end
	return obj
end

function CategoryHandler:parameter(key)
	local parameterNames = self._data.parameters[key]
	local pntype = type(parameterNames)
	if pntype == 'string' or pntype == 'number' then
		return self._args[parameterNames]
	elseif pntype == 'table' then
		for _, name in ipairs(parameterNames) do
			local value = self._args[name]
			if value ~= nil then
				return value
			end
		end
		return nil
	else
		error(string.format(
			'invalid config key "%s"',
			tostring(key)
		), 2)
	end
end

function CategoryHandler:isSuppressedByArguments()
	return
		-- See if a category suppression argument has been set.
		self._nocat == true
		or self._categories == false
		or (
			self._category2
			and self._category2 ~= self._data.category2Yes
			and self._category2 ~= self._data.category2Negative
		)

		-- Check whether we are on a subpage, and see if categories are
		-- suppressed based on our subpage status.
		or self._subpage == self._data.subpageNo and self.title.isSubpage
		or self._subpage == self._data.subpageOnly and not self.title.isSubpage
end

function CategoryHandler:shouldSkipBlacklistCheck()
	-- Check whether the category suppression arguments indicate we
	-- should skip the blacklist check.
	return self._nocat == false
		or self._categories == true
		or self._category2 == self._data.category2Yes
end

function CategoryHandler:matchesBlacklist()
	if self._usesCurrentTitle then
		return self._data.currentTitleMatchesBlacklist
	else
		mShared = mShared or require('Module:Category handler/shared')
		return mShared.matchesBlacklist(
			self.title.prefixedText,
			mw.loadData('Module:Category handler/blacklist')
		)
	end
end

function CategoryHandler:isSuppressed()
	-- Find if categories are suppressed by either the arguments or by
	-- matching the blacklist.
	return self:isSuppressedByArguments()
		or not self:shouldSkipBlacklistCheck() and self:matchesBlacklist()
end

function CategoryHandler:getNamespaceParameters()
	if self._usesCurrentTitle then
		return self._data.currentTitleNamespaceParameters
	else
		if not mappings then
			mShared = mShared or require('Module:Category handler/shared')
			mappings = mShared.getParamMappings(true) -- gets mappings with mw.loadData
		end
		return mShared.getNamespaceParameters(
			self.title,
			mappings
		)
	end
end

function CategoryHandler:namespaceParametersExist()
	-- Find whether any namespace parameters have been specified.
	-- We use the order "all" --> namespace params --> "other" as this is what
	-- the old template did.
	if self:parameter('all') then
		return true
	end
	if not mappings then
		mShared = mShared or require('Module:Category handler/shared')
		mappings = mShared.getParamMappings(true) -- gets mappings with mw.loadData
	end
	for ns, params in pairs(mappings) do
		for i, param in ipairs(params) do
			if self._args[param] then
				return true
			end
		end
	end
	if self:parameter('other') then
		return true
	end
	return false
end

function CategoryHandler:getCategories()
	local params = self:getNamespaceParameters()
	local nsCategory
	for i, param in ipairs(params) do
		local value = self._args[param]
		if value ~= nil then
			nsCategory = value
			break
		end
	end
	if nsCategory ~= nil or self:namespaceParametersExist() then
		-- Namespace parameters exist - advanced usage.
		if nsCategory == nil then
			nsCategory = self:parameter('other')
		end
		local ret = {self:parameter('all')}
		local numParam = tonumber(nsCategory)
		if numParam and numParam >= 1 and math.floor(numParam) == numParam then
			-- nsCategory is an integer
			ret[#ret + 1] = self._args[numParam]
		else
			ret[#ret + 1] = nsCategory
		end
		if #ret < 1 then
			return nil
		else
			return table.concat(ret)
		end
	elseif self._data.defaultNamespaces[self.title.namespace] then
		-- Namespace parameters don't exist, simple usage.
		return self._args[1]
	end
	return nil
end

--------------------------------------------------------------------------------
-- Exports
--------------------------------------------------------------------------------

local p = {}

function p._exportClasses()
	-- Used for testing purposes.
	return {
		CategoryHandler = CategoryHandler
	}
end

function p._main(args, data)
	data = data or mw.loadData('Module:Category handler/data')
	local handler = CategoryHandler.new(data, args)
	if handler:isSuppressed() then
		return nil
	end
	return handler:getCategories()
end

function p.main(frame, data)
	data = data or mw.loadData('Module:Category handler/data')
	local args = require('Module:Arguments').getArgs(frame, {
		wrappers = data.wrappers,
		valueFunc = function (k, v)
			v = trimWhitespace(v)
			if type(k) == 'number' then
				if v ~= '' then
					return v
				else
					return nil
				end
			else
				return v
			end
		end
	})
	return p._main(args, data)
end

return p