Module:CanonicalForms: Difference between revisions

(7 intermediate revisions by the same user not shown)

Line 1:

-- ~~Module~~:CanonicalForms

--[[

~~-- This module provides generic normalization functions.~~

* Name: CanonicalForms

~~-- Templates can supply their own mapping tables (arrays of mapping groups)~~

* Author: Mark W. Datysgeld

~~-- where each group is a table with:~~

* Description: Text normalization utility that removes wiki markup and maps user input to canonical values using configurable lookup tables

~~-- • canonical: The canonical display value.~~

* Notes: Example usage: local mapping = { { canonical = "gTLD", synonyms = {"generic", "g"} }, { canonical = "ccTLD", synonyms = {"country", "cc"} } }; local canonical, css, category = require('Module:CanonicalForms').normalize(inputString, mapping)

~~-- • synonyms~~: ~~A table (array) of synonyms (case insensitive)~~.

]]

~~-- • [optional] css~~: ~~Any extra property (e.g. CSS class) associated with the canonical value.~~

--

~~-- The functions in this module will process an~~ input ~~string against the provided mapping table~~

~~-- and return the~~ canonical ~~value (and optionally any extra property) if a match is found.~~

--

~~-- Usage example~~:

-- local mapping = {

-- { canonical = "gTLD", synonyms = {"generic", "~~gtld", "tld~~"} },

-- { canonical = "ccTLD", synonyms = {~~"cctld",~~ "country", "cc"} }

-- }

-- local canonical, css = require('Module:CanonicalForms').normalize(inputString, mapping)

--

local CanonicalForms = {}

--- Normalize an input string ~~based on a provided mapping table.~~

-- Normalize an input string

-- ~~@param input The input string~~ to ~~normalize.~~

-- Removes wiki markup, converts to lowercase, and maps to canonical form

~~-- @param mappingTable An array of mapping groups.~~

~~-- @return~~ canonical ~~The canonical display value (if a match is found), otherwise the original input.~~

~~-- @return css (optional) The associated extra property (e.g. CSS class) if available.~~

function CanonicalForms.normalize(input, mappingTable)

if not input or input == "" then

return nil, nil

return nil, nil, nil

end

local ~~lowerInput~~ = input:lower()

-- Remove wiki internal link markup (e.g., "[[Brand TLD]]" → "Brand TLD")

for _, group in ipairs(mappingTable) do

local cleanInput = input:gsub("%[%[([^|%]]+)|?[^%]]*%]%]", "%1"):lower()

if group.~~synonyms and type~~(group.synonyms~~) == "table" then~~

for _, syn in ipairs(group.synonyms) do

-- Create lookup table for faster matching (first call only)

~~if lowerInput ==~~ syn:lower() ~~then~~

if not mappingTable._lookupCache then

~~return~~ group~~.canonical, group.css~~

local lookupCache = {}

~~end~~

for _, group in ipairs(mappingTable) do

-- Add the canonical form itself to the lookup (in lowercase)

lookupCache[group.canonical:lower()] = group

-- Add all synonyms to the lookup

for _, syn in ipairs(group.synonyms or {}) do

lookupCache[syn:lower()] = group

end

mappingTable._lookupCache = lookupCache

end

return ~~input~~, nil

-- Direct lookup via cache

local match = mappingTable._lookupCache[cleanInput]

if match then

return match.canonical, match.css, match.category

end

return cleanInput, nil, nil

end

return CanonicalForms

@@ Line 1: / Line 1: @@
--- Module:CanonicalForms
+--[[
--- This module provides generic normalization functions.
+* Name: CanonicalForms
--- Templates can supply their own mapping tables (arrays of mapping groups)
+* Author: Mark W. Datysgeld
--- where each group is a table with:
+* Description: Text normalization utility that removes wiki markup and maps user input to canonical values using configurable lookup tables
---   • canonical: The canonical display value.
+* Notes: Example usage: local mapping = { { canonical = "gTLD", synonyms = {"generic", "g"} }, { canonical = "ccTLD", synonyms = {"country", "cc"} } }; local canonical, css, category = require('Module:CanonicalForms').normalize(inputString, mapping)
---   • synonyms:  A table (array) of synonyms (case insensitive).
+]]
---   • [optional] css: Any extra property (e.g. CSS class) associated with the canonical value.
---
--- The functions in this module will process an input string against the provided mapping table
--- and return the canonical value (and optionally any extra property) if a match is found.
---
--- Usage example:
---   local mapping = {
---     { canonical = "gTLD", synonyms = {"generic", "gtld", "tld"} },
---     { canonical = "ccTLD", synonyms = {"cctld", "country", "cc"} }
---   }
---   local canonical, css = require('Module:CanonicalForms').normalize(inputString, mapping)
---
 local CanonicalForms = {}
---- Normalize an input string based on a provided mapping table.
+-- Normalize an input string
--- @param input The input string to normalize.
+-- Removes wiki markup, converts to lowercase, and maps to canonical form
--- @param mappingTable An array of mapping groups.
--- @return canonical The canonical display value (if a match is found), otherwise the original input.
--- @return css (optional) The associated extra property (e.g. CSS class) if available.
 function CanonicalForms.normalize(input, mappingTable)
      if not input or input == "" then
-         return nil, nil
+         return nil, nil, nil
      end
-     local lowerInput = input:lower()
+    -- Remove wiki internal link markup (e.g., "[[Brand TLD]]" → "Brand TLD")
-     for _, group in ipairs(mappingTable) do
+     local cleanInput = input:gsub("%[%[([^|%]]+)|?[^%]]*%]%]", "%1"):lower()
-        if group.synonyms and type(group.synonyms) == "table" then
-             for _, syn in ipairs(group.synonyms) do
+     -- Create lookup table for faster matching (first call only)
-                 if lowerInput == syn:lower() then
+    if not mappingTable._lookupCache then
-                    return group.canonical, group.css
+        local lookupCache = {}
-                end
+        for _, group in ipairs(mappingTable) do
+            -- Add the canonical form itself to the lookup (in lowercase)
+            lookupCache[group.canonical:lower()] = group
+            -- Add all synonyms to the lookup
+             for _, syn in ipairs(group.synonyms or {}) do
+                 lookupCache[syn:lower()] = group
              end
          end
+        mappingTable._lookupCache = lookupCache
      end
-     return input, nil
+    -- Direct lookup via cache
+    local match = mappingTable._lookupCache[cleanInput]
+    if match then
+        return match.canonical, match.css, match.category
+    end
+     return cleanInput, nil, nil
 end
 return CanonicalForms