Module:SMWUtils - Revision history

Arclight-MCP-Bot: Fix deduplication: key on printout columns instead of subject (subobject subjects are unique per-input) (via update-page on MediaWiki MCP Server)

2026-02-19T07:29:32Z

Fix deduplication: key on printout columns instead of subject (subobject subjects are unique per-input) (via update-page on MediaWiki MCP Server)

← Older revision		Revision as of 07:29, 19 February 2026
Line 42:		Line 42:
	end		end

	-- Deduplicate rows by ~~subject column~~ (~~row[~~1]), ~~keeping~~ the ~~first occurrence~~		-- Deduplicate rows by printout columns, keeping the first occurrence
	local function ~~deduplicateBySubject~~(rows)		-- Uses stripped text of printout columns (2..n+1) as the dedup key,
			-- since subobject queries produce unique subjects (e.g. Page#sub1, Page#sub2)
			-- but identical printout values for the same parent page
			local function deduplicateRows(rows, numPrintouts)
	local seen = {}		local seen = {}
	local deduped = {}		local deduped = {}
	for _, row in ipairs(rows) do		for _, row in ipairs(rows) do
	local ~~key~~ = stripHtml(row[1]) or ~~row[1] or~~ ''		local keyParts = {}
			for j = 2, numPrintouts + 1 do
			table.insert(keyParts, stripHtml(row[j]) or '')
			end
			local key = table.concat(keyParts, '\|')
	if not seen[key] then		if not seen[key] then
	seen[key] = true		seen[key] = true
Line 191:		Line 198:
	end		end

	-- Deduplicate ~~by subject~~ (multiple subobjects on same page produce duplicate rows)		-- Deduplicate (multiple subobjects on same page produce duplicate rows)
	filtered = ~~deduplicateBySubject~~(filtered)		filtered = deduplicateRows(filtered, #printoutProps)

	-- Build headers list		-- Build headers list
Line 272:		Line 279:
	end		end

	-- Deduplicate ~~by subject~~ (multiple subobjects on same page produce duplicate rows)		-- Deduplicate (multiple subobjects on same page produce duplicate rows)
	filtered = ~~deduplicateBySubject~~(filtered)		filtered = deduplicateRows(filtered, #printoutProps)

	-- Build headers list		-- Build headers list

Arclight-MCP-Bot: Add deduplication by subject column to eliminate duplicate rows from multiple matching subobjects (via update-page on MediaWiki MCP Server)

2026-02-19T07:26:17Z

Add deduplication by subject column to eliminate duplicate rows from multiple matching subobjects (via update-page on MediaWiki MCP Server)

@@ Line 40: / Line 40: @@
      if not s then return nil end
      return trim(s:gsub('<[^>]+>', ''):gsub('&[^;]+;', ''))
 end
@@ Line 176: / Line 190: @@
          end
      end
      -- Build headers list
@@ Line 254: / Line 271: @@
          end
      end
      -- Build headers list

Arclight-MCP-Bot: Add mainlabel support to include subject/page name column in output (via update-page on MediaWiki MCP Server)

2026-02-19T07:10:44Z

Add mainlabel support to include subject/page name column in output (via update-page on MediaWiki MCP Server)

← Older revision		Revision as of 07:10, 19 February 2026
Line 112:		Line 112:
	-- printouts: Comma-separated properties to display		-- printouts: Comma-separated properties to display
	-- headers: Comma-separated header names		-- headers: Comma-separated header names
			-- mainlabel: Optional header for the subject/page name column
	-- format: table, ul, ol (default: table)		-- format: table, ul, ol (default: table)
	-- intro: Custom intro text		-- intro: Custom intro text
Line 122:		Line 123:
	local headers = trim(args.headers) or ''		local headers = trim(args.headers) or ''
	local intro = trim(args.intro)		local intro = trim(args.intro)
			local mainlabel = trim(args.mainlabel)

	if not minProperty then		if not minProperty then
Line 184:		Line 186:

	if format == 'ul' then		if format == 'ul' then
	return p._formatList(filtered, printoutProps, headerList, 'ul', introText)		return p._formatList(filtered, printoutProps, headerList, 'ul', introText, mainlabel)
	elseif format == 'ol' then		elseif format == 'ol' then
	return p._formatList(filtered, printoutProps, headerList, 'ol', introText)		return p._formatList(filtered, printoutProps, headerList, 'ol', introText, mainlabel)
	else		else
	return p._formatTable(filtered, printoutProps, headerList, introText)		return p._formatTable(filtered, printoutProps, headerList, introText, mainlabel)
	end		end
	end		end
Line 201:		Line 203:
	local headers = trim(args.headers) or ''		local headers = trim(args.headers) or ''
	local intro = trim(args.intro)		local intro = trim(args.intro)
			local mainlabel = trim(args.mainlabel)

	if not maxProperty then		if not maxProperty then
Line 261:		Line 264:

	if format == 'ul' then		if format == 'ul' then
	return p._formatList(filtered, printoutProps, headerList, 'ul', introText)		return p._formatList(filtered, printoutProps, headerList, 'ul', introText, mainlabel)
	elseif format == 'ol' then		elseif format == 'ol' then
	return p._formatList(filtered, printoutProps, headerList, 'ol', introText)		return p._formatList(filtered, printoutProps, headerList, 'ol', introText, mainlabel)
	else		else
	return p._formatTable(filtered, printoutProps, headerList, introText)		return p._formatTable(filtered, printoutProps, headerList, introText, mainlabel)
	end		end
	end		end
Line 271:		Line 274:
	-- Format results as a wikitable		-- Format results as a wikitable
	-- Note: row[1] is subject, row[2..n+1] are printouts, row[n+2] is min/max value		-- Note: row[1] is subject, row[2..n+1] are printouts, row[n+2] is min/max value
	function p._formatTable(rows, printouts, headers, introText)		function p._formatTable(rows, printouts, headers, introText, mainlabel)
	local output = {}		local output = {}

Line 279:		Line 282:
	-- Header row		-- Header row
	table.insert(output, '\|-\n! #')		table.insert(output, '\|-\n! #')
			if mainlabel then
			table.insert(output, ' !! ' .. mainlabel)
			end
	for i, prop in ipairs(printouts) do		for i, prop in ipairs(printouts) do
	local headerText = headers[i] or prop:gsub('.*%.', ''):gsub('Has ', '')		local headerText = headers[i] or prop:gsub('.*%.', ''):gsub('Has ', '')
Line 288:		Line 294:
	for i, row in ipairs(rows) do		for i, row in ipairs(rows) do
	local tr = '\|-\n\| ' .. tostring(i)		local tr = '\|-\n\| ' .. tostring(i)
			if mainlabel then
			tr = tr .. ' \|\| ' .. (row[1] or '')
			end
	for j = 1, #printouts do		for j = 1, #printouts do
	-- Printout j is in column j+1 (offset by subject column)		-- Printout j is in column j+1 (offset by subject column)
Line 301:		Line 310:

	-- Format results as a list		-- Format results as a list
	function p._formatList(rows, printouts, headers, listType, introText)		function p._formatList(rows, printouts, headers, listType, introText, mainlabel)
	local output = {}		local output = {}
	local bullet = listType == 'ol' and '#' or '*'		local bullet = listType == 'ol' and '#' or '*'
Line 309:		Line 318:
	for _, row in ipairs(rows) do		for _, row in ipairs(rows) do
	local parts = {}		local parts = {}
			if mainlabel then
			local subject = row[1]
			if subject and subject ~= '' then
			table.insert(parts, subject)
			end
			end
	for j = 1, #printouts do		for j = 1, #printouts do
	-- Printout j is in column j+1 (offset by subject column)		-- Printout j is in column j+1 (offset by subject column)

Arclight-MCP-Bot: Fix column index calculation - account for subject column in broadtable format (via update-page on MediaWiki MCP Server)

2026-01-15T07:10:24Z

Fix column index calculation - account for subject column in broadtable format (via update-page on MediaWiki MCP Server)

← Older revision		Revision as of 07:10, 15 January 2026
Line 14:		Line 14:

	-- Parse an HTML table from #ask broadtable format into rows of cells		-- Parse an HTML table from #ask broadtable format into rows of cells
			-- Filters out the "further results" row
	local function parseHtmlTable(html)		local function parseHtmlTable(html)
	local rows = {}		local rows = {}
Line 19:		Line 20:

	for rowContent in html:gmatch('<tr[^>]*>(.-)</tr>') do		for rowContent in html:gmatch('<tr[^>]*>(.-)</tr>') do
	local cells = {}		-- Skip "further results" rows
	for cellContent in rowContent:gmatch('<td[^>]*>(.-)</td>') do		if not rowContent:match('smw%-broadtable%-furtherresults') then
	-- Keep HTML intact for links		local cells = {}
	table.insert(cells, trim(cellContent) or '')		for cellContent in rowContent:gmatch('<td[^>]*>(.-)</td>') do
	end		-- Keep HTML intact for links
	if #cells > 0 then		table.insert(cells, trim(cellContent) or '')
	table.insert(rows, cells)		end
			if #cells > 0 then
			table.insert(rows, cells)
			end
	end		end
	end		end
Line 85:		Line 89:
	end		end

	-- Show what we'd calculate as min		-- Show what we'd calculate as min (corrected: +2 for subject column)
	if #rows > 0 then		if #rows > 0 then
	local minValueCol = #printoutProps + 1		local minValueCol = #printoutProps + 2 -- +1 for subject column, +1 for minProperty column
	table.insert(output, "'''Min value column index:''' " .. minValueCol .. "\n")		table.insert(output, "'''Printout count:''' " .. #printoutProps .. "\n")
			table.insert(output, "'''Min value column index:''' " .. minValueCol .. " (printouts + 2 for subject col)\n")
	if rows[1][minValueCol] then		if rows[1][minValueCol] then
	local minText = stripHtml(rows[1][minValueCol])		local minText = stripHtml(rows[1][minValueCol])
Line 151:		Line 156:
	end		end

	-- ~~The~~ minProperty ~~value is in the last column (we added it with =#)~~		-- broadtable format: Column 1 = subject, Columns 2..n+1 = printouts, Column n+2 = minProperty
	-- ~~Find the minimum value from the first row (already sorted asc)~~		-- So minValueCol = #printoutProps + 2
	local minValueCol = #printoutProps + 1		local minValueCol = #printoutProps + 2
	local minValueText = stripHtml(rows[1][minValueCol])		local minValueText = stripHtml(rows[1][minValueCol])
	local minValue = tonumber(minValueText)		local minValue = tonumber(minValueText)
Line 229:		Line 234:
	end		end

	local maxValueCol = #printoutProps + 1		-- broadtable format: Column 1 = subject, Columns 2..n+1 = printouts, Column n+2 = maxProperty
			local maxValueCol = #printoutProps + 2
	local maxValueText = stripHtml(rows[1][maxValueCol])		local maxValueText = stripHtml(rows[1][maxValueCol])
	local maxValue = tonumber(maxValueText)		local maxValue = tonumber(maxValueText)
Line 264:		Line 270:

	-- Format results as a wikitable		-- Format results as a wikitable
			-- Note: row[1] is subject, row[2..n+1] are printouts, row[n+2] is min/max value
	function p._formatTable(rows, printouts, headers, introText)		function p._formatTable(rows, printouts, headers, introText)
	local output = {}		local output = {}
Line 278:		Line 285:
	table.insert(output, '\n')		table.insert(output, '\n')

	-- Data rows ~~(only show~~ printout columns~~, not the min/max value~~ column)		-- Data rows: printout columns start at index 2 (after subject column)
	for i, row in ipairs(rows) do		for i, row in ipairs(rows) do
	local tr = '\|-\n\| ' .. tostring(i)		local tr = '\|-\n\| ' .. tostring(i)
	for j = 1, #printouts do		for j = 1, #printouts do
	tr = tr .. ' \|\| ' .. (row[j] or '')		-- Printout j is in column j+1 (offset by subject column)
			tr = tr .. ' \|\| ' .. (row[j + 1] or '')
	end		end
	table.insert(output, tr .. '\n')		table.insert(output, tr .. '\n')
Line 302:		Line 310:
	local parts = {}		local parts = {}
	for j = 1, #printouts do		for j = 1, #printouts do
	if row[j] and ~~row[j]~~ ~= '' then		-- Printout j is in column j+1 (offset by subject column)
	table.insert(parts, ~~row[j]~~)		local cell = row[j + 1]
			if cell and cell ~= '' then
			table.insert(parts, cell)
	end		end
	end		end

Arclight-MCP-Bot: Add debug function to diagnose parsing issues (via update-page on MediaWiki MCP Server)

2026-01-15T07:08:58Z

Add debug function to diagnose parsing issues (via update-page on MediaWiki MCP Server)

@@ Line 35: / Line 35: @@
 local function stripHtml(s)
      if not s then return nil end
-     return s:gsub('<[^>]+>', ''):gsub('&[^;]+;', '')
+     return trim(s:gsub('<[^>]+>', ''):gsub('&[^;]+;', ''))
 end
@@ Line 63: / Line 126: @@
      local printoutQuery = {}
      for prop in printouts:gmatch('[^,]+') do
-         local p = trim(prop)
+         local propTrimmed = trim(prop)
-         table.insert(printoutProps, p)
+         table.insert(printoutProps, propTrimmed)
-         table.insert(printoutQuery, '|?' .. p)
+         table.insert(printoutQuery, '|?' .. propTrimmed)
      end
      -- Add minProperty to query
@@ Line 95: / Line 158: @@
      if not minValue then
-         return "Could not determine minimum value."
+         return "Could not determine minimum value. Column " .. minValueCol .. ", raw: " .. tostring(rows[1][minValueCol]) .. ", stripped: " .. tostring(minValueText)
      end
@@ Line 142: / Line 205: @@
      local printoutQuery = {}
      for prop in printouts:gmatch('[^,]+') do
-         local p = trim(prop)
+         local propTrimmed = trim(prop)
-         table.insert(printoutProps, p)
+         table.insert(printoutProps, propTrimmed)
-         table.insert(printoutQuery, '|?' .. p)
+         table.insert(printoutQuery, '|?' .. propTrimmed)
      end
      table.insert(printoutQuery, '|?' .. maxProperty .. '=#')
@@ Line 171: / Line 234: @@
      if not maxValue then
-         return "Could not determine maximum value."
+         return "Could not determine maximum value. Column " .. maxValueCol .. ", raw: " .. tostring(rows[1][maxValueCol]) .. ", stripped: " .. tostring(maxValueText)
      end

Arclight-MCP-Bot: Rewrite to use frame:preprocess() with #ask queries since mw.smw.ask() doesn't support property chains (via update-page on MediaWiki MCP Server)

2026-01-15T07:07:02Z

Rewrite to use frame:preprocess() with #ask queries since mw.smw.ask() doesn't support property chains (via update-page on MediaWiki MCP Server)

Show changes

Arclight-MCP-Bot: Add debug function and improve property chain handling (via update-page on MediaWiki MCP Server)

2026-01-15T07:05:14Z

Add debug function and improve property chain handling (via update-page on MediaWiki MCP Server)

Show changes

Arclight-MCP-Bot: Rewrite to use mw.smw.ask() now that Semantic Scribunto is installed (via update-page on MediaWiki MCP Server)

2026-01-15T07:03:27Z

Rewrite to use mw.smw.ask() now that Semantic Scribunto is installed (via update-page on MediaWiki MCP Server)

Show changes

Arclight-MCP-Bot: Rewrite to use frame:preprocess() instead of mw.smw.ask() (Semantic Scribunto not installed) (via update-page on MediaWiki MCP Server)

2026-01-15T07:01:09Z

Rewrite to use frame:preprocess() instead of mw.smw.ask() (Semantic Scribunto not installed) (via update-page on MediaWiki MCP Server)

Show changes

Arclight: Arclight changed the content model of the page Module:SMWUtils from "wikitext" to "Scribunto module"

2026-01-15T07:00:05Z

Arclight changed the content model of the page Module:SMWUtils from "wikitext" to "Scribunto module"

← Older revision	Revision as of 07:00, 15 January 2026
(No difference)