Module:Performances
This module is supposed to provide access to the list of performances Angelina did. At the heart of it is a very generic, configurable table-generator, that could also be used in different contexts.
Usage
Overview
{{#invoke: Performances | createTable -- createTable is the generic table generator
|page=Data:Performances.json -- link to page with JSON data
|supplements=Data:VideoMetaData.json;url -- supplemental data sources
|headers=Song,Date,Type,With,Video -- List of titles used in the group section headers
|keys=[[<song>]],{{d|<date>}},<type>,<with>,[<url> <song> - <event>] -- computed data items corresponding to those titles
|sort=<date> -- computed sort keys for sorting outer table
|sort1=<date> -- (optional): outer sortkey, same as inner if omitted
|filters=date:2018,event:Kongsberg -- filters to select what items are displayed
|char_limit=7 -- The amount of characters of the sort key considered for grouping the sections
|char_limit1=4 -- (optional): outer char limit, no outer grouping if ommited
|caption=Kongsberg 2018 by Month -- Title displayed in Header of outer Table
|group_sort=<date> ! <type> ! <pos> -- how the group section sub-tables are sorted
|id=1 -- Id needed if multiple tables are generated within a page
}}
Options
page: Provides page that contains JSON formated input data.
cargo_query: Alternative way to access data via Cargo query (see below for details)
supplements (optional): comma seperated list of supplemental JSON data source entries of "supplementalDataPage;keyName:charLimit;targetKeyName", with charLimit and targetKeyName being optional. Data from the supplemental source will be matched based on the keys (if the target's key name is ommitted, the same will be used). The charLimit option shortens the primary key value to this amount of characters. The target data can also be in a form, where the matching key data is used as keys itself instead of value of a "key" field.
headers: Comma seperated list of titles used in the group section headers.
keys: Comma seperated list of computed fields used for the colums in group sections defined by the corrosponding headers. Number of entries should match the number of headers.
Key names representing data are enclosed in "<",">" brackets. Each field can reference multiple keys. Other text will be still interpreted e.g. as templates. For the key's values, substitions can be defined by appending them with colons. A pair of strings will be interpretated as the search and replacement strings for a gsub() substition with lua pattern matching. The special keywords "toUpper", "toLower", and "toTitleCase" will perform the appropriate conversions, "limit:n" will limit the resresult to n characters. Several substitutions can be concatenated.
sort: Computed sort fields for sorting outer table, and for labeling of section names. Same syntax as for keys.
group (optional): Definition of how to label section names differently from 'sort' definition.
char_limit (optional): The amount of characters of the sort key considered for grouping the sections. This will also affect the displayed section name if 'group' is not provided.
filters: Comma seperated list of filters to select what items are displayed. A filter is described by "keyname:values". Alternatives within an filter value item can be described with "/". Negations can be described with "!( . )".
sort1 (optional): Outer sortkey. If ommited, the inner sort key will be used.
group1 (optional): Definition of how to label outer section names differently from 'sort1' definition.
char_limit1 (optional): The amount of characters of the sort key considered for the outer grouping. This will also affect the displayed section name, if 'group1' is ommited.
If ommited, no outer grouping will be performed.
id: A unique Id for each table is needed if multiple tables are generated within a page, to provide a basis for unique section ids.
caption (optional): Title displayed in the header of the outer table
group_sort: Computed field. that determines how the group section sub-tables are sorted. Same syntax as for keys.
Features
The generated table may have one or two levels of grouping, and up to three levels of sorting. It has the follwing structure, with the outer grouping being optional (click on the [n Items] buttons to reveal the inner collapsed tables):
| Caption (provided as parameter) | |||
|---|---|---|---|
| 1 (outer grouping level seperator) | |||
| A (inner grouping level header) [2 Items] | |||
| B (inner grouping level header) [1 Item] | |||
| 2 (outer grouping level seperator) | |||
| C (inner grouping level header) [1 Item] | |||
The inner tables are enclosed in
<div youtube-player-placeholder> ... </div>
tags, which can optionally be used by Javascript code contained in a Gadget (MediaWiki:Gadget-embeddedYouTubePlayer.js) to attach an embedded YouTube player to each section, that is loaded with a playlist of any YouTube video link contained in that section. Clicking on those links will then also load and start the linked video in that player.
Each grouping separator/header will set an anchor id, so that the table rows can be jumped at with internal references, e.g. C.
There is also code available, that will expand the inner table of a jump target when jumped to.
Use of Cargo queries for input
This was recently added as an alternative source for input. It supports:
- Multiple Cargo tables with configurable fields
- Nested child tables (1–3+ levels)
- List field splitting (# or ,)
- Optional pretty-printed JSON
- Flexible configuration in JSON or Lua-like syntax
The getJSON() function generates nested JSON structures from Cargo tables. It is meant to verify that queries produce the right structures before using them in the createTable() function.
All table and field configuration is contained in a single parameter: cargo_query.
{{#invoke:CargoQueryTest|getJSON
|cargo_query=<JSON or Lua-like table>
|pretty=true
}}- cargo_query – required; configuration of tables, fields, nesting, and root table.
- pretty – optional; if "true", outputs indented JSON (requires mw.text.JSON_PRETTY).
cargo_query Parameters
| Parameter | Type | Description |
|---|---|---|
| tables | array of strings | List of Cargo table names (CamelCase). Order does not determine root. |
| fields | table | Map of tableName → comma-separated list of field names to return. Fields can be lowercase. |
| where | table | Map of tableName → Cargo WHERE clause for filtering rows. Empty string for no filter. |
| nest | array of tables | Each table describes a parent-child nesting rule: { parent = "ParentTable", child = "ChildTable", parentKey = "ParentField", childKey = "ChildField", as = "ChildLabel" } |
| listFields | array of strings | Field names that should be returned as arrays (split on # or ,). |
| root | string | The Cargo table to use as the top-level JSON object. |
Field Splitting Rules
- # delimiter → splits Cargo lists into an array, preserves empty entries.
Example: "Concert##US Concert" → ["Concert","","US Concert"]
- ', delimiter → splits only if there are no spaces around the comma.
Example 1 — Two-level nesting (Performances → Videos)
{{#invoke:CargoQueryTest|getJSON
|cargo_query={
"tables": ["PerformancesDevel","VideosDevel"],
"fields": {
"PerformancesDevel": "song,event,context,date,type,pos,partners,comment,perfID",
"VideosDevel": "perfID,url,duration,quality"
},
"where": {
"PerformancesDevel": "",
"VideosDevel": ""
},
"nest": [
{ "parent": "PerformancesDevel", "child": "VideosDevel", "parentKey": "perfID", "childKey": "perfID", "as": "videos" }
],
"listFields": ["context","partners"],
"root": "PerformancesDevel"
}
|pretty=true
}}Lua-like table style (sanitized)
{{#invoke:CargoQueryTest|getJSON
|cargo_query={
tables = ['PerformancesDevel','VideosDevel'],
fields = {
PerformancesDevel = 'song,event,context,date,type,pos,partners,comment,perfID',
VideosDevel = 'perfID,url,duration,quality'
},
nest = {
{ parent = 'PerformancesDevel', child = 'VideosDevel', parentKey = 'perfID', childKey = 'perfID', as = 'videos' }
},
listFields = ['context','partners'],
root = 'PerformancesDevel'
}
|pretty=true
}}Example 2 — Three-level nesting (Songs → Performances → Videos)
{{#invoke:CargoQueryTest|getJSON
|cargo_query={
tables = ['SongsDevel','PerformancesDevel','VideosDevel'],
fields = {
SongsDevel = 'songID,title,artist',
PerformancesDevel = 'song,event,context,date,type,pos,partners,comment,perfID',
VideosDevel = 'perfID,url,duration,quality'
},
nest = {
{ parent = 'SongsDevel', child = 'PerformancesDevel', parentKey = 'songID', childKey = 'song', as = 'performances' },
{ parent = 'PerformancesDevel', child = 'VideosDevel', parentKey = 'perfID', childKey = 'perfID', as = 'videos' }
},
listFields = ['context','partners'],
root = 'SongsDevel'
}
|pretty=true
}}Tips
- Always specify root in cargo_query to avoid ambiguity.
- Use listFields for any Cargo field that should become a Lua/JSON array.
- pretty=true is optional but recommended for debugging and readability.
- Lua-table style is allowed inline — single quotes ' are fine; the module sanitizes them into valid JSON.
- Multi-level nesting must specify both parent and child explicitly in each rule.
- Use the AJW:Cargo_query_test page to develop your query and check if it delivers the expected JSON equivalent structure.
ToDo
- Make sorting more flexible, e.g. by applying regex substitutions to data values before being sorted.
- Have an option, to not start collapsed (usefull for small tables)
- Implement access to sublists below keys.
- provide default values for empty fields of computed fields, e.g. like <keyName=defaultValue>
- Provide a means to print additional data behind the sort field that does not effect the id-anchor
Examples
Multiple Filters, combined as AND
{{#invoke: Performances | createTable
|page=Data:Performances.json
|headers=Song,Date,Type,With,Video
|keys=[[<song>]],{{d|<date>}},<type>,<with>,[<url> <song> - <event>]
|sort=<date>
|filters=date:2018,event:Kongsberg
|char_limit=7
|caption=Kongsberg 2018 by Month
|group_sort=<date> ! <type> ! <pos>
|id=1
}}
| Kongsberg 2018 by Month | ||||
|---|---|---|---|---|
| 2018-06 [12 Items] | ||||
| 2018-07 [22 Items] | ||||
Single Event with Supplemental Video Metadata
{{#invoke: Performances | createTable
|page=Data:Performances.json
|supplements=Data:VideoMetaData.json;url:43
|headers=Song,Date,Type,Video
|keys=[[<song>]],{{d|<date>}},<type>,[<url> <url-title>; <url-User>]
|sort=<event>
|group_sort=<date> ! <type> ! <pos> ! <song>
|filters=event:Bjerke
|id=2
}}
| Bjerke Lydstudio [8 Items] | |||
|---|---|---|---|
Alternatives in Filters, combinded as OR
{{#invoke: Performances | createTable
|page=Data:Performances.json
|headers=Song,Date,Type,Comment
|keys=[[#<song>|<song>]],{{d|<date>}}: <pos>,<type> <duration> [<url> play],<comment>; <with>
|sort=[[<event>]]
|caption=Repetitive Live Events
|group_sort=<date> ! <type> ! <pos> ! <song>
|filters=event:Allsang på Grensen/TV 2's Artist Gala
|id=3
}}
| Repetitive Live Events |
|---|
Demonstrating Different Outer Sorting/Grouping and Negation in Filters
In the following example, the highest level (sort1) is sorted by year, then the middle one (sort) by date, as well as the inner table one, which also features secondary sorting keys (type, pos, song). The seperator "!" that was chosen for these has no special meaning, but it is the lowest character encoding value above a space, which should ensure that the correct sorting is applied.
"duration:!(fragment)" is used for filtering for those videos, that are not fragments.
Also, wie use substitutions to the computed "url" field
{{#invoke: Performances | createTable
|page=Data:Performances.json
|headers=Song,Date,Type,Comment,Video
|keys=[[#<song>|<song>]],{{d|<date>}}; <pos>,<type> <duration>, <comment>; <with>,[<url> <url:.*www.: :.com.*: >]
|sort=[[<event>]]
|sort1=<date>
|char_limit1=4
|filters=type:live,duration:!(fragment)
|caption=Live Events by Year, excluding fragments
|group_sort=<date> ! <type> ! <pos> ! <song>
|id=4
}}
| Live Events by Year, excluding fragments |
|---|
Demonstrating Same Type Sorting, with Different Grouping Levels, and Supplemental Data
Here, the different char_limits on the sorting of song lead to two different grouping levels, by 1st character and by unique title.
{{AtoZ}}
{{#invoke: Performances | createTable
|page=Data:Performances.json
|supplements=Data:Songs.json;song;title,Data:VideoMetaData.json;url:43 |headers=Event,Date,Type,Video,Pos,With,Comment
|keys=[[#<event>|<event>]],{{d|<date>}},<type> <duration>,[<url> play <song-type>],<pos>,<with>,<comment>; <url-channelName>
|sort=[[<song>]]
|char_limit1=1
|caption=Live Songs, excluding fragments
|group_sort=<date> ! <type> ! <pos>
|filters=duration:!(fragment),type:live
|id=5
}}
{{AtoZ}}
A · B · C · D · E · F · G · H · I · J · K · L · M · N · O · P · Q · R · S · T · U · V · W · X · Y · Z
| Live Songs, excluding fragments |
|---|
Code
local p = {}
-- Load required libraries
local mw_text = mw.text
local mw_title = require('mw.title')
-- Function to fetch JSON from a wiki page
function p.fetchJSONFromPage(pageName)
local title = mw_title.new(pageName)
if not title then
return nil, 'Invalid page name'
end
local content = title:getContent()
if not content then
return nil, 'Page not found or empty'
end
return mw_text.jsonDecode(content)
end
-- Function to compute the value of a field based on a formula
function p.computeField(item, formula)
return (formula:gsub('(%b<>)', function(placeholder)
local key = placeholder:sub(2, -2)
return tostring(item[key] or '')
end))
end
-- Function to filter data based on provided filters
function p.filterData(data, filters)
local filteredData = {}
for _, item in ipairs(data) do
local match = true
for key, pattern in pairs(filters) do
local luaPattern = pattern:gsub('%*', '.*')
if not tostring(item[key]):match(luaPattern) then
match = false
break
end
end
if match then
table.insert(filteredData, item)
end
end
return filteredData
end
-- Function to sort data based on a sorting key formula
function p.sortData(data, sortKeyFormula)
table.sort(data, function(a, b)
local aKey = p.computeField(a, sortKeyFormula):gsub('[^%w%?%!]', ''):lower():gsub('%?',"%#")
local bKey = p.computeField(b, sortKeyFormula):gsub('[^%w%?%!]', ''):lower():gsub('%?',"%#")
return tostring(aKey) < tostring(bKey)
end)
return data
end
-- Function to group data based on a grouping key formula
function p.groupData(data, groupKeyFormula, charLimit)
local groupedData = {}
local currentGroup = nil
for _, item in ipairs(data) do
local groupValue = p.computeField(item, groupKeyFormula)
if charLimit and tonumber(charLimit) then
groupValue = mw.ustring.sub(groupValue:gsub('[%(%)%[%]%{%}]','') , 1, tonumber(charLimit))
end
if groupValue ~= currentGroup then
currentGroup = groupValue
groupedData[#groupedData + 1] = {header = groupValue, items = {}}
end
table.insert(groupedData[#groupedData].items, item)
end
return groupedData
end
-- Function to render table headers
function p.renderHeaders(headers)
return '! ' .. table.concat(headers, ' !! ') .. '\n'
end
-- Function to render table rows
function p.renderRows(items, computedKeys)
local rows = ''
for _, item in ipairs(items) do
rows = rows .. '|-\n'
local row = {}
for _, keyFormula in ipairs(computedKeys) do
row[#row + 1] = p.computeField(item, keyFormula) or ''
end
rows = rows .. '| ' .. table.concat(row, ' || ') .. '\n'
end
return rows
end
-- Function to render the complete table
function p.renderTable(data, headers, computedKeys, sortKeyFormula, charLimit, sortKeyFormula1, charLimit1, groupSortKeyFormula, caption, tableId)
local outerTable = '{| class="wikitable" style="width: 100%; margin: 0;"\n'
outerTable = outerTable .. '|-\n! colspan="' .. #headers .. '" style="text-align:left;" | ' .. caption .. '\n'
local sortedData1 = p.sortData(data, sortKeyFormula1) -- this should be more flexible with different sortKeyForumlar
local groupedData1 = p.groupData(sortedData1, sortKeyFormula1, charLimit1) -- this should be more flexible with flexibile charcount
for groupIndex1, group1 in ipairs(groupedData1) do
local sortedData = p.sortData(group1.items, sortKeyFormula)
local groupedData = p.groupData(sortedData, sortKeyFormula, charLimit)
outerTable = outerTable ..'|- \n| style="text-align:center;" | <span id="' .. group1.header .. '">' .. group1.header .. '</span>\n'
for groupIndex, group in ipairs(groupedData) do
if groupSortKeyFormula then
p.sortData(group.items, groupSortKeyFormula)
end
local groupId = tableId .. '-' .. groupIndex1 .. '-' .. groupIndex
local itemCount = #group.items
local toggleLabel = '[' .. itemCount .. ' Item' .. (itemCount > 1 and "s" or "") .. ']'
local toggleSpan = '<span class="mw-customtoggle-' .. groupId .. '" style="cursor:pointer; color:blue;">' .. toggleLabel .. '</span>'
outerTable = outerTable .. '|-\n! colspan="' .. #headers .. '" style="background-color:#f5f5f5; text-align:left;" | <span id="' .. group.header .. '">' .. group.header .. '</span> ' .. toggleSpan .. '\n'
outerTable = outerTable .. '<tr class="mw-collapsible mw-collapsed" id="mw-customcollapsible-' .. groupId .. '" style="display: none">\n'
outerTable = outerTable .. '| <div class="youtube-player-placeholder">\n' -- this would be the place to possibly reduce the padding for the inner table
outerTable = outerTable .. '{| class="wikitable sortable" style="width: 100%; margin: 0;"\n'
outerTable = outerTable .. '|-\n' .. p.renderHeaders(headers)
outerTable = outerTable .. p.renderRows(group.items, computedKeys)
outerTable = outerTable .. '|}\n</div>\n'
outerTable = outerTable .. '</tr>\n'
end
end
outerTable = outerTable .. '|}\n'
return outerTable
end
-- Main function to create the table
function p.createTable(frame)
local args = frame:getParent().args
for k, v in pairs(frame.args) do
args[k] = v
end
local pageName = args['page']
if not args['headers'] or not args['keys'] or not args['sort'] then
return 'Error: headers, keys, and sort arguments are required.'
end
local headers = mw_text.split(args['headers'], ',')
local computedKeys = mw_text.split(args['keys'], ',')
local sortKeyFormula = args['sort']
local charLimit = args['char_limit']
local sortKeyFormula1 = args['sort1']
local charLimit1 = args['char_limit1']
local groupSortKeyFormula = args['group_sort']
local caption = args['caption'] or ''
local tableID = args['id']
local filters = {}
local filterString = args['filters']
if filterString then
for key, pattern in filterString:gmatch('([^:]+):([^,]+),?') do
filters[key] = pattern
end
end
if not pageName then
return 'Error: page argument is required to fetch JSON data.'
end
local data, err = p.fetchJSONFromPage(pageName)
if not data then
return 'Error fetching JSON from page: ' .. err
end
data = p.filterData(data, filters)
if not sortKeyFormula1 then
sortKeyFormula1 = sortKeyFormula
end
if not charLimit1 then
charLimit1 = 0
end
return p.renderTable(data, headers, computedKeys, sortKeyFormula, charLimit, sortKeyFormula1, charLimit1, groupSortKeyFormula, caption, tableID)
end
return p
