View source for Help:Extension:ParserFunctions

<noinclude>
<languages/>
</noinclude>
{{PD Help Page}}
{{Magic words|general=1}}
<translate><!--T:1--> The <tvar name=1>{{ll|Extension:ParserFunctions|nsp=0}}</tvar> extension provides additional [[<tvar name=3>Special:MyLanguage/Parser functions</tvar>|parser functions]] to supplement those already present in MediaWiki core.</translate>
<translate><!--T:2769--> (See <tvar name=1>{{ll|help:magic words#parser functions}}</tvar>.)</translate>
<translate><!--T:2447--> (It ''may'' be configured to provide additional parser functions for string handling; these are only available to non-Wikimedia Foundation wikis hence these string functions are documented {{<tvar name=ext-sf>ll|Extension:ParserFunctions/String functions</tvar>|elsewhere}}.)</translate>
<translate><!--T:2448--> All the parser functions provided by this extension take the form:</translate>
:<code><nowiki>{{</nowiki><translate><!--T:2--> '''#functionname''': ''argument 1'' | ''argument 2'' | ''argument 3'' ... </translate><nowiki>}}</nowiki></code>

== #expr ==
{{for|1=<translate><!--T:2711--> a more '''in-depth manual''' on the finer points of how the expression evaluator works, including some additional operators not covered here</translate>|2=Special:MyLanguage/Manual:Expr parser function syntax{{!}}Manual:Expr parser function syntax}}
{| class="wikitable" style="float:{{dir|{{pagelang}}|left|right}}; width:30%; margin-{{dir|{{pagelang}}|right|left}}:1em"
|-
! {{hl1}} | <translate><!--T:3--> Type</translate>
! {{hl1}} | <translate><!--T:4--> Operators</translate>
|-
| {{hl2}} | <translate><!--T:5--> Grouping (parentheses)</translate>
| style="word-spacing: 0.5em; font-weight: bold;" | <code>( )</code>
|-
| {{hl2}} | <translate><!--T:6--> Numbers</translate>
| <code>'''1234.5'''</code> &nbsp; <code>'''e'''</code>&nbsp;(2.718) &nbsp; <code>'''pi'''</code>&nbsp;(3.142)
|-
| {{hl2}} |
| <translate><!--T:7--> binary operator</translate> <code>'''e'''</code> &nbsp; <translate><!--T:8--> unary</translate> <code>'''+'''</code>,<code>'''-'''</code>
|-
| {{hl2}} | <translate><!--T:9--> Unary</translate>
| style="word-spacing: 0.5em; font-weight: bold;" | <code>not ceil trunc floor abs exp ln sin cos tan acos asin atan sqrt</code>
|-
| {{hl2}} rowspan="3" | <translate><!--T:10--> Binary</translate>
| style="word-spacing: 0.5em; font-weight: bold;" | <code>^</code>
|-
| style="word-spacing: 0.5em; font-weight: bold;" | <code>* / div mod fmod</code>
|-
| style="word-spacing: 0.5em; font-weight: bold;" | <code>+ -</code>
|-
| {{hl2}} | <translate><!--T:11--> Round</translate>
| style="word-spacing: 0.5em; font-weight: bold;" | <code>round</code>
|-
| {{hl2}} rowspan="3" | <translate><!--T:12--> Logic</translate>
| style="word-spacing: 0.5em; font-weight: bold;" | <code>= != <> > < >= <=</code>
|-
| style="word-spacing: 0.5em; font-weight: bold;" | <code>and</code>
|-
| style="word-spacing: 0.5em; font-weight: bold;" | <code>or</code>
|}

<translate><!--T:13--> This function evaluates a mathematical expression and returns the calculated value.</translate>
<translate><!--T:2418--> This function is also available in <tvar name=1>{{ll|Extension:Scribunto|nsp=0}}</tvar> via the <tvar name=expr><code>mw.ext.ParserFunctions.expr</code></tvar> function.</translate>

:<code><nowiki>{{#expr:</nowiki> ''<translate><!--T:14--> expression</translate>'' <nowiki>}}</nowiki></code>

<translate><!--T:2747--> Basic example</translate>
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 + 1 }}</syntaxhighlight> → '''{{#expr: 1 + 1 }}'''

<translate><!--T:15--> The available operators are listed to the right, in order of precedence. See <tvar name=1>{{ll|Manual:Expr parser function syntax}}</tvar> for more details of the function of each operator.</translate>
<translate><!--T:2710--> The accuracy and format of the result returned will vary depending on the operating system of the server running the wiki and the number format of the site language.</translate>

<translate>
<!--T:16-->
When evaluating using [[w:boolean algebra|boolean algebra]], zero evaluates to <tvar name=1>{{phpi|false}}</tvar>, and any nonzero value, positive or negative, evaluates to <tvar name=2>{{phpi|true}}</tvar>:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 and -1 }}</syntaxhighlight> → '''{{#expr: 1 and -1 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 and 0 }}</syntaxhighlight> → '''{{#expr: 1 and 0 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 or -1 }}</syntaxhighlight> → '''{{#expr: 1 or -1 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: -1 or 0 }}</syntaxhighlight> → '''{{#expr: -1 or 0 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 0 or 0 }}</syntaxhighlight> → '''{{#expr: 0 or 0 }}'''

<translate>
<!--T:17-->
An empty input expression returns an empty string. Invalid expressions return one of several error messages, which can be caught using the <tvar name=iferror><code>[[##iferror|#iferror]]</code></tvar> function:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#expr: }}</syntaxhighlight> → ''' {{#expr: }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1+ }}</syntaxhighlight> → '''{{#expr: 1+ }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 = }}</syntaxhighlight> → '''{{#expr: 1 = }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: 1 foo 2 }}</syntaxhighlight> → '''{{#expr: 1 foo 2 }}'''

<translate>
<!--T:18-->
The order of addition and subtraction operands before or after a number is meaningful and may be treated as a positive or negative value instead of as an operand with an erroneous input:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#expr: +1 }}</syntaxhighlight> → '''{{#expr: +1 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: -1 }}</syntaxhighlight> → '''{{#expr: -1 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: + 1 }}</syntaxhighlight> → '''{{#expr: + 1 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: - 1 }}</syntaxhighlight> → '''{{#expr: - 1 }}'''

<translate><!--T:19--> Note, if using the output of magic words, you must raw-format them in order to remove commas and translate the numerals.</translate>
<translate><!--T:2674--> For example, <tvar name=1><nowiki>{{NUMBEROFUSERS}}</nowiki></tvar> results in <tvar name=2>{{NUMBEROFUSERS}}</tvar>, where we want <tvar name=3>{{formatnum:{{NUMBEROFUSERS}}|R}}</tvar>, which can be obtained using <tvar name=4><code><nowiki>{{</nowiki>{{ll|Help:Magic words#Formatting|formatnum}}<nowiki>:{{NUMBEROFUSERS}}|R}}</nowiki></code></tvar>.</translate>
<translate><!--T:2675--> This is especially important in some languages, where numerals are translated.</translate>
<translate><!--T:2676--> For example, in Bengali, <tvar name=1><nowiki>{{NUMBEROFUSERS}}</nowiki></tvar> produces <tvar name=bengali>৩০,০৬১</tvar>.</translate>

:<syntaxhighlight lang="wikitext" inline>{{#expr:{{NUMBEROFUSERS}}+100}}</syntaxhighlight> → ''' {{#expr:{{NUMBEROFUSERS}}+100}}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr:{{formatnum:{{NUMBEROFUSERS}}|R}}+100}}</syntaxhighlight> → '''{{#expr:{{formatnum:{{NUMBEROFUSERS}}|R}}+100}}'''

{{Warning|1=<translate><!--T:20--> The operator <tvar name=mod><code>mod</code></tvar> gives wrong results for some values of the second argument:</translate>
<!-- the following line now produces the correct value:
:<syntaxhighlight lang="wikitext" inline>{{#expr: 123 mod (2^32-1)}}</syntaxhighlight> → '''{{#expr: 123 mod (2^32-1)}}''' (should be 123)<br />
but see this one: -->
:<syntaxhighlight lang="wikitext" inline>{{#expr: 123 mod (2^64-1)}}</syntaxhighlight> → {{#expr: 123 mod (2^64-1)}} <translate><!--T:21--> (produces an empty string; should be 123)</translate><br />
<!-- and the following bug has now been solved, as shown on the linked bug report:
Depending on the specification and configuration of the server running the wiki, there may also be other errors:
:<syntaxhighlight lang="wikitext" inline>{{#expr: 20060618093259 mod 10000}}</syntaxhighlight> → '''{{#expr: 20060618093259 mod 10000}}''' in most cases, but may occasionally give '''-6357'''. See {{bugzilla|6356}}.
-->
}}
{{tip|1=<translate><!--T:84--> If you want to do calculations based on dates (ex. test whether current date and time is after some other date and time), first convert the time to number of seconds after January 1, 1970 (UTC) using <tvar name=time><nowiki>{{#time: xNU }}</nowiki></tvar>, then you can simply add and subtract dates as numbers.</translate>}}

<translate>
=== Rounding === <!--T:22-->

<!--T:23-->
[<tvar name=php>https://php.net/function.round</tvar> Rounds] off the number on the left to a multiple of 1/10 raised to a power, with the exponent equal to the truncated value of the number given on the right.

<!--T:24-->
To round up or down use unary <tvar name=ceil><code>ceil</code></tvar> or <tvar name=floor><code>floor</code></tvar> respectively.
</translate>

{| class="wikitable"
! {{hl1}} | <translate><!--T:25--> Test case</translate>
! {{hl1}} | <translate><!--T:26--> Result</translate>
! {{hl1}} | <translate><!--T:27--> Method of rounding</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1/3 round 5 }}</nowiki></code> || {{#expr: 1/3 round 5 }} || <translate><!--T:28--> Final digit is < 5, so no apparent rounding occurs</translate> (0.333333… → 0.33333)
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1/6 round 5 }}</nowiki></code> || {{#expr: 1/6 round 5 }} || <translate><!--T:29--> Final digit is ≥ 5, so it is rounded up</translate> (0.166666… → 0.16667)
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 8.99999/9 round 5 }}</nowiki></code> || {{#expr: 8.99999/9 round 5 }} || <translate><!--T:30--> Again, the result is rounded up on the last digit, which results in additional rounding</translate> (0.999998… → 1.00000 → 1)
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1234.5678 '''round -2''' }}</nowiki></code> || {{#expr: 1234.567 round -2 }} || <translate><!--T:31--> Rounded to nearest 100 because negative values round to the left of the decimal point</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1234.5678 '''round 2''' }}</nowiki></code> || {{#expr: 1234.567 round 2 }} || <translate><!--T:32--> Rounded to nearest 100th because positive values round to the right of the decimal point</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1234.5678 round 2'''.3''' }}</nowiki></code> || {{#expr: 1234.567 round 2.3 }} || <translate><!--T:33--> Decimals in the rounding index make no difference in the rounded result</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''trunc''' 1234.5678 }}</nowiki></code> || {{#expr: trunc 1234.567 }} || <translate><!--T:34--> Decimal portion truncated (chopped off)</translate>
|-
! colspan=3 {{hl2}} | <translate><!--T:35--> Rounding to the nearest integer</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1/3 '''round 0''' }}</nowiki></code> || {{#expr: 1/3 round 0 }} || <translate><!--T:36--> Down to the ''nearest'' integer, which is zero</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1/2 '''round 0''' }}</nowiki></code> || {{#expr: 1/2 round 0 }} || <translate><!--T:37--> Up to the nearest integer, which is one</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 3/4 '''round 0''' }}</nowiki></code> || {{#expr: 3/4 round 0 }} || <translate><!--T:38--> Up to the nearest integer, which is one</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: -1/3 '''round 0''' }}</nowiki></code> || {{#expr: -1/3 round 0 }} || <translate><!--T:39--> Up to the nearest integer, which is zero</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: -1/2 '''round 0''' }}</nowiki></code> || {{#expr: -1/2 round 0 }} || <translate><!--T:41--> Down to the nearest integer, which is negative one</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: -3/4 '''round 0''' }}</nowiki></code> || {{#expr: -3/4 round 0 }} || <translate><!--T:42--> Down to the nearest integer, which is negative one</translate>
|-
! colspan=3 {{hl2}} | <translate><!--T:43--> Rounding up or down with ''ceil'' and ''floor''</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''ceil('''1/3''')''' }}</nowiki></code> || {{#expr: ceil(1/3) }} || <translate><!--T:44--> Up to the next ''larger'' integer, which is one</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''floor('''1/3''')''' }}</nowiki></code> || {{#expr: floor(1/3) }} || <translate><!--T:45--> Down to the next ''smaller'' integer, which is zero</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''ceil('''-1/3''')''' }}</nowiki></code> || {{#expr: ceil(-1/3) }} || <translate><!--T:46--> Up to the next larger integer, which is zero</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''floor('''-1/3''')''' }}</nowiki></code> || {{#expr: floor(-1/3) }} || <translate><!--T:48--> Down to the next smaller integer, which is negative one</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: '''ceil''' 1/3 }}</nowiki></code> || {{#expr: ceil 1/3 }} || <translate><!--T:49--> Not rounded, since 1 already is an integer</translate><br />{{Warning|1=<translate><!--T:50--> Interpreted as <tvar name=ceil1>'''(ceil''' 1''')'''/3</tvar>, not <tvar name=ceil2>'''ceil('''1/3''')'''</tvar>, as you might expect</translate>}}
|-
! colspan=3 {{hl2}} | <translate><!--T:2770--> Rounding large numbers</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1e-92 round 400 }}</nowiki></code> || {{#expr:1e-92 round 400}}
| <translate><!--T:2739--> Rounding to a very large number leads to infinity.</translate> <translate><!--T:2740--> Hence, the original value without the infinity is given as the answer.</translate>
|-
| <code style="white-space:nowrap;"><nowiki>{{#expr: 1e108 round 200 }}</nowiki></code> || {{#expr:1e108 round 200}}
| <translate><!--T:2741--> Same as above.</translate>
|}

<translate>
=== Strings === <!--T:2419-->

<!--T:2420-->
Expressions only work with number-like values, they cannot compare strings or characters. <tvar name=ifeq>[[#ifeq]]</tvar> can be used instead.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#expr: "a" = "a" }}</syntaxhighlight> → '''{{#expr: "a" = "a" }}'''
:<syntaxhighlight lang="wikitext" inline>{{#expr: a = a }}</syntaxhighlight> → '''{{#expr: a = a }}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: a | a | 1 | 0 }}</syntaxhighlight> → '''{{#ifeq: a | a | 1 | 0 }}'''

== #if ==
<translate>
<!--T:51-->
This function evaluates a test string and determines whether or not it is empty. A test string containing only white space is considered to be empty.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#if: <translate nowrap><!--T:52--> test string</translate> | <translate nowrap><!--T:53--> value if test string is not empty</translate> | <translate nowrap><!--T:54--> value if test string is empty (or only white space)</translate>}}</syntaxhighlight>

:<syntaxhighlight lang="wikitext" inline>{{#if: <translate nowrap><!--T:55--> first parameter</translate> | <translate nowrap><!--T:56--> second parameter</translate> | <translate nowrap><!--T:57--> third parameter</translate> }}</syntaxhighlight>

<translate>
<!--T:58-->
This function first tests whether the first parameter is not empty. If the first parameter is not empty, the function displays the second argument. If the first parameter is empty or contains only whitespace characters (spaces, newlines, etc.) it displays the third argument.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#if: | yes | no}}</syntaxhighlight> → '''{{#if: | yes | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: string | yes | no}}</syntaxhighlight> → '''{{#if: string | yes | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: &nbsp;&nbsp;&nbsp;&nbsp; | yes | no}}</syntaxhighlight> → '''{{#if: &nbsp;&nbsp;&nbsp;&nbsp; | yes | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: | yes | no}}</syntaxhighlight> → '''{{#if: | yes | no}}'''

<translate>
<!--T:59-->
The test string is always interpreted as pure text, so mathematical expressions are not evaluated (see <tvar name=1>[[##ifexpr|#ifexpr]]</tvar> for that):
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#if: 1==2 | yes | no }}</syntaxhighlight> → '''{{#if: 1==2 | yes | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: 0 | yes | no }}</syntaxhighlight> → '''{{#if: 0 | yes | no }}'''

<translate>
<!--T:60-->
The last parameter (false) may be omitted:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#if: foo | yes }}</syntaxhighlight> → ''' {{#if: foo | yes }}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: | yes }}</syntaxhighlight> → ''' {{#if: | yes }}'''
:<syntaxhighlight lang="wikitext" inline>{{#if: foo | | no}}</syntaxhighlight> → ''' {{#if: foo | | no}}'''

<translate><!--T:61--> The function may be nested. To do so, nest the inner <tvar name=1><code>#if</code></tvar> function in its full form in place of a parameter of the enclosing <tvar name=1><code>#if</code></tvar> function. Up to seven levels of nesting is possible, although that may depend on the wiki or a memory limit.</translate>
<!--
Here's a nested #if that returns "wowzers" if <nowiki>{{{1}}}</nowiki> is supplied, and "zowzers" if <nowiki>{{{2}}}</nowiki> is supplied (hence "wowzers zowzers" if both are supplied).
In this example, the code is indented; unindented markup for nested parserfunctions quickly becomes unreadable.
Note that a HTML comment is used here, which allows a line break in the source code without including it in the output.
-->

<syntaxhighlight lang="wikitext">
{{#if:<translate nowrap><!--T:2661--> test string</translate>
| <translate nowrap><!--T:2662--> value if test string is not empty</translate>
|{{#if:<translate nowrap><!--T:2689--> test string</translate>
  |<translate nowrap><!--T:2690--> value if test string is not empty</translate>
  |<translate nowrap><!--T:2663--> value if test string is empty (or only white space)</translate>
  }}
}}
</syntaxhighlight>

<translate><!--T:2373--> You can also use a parameter as the test string in your <tvar name=2><code>#if</code></tvar> statement.</translate>
<translate><!--T:2775--> You need to ensure you add the <tvar name=1><code>|</code></tvar> (pipe symbol) after the name of the variable.</translate>
<translate><!--T:2463--> (So that if the parameter does not have a value, it evaluates to an empty string instead of the string "<tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar>".)</translate>

:{{tmpl|0=<code class="mw-highlight"><!--
-->{{((}}#if:{{#tag:span|{{(((}}|class=nb}}{{#tag:span|1|class=cp}}{{red|{{!}}}}{{#tag:span|{{)))}}|class=nb}}{{!}}$1{{!}}$2{{))}}<!--
--></code>
|1=<translate><!--T:2374--> You entered text in variable 1</translate>
|2=<translate><!--T:2655--> There is no text in variable 1</translate>
}}

<translate>
<!--T:62-->
See <tvar name=1>{{ll|Help:Parser functions in templates}}</tvar> for more examples of this parser function.
</translate>

{{Anchor|ifeq|IFEQ|Ifeq|IfEq|If equal}}

== #ifeq ==
<translate><!--T:63--> This parser function compares two input strings, determines whether they are identical, and returns one of two strings based on the result.</translate>
<translate><!--T:2558--> If more comparisons and output strings are required, consider using <tvar name=switch><code>#switch</code></tvar>.</translate>

:<code><nowiki>{{#ifeq:</nowiki> ''string 1'' <nowiki>|</nowiki> ''string 2'' <nowiki>|</nowiki> ''value if identical'' <nowiki>|</nowiki> ''value if different'' <nowiki>}}</nowiki></code>

<translate>
<!--T:64-->
If both strings are valid numerical values, the strings are compared numerically:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: 01 | 1 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: 01 | 1 | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: 0 | -0 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: 0 | -0 | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: 1e3 | 1000 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: 1e3 | 1000 | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: {{#expr:10^3}} | 1000 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: {{#expr:10^3}} | 1000 | equal | not equal}}'''

<translate>
<!--T:65-->
Otherwise, the comparison is made as text; this comparison is case-sensitive:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: foo | bar | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: foo | bar | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: foo | Foo | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: foo | Foo | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: "01" | "1" | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: "01" | "1" | equal | not equal}}''' &nbsp;''<translate><!--T:66--> (compare to similar example above, without the quotes)</translate>''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: 10^3 | 1000 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: 10^3 | 1000 | equal | not equal}}''' &nbsp;''<translate><!--T:67--> (compare to similar example above, with <tvar name=expr><code>[[##expr|#expr]]</code></tvar> returning a valid number first)</translate>''

<translate><!--T:68--> As a practical example, consider an existing {{<tvar name=help-templates>ll|Help:Templates</tvar>|template}} <tvar name=1><code>Template:Timer</code></tvar> using the parser to choose between two standard times, short and long.</translate>
<translate><!--T:2559--> It takes the parameter as the first input to compare against the string "short" &ndash; there is no convention for the order, but it is simpler to read if the parameter goes first.</translate>
<translate><!--T:2560--> The template code is defined as:</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: {{{1|}}} | short | 20 | 40 }}</syntaxhighlight>

<translate>
<!--T:69-->
the following ensue:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{timer|short}}</syntaxhighlight> → '''20'''
:<syntaxhighlight lang="wikitext" inline>{{timer|20}}</syntaxhighlight> → '''40'''
:<syntaxhighlight lang="wikitext" inline>{{timer}}</syntaxhighlight> → '''40'''

{{Warning|1=<translate><!--T:73--> When used inside a parser function, any parser tags and other parser functions must be temporarily replaced with {{<tvar name=1>ll|Strip marker</tvar>|a unique code}}. This affects comparisons:</translate>

:<code><nowiki>{{#ifeq: &lt;nowiki>foo&lt;/nowiki> | &lt;nowiki>foo&lt;/nowiki> | equal | not equal}}</nowiki></code> → '''{{#ifeq: <nowiki>foo</nowiki> | <nowiki>foo</nowiki> | equal | not equal}} '''
:<code><nowiki>{{#ifeq: &lt;math>foo&lt;/math> | &lt;math>foo&lt;/math> | equal | not equal}}</nowiki></code> → '''{{#ifeq: <math>foo</math> | <math>foo</math> | equal | not equal}} '''
:<code><nowiki>{{#ifeq: {{</nowiki>[[Help:Magic words#Miscellaneous|#tag:]]<nowiki>math|foo}} | {{#tag:math|foo}} | equal | not equal}}</nowiki></code> → '''{{#ifeq: {{#tag:math|foo}} | {{#tag:math|foo}} | equal | not equal}} '''
:<syntaxhighlight lang="wikitext" inline>{{#ifeq: [[foo]] | [[foo]] | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: [[foo]] | [[foo]] | equal | not equal}} '''

<translate>
<!--T:74-->
If the strings to be compared are given as equal calls to the same {{<tvar name=templates>ll|Help:Templates</tvar>|template}} containing such tags, then the condition is true, but in the case of two templates with identical content containing such tags it is false.</translate>
}}

{{Warning|1=<translate><!--T:2413--> Literal comparisons to {{<tvar name=magic>ll|Help:Magic words#Page names</tvar>|page-name magic words}} may fail depending on site configuration.</translate> <translate><!--T:2421--> For example,  <tvar name=fullpagename><nowiki>{{FULLPAGENAME}}</nowiki></tvar>, depending on wiki, may capitalize the first letter, and will replace all underscores with spaces.</translate>

<translate>
<!--T:2572-->
To work around this, apply the magic word to both parameters:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: {{FULLPAGENAME: L'Aquila}} | {{FULLPAGENAME}} | equal | not equal}}</syntaxhighlight> → '''equal'''
}}

== #iferror ==

<translate>
<!--T:75-->
This function takes an input string and returns one of two results; the function evaluates to <tvar name=true><code>true</code></tvar> if the input string contains an HTML object with <tvar name=error><code>class="error"</code></tvar>, as generated by other parser functions such as <tvar name=expr><code>[[##expr|#expr]]</code></tvar>, <tvar name=time><code>[[##time|#time]]</code></tvar> and <tvar name=rel2abs><code>[[##rel2abs|#rel2abs]]</code></tvar>, [[<tvar name=templates>Special:MyLanguage/Help:Templates</tvar>|template]] errors such as loops and recursions, and other "failsoft" parser errors.
</translate>

:<code><nowiki>{{#iferror:</nowiki> ''test string'' <nowiki>|</nowiki> ''value if error'' <nowiki>|</nowiki> ''value if correct'' <nowiki>}}</nowiki></code>

<translate><!--T:76--> One or both of the return strings can be omitted.</translate>
<translate><!--T:2771--> If the <tvar name=1><code>correct</code></tvar> string is omitted, the <tvar name=2><code>test string</code></tvar> is returned if it is not erroneous.</translate>
<translate><!--T:2772--> If the <tvar name=3><code>error</code></tvar> string is also omitted, an empty string is returned on an error:</translate>

:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + 2 }} | error | correct }}</syntaxhighlight> → '''{{#iferror: {{#expr: 1 + 2 }} | error | correct }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + X }} | error | correct }}</syntaxhighlight> → '''{{#iferror: {{#expr: 1 + X }} | error | correct }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + 2 }} | error }}</syntaxhighlight> → '''{{#iferror: {{#expr: 1 + 2 }} | error }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + X }} | error }}</syntaxhighlight> → '''{{#iferror: {{#expr: 1 + X }} | error }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + 2 }} }}</syntaxhighlight> → '''{{#iferror: {{#expr: 1 + 2 }} }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: 1 + X }} }}</syntaxhighlight> → '''&zwnj;{{#iferror: {{#expr: 1 + X }} }}'''<!-- maybe the zwnj protects the trailing space before no output -->
:<syntaxhighlight lang="wikitext" inline>{{#iferror: {{#expr: . }} | error | correct }}</syntaxhighlight> → '''{{#iferror: {{#expr: . }} | error | correct }}'''
:<syntaxhighlight lang="wikitext" inline>{{#iferror: <strong class="error">a</strong> | error | correct }}</syntaxhighlight> → '''{{#iferror: <strong class="error">a</strong> | error | correct }}'''
<translate><!--T:2682--> Some errors may cause a tracking category to be added, using <tvar name=1><code><nowiki>{{#iferror:}}</nowiki></code></tvar> will not suppress the addition of the category.</translate>

== #ifexpr ==

<translate>
<!--T:77-->
This function evaluates a mathematical expression and returns one of two strings depending on the boolean value of the result:
</translate>

:<code><nowiki>{{#ifexpr:</nowiki> ''expression'' <nowiki>|</nowiki> ''value if true'' <nowiki>|</nowiki> ''value if false'' <nowiki>}}</nowiki></code>

<translate>
<!--T:78-->
The <tvar name=1><code>expression</code></tvar> input is evaluated exactly as for <tvar name=expr><code>[[##expr|#expr]]</code></tvar> above, with the same operators being available. The output is then evaluated as a boolean expression.

<!--T:79-->
An empty input expression evaluates to <tvar name=false>{{phpi|false}}</tvar>:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: | yes | no}}</syntaxhighlight> → '''{{#ifexpr: | yes | no}}'''

<translate>
<!--T:80-->
As mentioned above, zero evaluates to <tvar name=false>{{phpi|false}}</tvar> and any nonzero value evaluates to <tvar name=true>{{phpi|true}}</tvar>, so this function is equivalent to one using <tvar name=ifeq><code>[[##ifeq|#ifeq]]</code></tvar> and <tvar name=expr><code>[[##expr|#expr]]</code></tvar> only:
</translate>

:<code><nowiki>{{#ifeq: {{#expr: </nowiki>''expression''<nowiki> }} | 0 | </nowiki> ''value if false'' <nowiki>|</nowiki> ''value if true'' <nowiki>}}</nowiki></code>

<translate>
<!--T:81-->
except for an empty or wrong input expression (an error message is treated as an empty string; it is not equal to zero, so we get <tvar name=1><code>value if true</code></tvar>).
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: = | yes | no }}</syntaxhighlight> → ''' {{#ifexpr: = | yes | no }}'''

<translate>
<!--T:82-->
comparing
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: {{#expr: = }} | 0 | no | yes }}</syntaxhighlight> → ''' {{#ifeq: {{#expr: = }} | 0 | no | yes }}'''

<translate>
<!--T:83-->
Either or both of the return values may be omitted; no output is given when the appropriate branch is left empty:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 1 > 0 | yes }}</syntaxhighlight> → '''{{#ifexpr: 1 > 0 | yes }}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 1 < 0 | yes }}</syntaxhighlight> → ''' {{#ifexpr: 1 < 0 | yes }}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 0 = 0 | yes }}</syntaxhighlight> → ''' {{#ifexpr: 0 = 0 | yes }}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 1 > 0 | | no}}</syntaxhighlight> → ''' {{#ifexpr: 1 > 0 | | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 1 < 0 | | no}}</syntaxhighlight> → ''' {{#ifexpr: 1 < 0 | | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 1 > 0 }}</syntaxhighlight> → ''' {{#ifexpr: 1 > 0 }}'''

<translate><!--T:2683--> Boolean operators of equality or inequality operators are supported.</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 0 = 0 or 1 = 0 | yes}}</syntaxhighlight> → '''{{#ifexpr: 0 = 0 or 1 = 0 | yes}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 0 = 0 and 1 = 0 | | no}}</syntaxhighlight> → '''{{#ifexpr: 0 = 0 and 1 = 0 | | no}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 2 > 0 or 1 < 0 | yes}}</syntaxhighlight> → '''{{#ifexpr: 2 > 0 or 1 < 0  | yes}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 2 > 0 and 1 > 0 | yes | no}}</syntaxhighlight> → '''{{#ifexpr: 2 > 0 and 1 > 0 | yes | no}}'''

{{Warning|1=<translate><!--T:70--> The results of numerical comparisons with <tvar name=expr><code>[[##ifexpr|#ifexpr]]</code></tvar> do not always match those of <tvar name=ifeq><code>[[##ifeq|#ifeq]]</code></tvar> and <tvar name=switch><code>[[##switch|#switch]]</code></tvar>.</translate> <translate><!--T:2561--> These latter two are more accurate than <tvar name=expr><code>[[##ifexpr|#ifexpr]]</code></tvar>, and so may not return equivalent results.</translate>
<translate><!--T:2562--> Consider these comparisons with the final digit changed:</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifeq: 12345678901234567 | 12345678901234568 | equal | not equal}}</syntaxhighlight> → '''{{#ifeq: 12345678901234567 | 12345678901234568 | equal | not equal}}'''
:<syntaxhighlight lang="wikitext" inline>{{#switch: 12345678901234567 | 12345678901234568 = equal | not equal}}</syntaxhighlight> → '''{{#switch: 12345678901234567 | 12345678901234568 = equal | not equal}}'''

<translate><!--T:71--> Because PHP used in <tvar name=ifeq><code>[[##ifeq|#ifeq]]</code></tvar> and <tvar name=switch><code>[[##switch|#switch]]</code></tvar> compares two numbers of type integer, it returns the expected result correctly.</translate>
<translate><!--T:2563--> Whereas with <tvar name=expr><code>[[##ifexpr|#ifexpr]]</code></tvar> and the same numbers:</translate>

:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 12345678901234567 = 12345678901234568 | equal | not equal}}</syntaxhighlight> → '''{{#ifexpr: 12345678901234567 = 12345678901234568 | equal | not equal}}'''

<translate><!--T:2564--> With the different digit, the result of equal is actually incorrect.</translate>
<translate><!--T:72--> This behaviour in <tvar name=expr><code>[[##ifexpr|#ifexpr]]</code></tvar> is caused because MediaWiki converts literal numbers in expressions to type float, which, for large integers like these, involves rounding.</translate><br /><br />
}}

{{Anchor|ifexist|Ifexist|ifExist|IfExist|IFexist|IFEXIST}}

== #ifexist ==
{{hatnote|1=<translate><!--T:2712--> See [[<tvar name=1>Manual:Checking for page existence</tvar>|Manual:Checking for page existence]] for other methods of checking if a page exists with different limitations</translate>}}

<translate>
<!--T:85-->
This function takes an input string, interprets it as a page title, and returns one of two values depending on whether or not the page exists on the local wiki.
</translate>

:<code><nowiki>{{#ifexist:</nowiki> ''page title'' | ''value if exists'' | ''value if doesn't exist'' <nowiki>}}</nowiki></code>

<translate><!--T:86--> The function evaluates to <tvar name=1>{{phpi|true}}</tvar> if the page exists, whether it contains content, is visibly blank (contains meta-data such as category links or {{<tvar name=help-magic>ll|Help:Magic words</tvar>|magic words}}, but no visible content), is blank, or is a {{<tvar name=help-redirects>ll|Help:Redirects</tvar>|redirect}}.</translate>
<translate><!--T:2773--> Only pages that are redlinked evaluate to <tvar name=2>{{phpi|false}}</tvar>, including if the page used to exist but has been deleted.</translate>

:<code><nowiki>{{#ifexist:</nowiki> {{FULLPAGENAME}} | exists | doesn't exist <nowiki>}}</nowiki></code> → '''{{#ifexist: {{FULLPAGENAME}} | exists | doesn't exist }}'''
:<code><nowiki>{{#ifexist:</nowiki> XX{{FULLPAGENAME}}XX | exists | doesn't exist <nowiki>}}</nowiki></code> → '''{{#ifexist: XX{{FULLPAGENAME}}XX | exists | doesn't exist }}'''

<translate>
<!--T:87-->
The function evaluates to <tvar name=1>{{phpi|true}}</tvar> for {{<tvar name=system>ll|Help:System message</tvar>|system messages}} that have been customized, and for {{<tvar name=special>ll|Help:Special pages</tvar>|special pages}} that are defined by the software.
</translate>

:<code><nowiki>{{#ifexist:</nowiki> {{#special:Watchlist}} | exists | doesn't exist <nowiki>}}</nowiki></code> → '''{{#ifexist: Special:Watchlist | exists | doesn't exist }}'''
:<code><nowiki>{{#ifexist:</nowiki> {{#special:CheckUser}} | exists | doesn't exist <nowiki>}}</nowiki></code> → {{#ifexist: Special:Checkuser | '''exists''' <translate><!--T:88--> (because the <tvar name=1>{{ll|Extension:Checkuser|nsp=0}}</tvar> extension is installed on this wiki)</translate> | '''doesn't exist''' <translate><!--T:89--> (because the <tvar name=1>{{ll|Extension:Checkuser|nsp=0}}</tvar> extension is not installed on this wiki)</translate> }}
:<code><nowiki>{{#ifexist: MediaWiki:Copyright | exists | doesn't exist }}</nowiki></code> → {{#ifexist: MediaWiki:Copyright | '''exists''' <translate><!--T:90--> (because [[{{ns:8}}:Copyright]] has been customized)</translate> | '''doesn't exist''' <translate><!--T:91--> (because [[{{ns:8}}:Copyright]] has not been customized)</translate> }}

<translate><!--T:92--> Prior to MediaWiki 1.45 if a page checks a target using <tvar name=ifexist><code>#ifexist:</code></tvar>, then that page will appear in the <tvar name=spec-wth>[[{{#special:WhatLinksHere}}]]</tvar> list for the target page.</translate>
<translate><!--T:2768--> So if the code <tvar name=code1><code><nowiki>{{#ifexist</tvar>:Foo <tvar name=code2>}}</nowiki></code></tvar> were included live on this page (<tvar name=fullpagename1>{{FULLPAGENAME}}</tvar>), [[<tvar name=wth>{{#special:WhatLinksHere}}</tvar>/Foo]] will list <tvar name=fullpagename2>{{FULLPAGENAME}}</tvar>.</translate>

<translate>
<!--T:93-->
On wikis using a shared media repository, <tvar name=ifexist><code>#ifexist:</code></tvar> can be used to check if a file has been uploaded to the repository but not to the wiki itself:
</translate>

:<code><nowiki>{{#ifexist:</nowiki> {{ns:File}}:Example.png <nowiki>| exists | doesn't exist }}</nowiki></code> → '''{{#ifexist: {{ns:File}}:Example.png | exists | doesn't exist }}'''
:<code><nowiki>{{#ifexist:</nowiki> Image:Example.png <nowiki>| exists | doesn't exist }}</nowiki></code> → '''{{#ifexist: Image:Example.png | exists | doesn't exist }}'''
:<code><nowiki>{{#ifexist:</nowiki> {{ns:-2}}:Example.png <nowiki>| exists | doesn't exist }}</nowiki></code> → '''{{#ifexist: {{ns:-2}}:Example.png | exists | doesn't exist }}'''

<translate>
<!--T:94-->
If a local description page has been created for the file, the result is '''exists''' for all of the above.

<!--T:95-->
<tvar name=ifexist><code>#ifexist:</code></tvar> does not work with interwiki links.

=== ifexist limits === <!--T:96-->
</translate>
<translate><!--T:263--> <tvar name=1><code>#ifexist:</code></tvar> is considered an "expensive parser function"; only a limited number of which can be included on any one page (including functions inside transcluded templates).</translate>
<translate><!--T:2668--> When this limit is exceeded, any further <tvar name=1><code>#ifexist:</code></tvar> functions automatically return false, whether the target page exists or not, and the page is categorized into <tvar name=2>[[:Category:{{MediaWiki:expensive-parserfunction-category}}]]</tvar>.</translate>
<translate><!--T:2669--> The name of the {{<tvar name=1>ll|Help:Tracking categories</tvar>|tracking category}} may vary depending on the content language of your wiki.</translate>

<translate><!--T:97--> For some use cases it is possible to emulate the ifexist effect with css, by using the selectors <tvar name=1><code>a.new</code></tvar> (to select links to nonexistent pages) or <tvar name=2><code>a:not(.new)</code></tvar> (to select links to existing pages).</translate>
<translate><!--T:2670--> Furthermore, since the number of expensive parser functions that can be used on a single page is controlled by <tvar name=1><code>{{ll|Manual:$wgExpensiveParserFunctionLimit|$wgExpensiveParserFunctionLimit}}</code></tvar>, one can also increase the limit in <tvar name=2>LocalSettings.php</tvar> if needed.</translate>

<translate>
=== ifexist and wanted pages === <!--T:2455-->
</translate>
<translate><!--T:2456--> Prior to MediaWiki 1.45, a page that does not exist and is tested for using <tvar name=1>#ifexist</tvar> will end up on the [[<tvar name=2>Special:MyLanguage/Manual:WantedPages</tvar>|Wanted Pages]].</translate>
<translate><!--T:2648--> See <tvar name=1>{{phab|T14019}}</tvar> for the reason, and [[w:Template:Linkless exists]] for a workaround.</translate>

== #rel2abs ==
<translate>
<!--T:98-->
This function converts a relative file path into an absolute filepath.
</translate>

:<code><nowiki>{{#rel2abs:</nowiki> ''path'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#rel2abs:</nowiki> ''path'' <nowiki>|</nowiki> ''base path'' <nowiki>}}</nowiki></code>

<translate>
<!--T:99-->
Within the <tvar name=1><code>path</code></tvar> input, the following syntax is valid:
</translate>

*<code>.</code> → <translate><!--T:2565--> the current level</translate>
*<code>..</code> → <translate><!--T:2566--> go up one level</translate>
*<code>/foo</code> → <translate><!--T:2567--> go down one level into the subdirectory <tvar name=1>/foo</tvar></translate>

<translate>
<!--T:100-->
If the <tvar name=1><code>base path</code></tvar> is not specified, the full page name of the page will be used instead:
</translate>

:<code><nowiki>{{#rel2abs:</nowiki> /quok   | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: /quok | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ./quok  | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ./quok | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ../quok | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../quok | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ../.    | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../. | {{ns:12}}:Foo/bar/baz }}'''

<translate><!--T:101--> Invalid syntax, such as <tvar name=code1><code>/.</code></tvar> or <tvar name=code2><code>/./</code></tvar>, is ignored.</translate>
<translate><!--T:2462--> Since no more than two consecutive full stops are permitted, sequences such as these can be used to separate successive statements:</translate>

:<code><nowiki>{{#rel2abs:</nowiki> ../quok/.        | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../quok/. | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ../../quok       | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../../quok | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ../../../quok    | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../../../quok | {{ns:12}}:Foo/bar/baz }}'''
:<code><nowiki>{{#rel2abs:</nowiki> ../../../../quok | {{ns:12}}:Foo/bar/baz <nowiki>}}</nowiki></code> → '''{{#rel2abs: ../../../../quok | {{ns:12}}:Foo/bar/baz }}'''

<translate><!--T:2672--> For a similar group of functions see also [[<tvar name=1>Special:MyLanguage/Help:Magic words#URL data</tvar>|Help:Magic words#URL data]].</translate>
<translate><!--T:2673--> Built-in parser functions include: <tvar name=1>'localurl:', 'fullurl:', 'anchorencode:'</tvar> etc.</translate>

== #switch ==
''<translate><!--T:2649--> See also:</translate> [[w:Help:Switch parser function]]''

<translate>
<!--T:102-->
This function compares one input value against several test cases, returning an associated string if a match is found.
</translate>

{{pre|1=
<nowiki>{{#switch:</nowiki> ''comparison string''
 {{!}} ''case'' = ''result''
 {{!}} ''case'' = ''result''
 {{!}} ''...''
 {{!}} ''case'' = ''result''
 {{!}} ''default result''
<nowiki>}}</nowiki>
}}

<translate>
<!--T:103-->
Examples:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: baz | foo = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' {{#switch: baz | foo = Foo | baz = Baz | Bar }} '''
:<syntaxhighlight lang="wikitext" inline>{{#switch: foo | foo = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' {{#switch: foo | foo = Foo | baz = Baz | Bar }} '''
:<syntaxhighlight lang="wikitext" inline>{{#switch: zzz | foo = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' {{#switch: zzz | foo = Foo | baz = Baz | Bar }} '''

<translate>
<!--T:104-->
<tvar name=1><nowiki>#</nowiki>switch</tvar> with partial transclusion tags can be used to make a configuration file that enables editors unfamiliar with template coding to view and edit configurable elements.

=== Default === <!--T:318-->

<!--T:105-->
The <code>''default result''</code> is returned if no <code>''case''</code> string matches the <code>''comparison string''</code>:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: test | foo = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' {{#switch: test | foo = Foo | baz = Baz | Bar }} '''

<translate><!--T:106--> In this syntax, the default result must be the last parameter and must not contain a raw equals sign (an equals sign without <tvar name=1><code><nowiki>{{}}</nowiki></code></tvar>).</translate>
<translate><!--T:2464--> If it does, it will be treated as a case comparison, and no text will display if no cases match.</translate>
<translate><!--T:2465--> This is because the default value has not been defined (is empty).</translate>
<translate><!--T:2466--> If a case matches however, its associated string will be returned.</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: test | Bar | foo = Foo | baz = Baz }}</syntaxhighlight>&nbsp;→ ''' {{#switch: test | Bar | foo =Foo | baz =Baz }} '''<!-- do not remove the spaces between «''' {{» and «}} '''» -->
:<syntaxhighlight lang="wikitext" inline>{{#switch: test | foo = Foo | baz = Baz | B=ar }}</syntaxhighlight>&nbsp;→ ''' {{#switch: test | foo = Foo | baz = Baz | B=ar }} '''<!-- do not remove the spaces between «''' {{» and «}} '''» -->
:<syntaxhighlight lang="wikitext" inline>{{#switch: test | test = Foo | baz = Baz | B=ar }}</syntaxhighlight>&nbsp;→ ''' {{#switch: test | test = Foo | baz = Baz | B=ar }} '''<!-- do not remove the spaces between «''' {{» and «}} '''» -->

<translate>
<!--T:107-->
Alternatively, the default result may be explicitly declared with a <code>''case''</code> string of "<code>#default</code>".
</translate>

{{pre|1=
<nowiki>{{#switch:</nowiki> ''comparison string''
 {{!}} ''case'' = ''result''
 {{!}} ''case'' = ''result''
 {{!}} ''...''
 {{!}} ''case'' = ''result''
 {{!}} #default = ''default result''
<nowiki>}}</nowiki>
}}

<translate>
<!--T:108-->
Default results declared in this way may be placed anywhere within the function:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: test | foo = Foo | #default = Bar | baz = Baz }}</syntaxhighlight> → ''' {{#switch: test | foo = Foo | #default = Bar | baz = Baz }} '''

<translate>
<!--T:109-->
If the <code>''default''</code> parameter is omitted and no match is made, no <code>''result''</code> is returned:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: test | foo = Foo | baz = Baz }}</syntaxhighlight> → ''' {{#switch: test | foo = Foo | baz = Baz }} '''<!-- do not remove the spaces between «''' {{» and «}} '''» -->

<translate>
=== Grouping results === <!--T:319-->

<!--T:110-->
It is possible to have 'fall through' values, where several <code>''case''</code> strings return the same <code>''result''</code> string. This minimizes duplication.
</translate>

{{pre|1=
<nowiki>{{#switch:</nowiki> ''comparison string''
 {{!}} ''case1'' = ''result1''
 {{!}} ''case2'' 
 {{!}} ''case3'' 
 {{!}} ''case4'' = ''result234''
 {{!}} ''case5'' = ''result5''
 {{!}} ''case6'' 
 {{!}} ''case7'' = ''result67''
 {{!}} #default = ''default result''
<nowiki>}}</nowiki>
}}

<translate><!--T:111--> Here cases 2, 3 and 4 all return <code>''result234''</code>; cases 6 and 7 both return <code>''result67''</code>.</translate>
<translate><!--T:2467--> The "<tvar name=1><code>#default = </code></tvar>" in the last parameter may be omitted in the above case.</translate>

<translate>
=== Use with parameters === <!--T:2476-->
</translate>
<translate><!--T:2477--> The function may be used with parameters as the test string.</translate>
<translate><!--T:2478--> In this case, it is not necessary to place the pipe after the parameter name, because it is very unlikely that you will choose to set a case to be the string "<tvar name=1><code><nowiki>{{{</nowiki>''parameter name''<nowiki>}}}</nowiki></code></tvar>".</translate>
(<translate><!--T:2479--> This is the value the parameter will default to if the pipe is absent and the parameter doesn't exist or have a value.</translate>
<translate><!--T:2480--> See <tvar name=1>{{ll|Help:Parser functions in templates}}</tvar>.</translate>)

:<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | foo = Foo | baz = Baz | Bar }}</syntaxhighlight>

<translate><!--T:2481--> In the above case, if <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> equals <tvar name=2><code>foo</code></tvar>, the function will return <tvar name=3><code>Foo</code></tvar>.</translate>
<translate><!--T:2482--> If it equals <tvar name=1><code>baz</code></tvar>, the function will return <tvar name=2><code>Baz</code></tvar>.</translate>
<translate><!--T:2483--> If the parameter is empty or does not exist, the function will return <tvar name=1><code>Bar</code></tvar>.</translate>

<translate>
<!--T:2484-->
As in the section above, cases can be combined to give a single result.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | foo | zoo | roo = Foo | baz = Baz | Bar }}</syntaxhighlight>

<translate><!--T:2505--> Here, if <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> equals <tvar name=2><code>foo</code></tvar>, <tvar name=3><code>zoo</code></tvar> or <tvar name=4><code>roo</code></tvar>, the function will return <tvar name=5><code>Foo</code></tvar>.</translate>
<translate><!--T:2506--> If it equals <tvar name=1><code>baz</code></tvar>, the function will return <tvar name=2><code>Baz</code></tvar>.</translate>
<translate><!--T:2507--> If the parameter is empty or does not exist, the function will return <tvar name=1><code>Bar</code></tvar>.</translate>

<translate>
<!--T:2508-->
Additionally, the default result can be omitted if you do not wish to return anything if the test parameter value does not match any of the cases.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | foo = Foo | bar = Bar }}</syntaxhighlight>

<translate>
<!--T:2568-->
In this case, the function returns an empty string unless <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> exists and equals <tvar name=2><code>foo</code></tvar> or <tvar name=3><code>bar</code></tvar>, in which case it returns <tvar name=4><code>Foo</code></tvar> or <tvar name=5><code>Bar</code></tvar>, respectively.

<!--T:2569-->
This has the same effect as declaring the default result as empty.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | foo | zoo | roo = Foo | baz = Baz | }}</syntaxhighlight>

<translate><!--T:2570--> If you decide to set a case as "<tvar name=1><code><nowiki>{{{</nowiki>''parameter name''<nowiki>}}}</nowiki></code></tvar>", the function will return that case's result when the parameter doesn't exist or doesn't have a value.</translate>
<translate><!--T:2571--> The parameter would have to exist and have a value other than the string "<tvar name=1><code><nowiki>{{{</nowiki>''parameter name''<nowiki>}}}</nowiki></code></tvar>" to return the function's default result.</translate>

:<translate><!--T:2624--> (when <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> doesn't exist or is empty):</translate>
::<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | {{{1}}} = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' Foo '''
:<translate><!--T:2625--> (when <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> has the value "<tvar name=2><code>test</code></tvar>"):</translate>
::<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | {{{1}}} = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' Bar '''
:<translate><!--T:2626--> (when <tvar name=1><code><nowiki>{{{1}}}</nowiki></code></tvar> has the value "<tvar name=2><code><nowiki>{{{1}}}</nowiki></code></tvar>"):</translate>
::<syntaxhighlight lang="wikitext" inline>{{#switch: {{{1}}} | {{{1}}} = Foo | baz = Baz | Bar }}</syntaxhighlight> → ''' Foo '''

<translate>
<!--T:2627-->
In this hypothetical case, you would need to add the pipe to the parameter (<tvar name=1><code><nowiki>{{{1</nowiki><span style="color:red; font-weight:bold">|</span><nowiki>}}}</nowiki></code></tvar>).

=== Comparison behavior === <!--T:320-->

<!--T:112-->
As with <tvar name=ifeq><code>[[##ifeq|#ifeq]]</code></tvar>, the comparison is made numerically if both the comparison string and the case string being tested are numeric; or as a case-sensitive string otherwise:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: 0 + 1 | 1 = one | 2 = two | three}}</syntaxhighlight>&nbsp;→ '''{{#switch: 0 + 1 | 1 = one | 2 = two | three}}'''
:<syntaxhighlight lang="wikitext" inline>{{#switch: {{#expr: 0 + 1}} | 1 = one | 2 = two | three}}</syntaxhighlight>&nbsp;→ '''{{#switch: {{#expr: 0 + 1}} | 1 = one | 2 = two | three}}'''
:<syntaxhighlight lang="wikitext" inline>{{#switch: 02 | +1 = one | +2 = two | three}}</syntaxhighlight>&nbsp;→ '''{{#switch: 02 | +1 = one | +2 = two | three}}'''
:<syntaxhighlight lang="wikitext" inline>{{#switch: 100 | 1e1 = ten | 1e2 = hundred | other}}</syntaxhighlight>&nbsp;→ '''{{#switch: 100 | 1e1 = ten | 1e2 = hundred | other}}'''

:<syntaxhighlight lang="wikitext" inline>{{#switch: a | a = A | b = B | C}}</syntaxhighlight>&nbsp;→ '''{{#switch: a | a = A | b = B | C}}'''
:<syntaxhighlight lang="wikitext" inline>{{#switch: A | a = A | b = B | C}}</syntaxhighlight>&nbsp;→ '''{{#switch: A | a = A | b = B | C}}'''

<translate>
<!--T:113-->
A <code>''case''</code> string may be empty:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: | = Nothing | foo = Foo | Something }}</syntaxhighlight> → '''{{#switch: | = Nothing | foo = Foo | Something }}'''

<translate>
<!--T:114-->
Once a match is found, subsequent <code>''cases''</code> are ignored:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: b | f = Foo | b = Bar | b = Baz | }}</syntaxhighlight> → '''{{#switch: b | f = Foo | b = Bar | b = Baz | }}'''

{{Warning|1=
<translate>
<!--T:115-->
Numerical comparisons with <tvar name=switch><code>[[##switch|#switch]]</code></tvar> and <tvar name=ifeq><code>[[##ifeq|#ifeq]]</code></tvar> are not equivalent to comparisons in expressions (see also above):
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#switch: 12345678901234567 | 12345678901234568 = A | B}}</syntaxhighlight>&nbsp;→ '''{{#switch: 12345678901234567 | 12345678901234568 = A | B}}'''
:<syntaxhighlight lang="wikitext" inline>{{#ifexpr: 12345678901234567 = 12345678901234568 | A | B}}</syntaxhighlight>&nbsp;→ '''{{#ifexpr: 12345678901234567 = 12345678901234568 | A | B}}'''
}}

<translate>
=== Raw equal signs === <!--T:116-->

<!--T:265-->
"Case" strings cannot contain raw equals signs. To work around this, use the <tvar name=1>[[Special:MyLanguage/Help:Magic words#Other|{{((}}={{))}}]]</tvar> magic word, or replace equals sign with HTML code <tvar name=code2><code><nowiki>&amp;#61;</nowiki></code></tvar>.

<!--T:117-->
Example:
</translate>

{| class="wikitable"
! <translate><!--T:2658--> You type</translate>
! <translate><!--T:2659--> You get</translate>
|-
| <syntaxhighlight lang=wikitext>
{{#switch: 1=2
 | 1=2 = raw
 | 1<nowiki>=</nowiki>2 = nowiki
 | 1{{=}}2 = template
 | default
}}
</syntaxhighlight>
| '''{{#switch: 1=2 | 1=2 = raw | 1<nowiki>=</nowiki>2 = nowiki | 1{{=}}2 = template | default }}'''
|-
| <syntaxhighlight lang=wikitext>
{{#switch: 1=2
 | 1&#61;2 = html
 | default
}}
</syntaxhighlight>
| '''{{#switch: 1=2 | 1&#61;2 = html | default }}'''
|}

{{note|1=<translate><!--T:118--> For a simple real life example of the use of this function, check [<tvar name=url>https://en.wikipedia.org/w/index.php?title=Template:NBA_color&action=edit&oldid=600531646</tvar> Template:NBA color]. Two complex examples can be found at <tvar name=2>{{ll|Template:Extension}}</tvar> and <tvar name=1>[[w:Template:BOTREQ]]</tvar>.</translate>
}}

<translate>
=== Replacing #ifeq === <!--T:119-->

<!--T:266-->
<tvar name=1><code>#switch</code></tvar> can be used to reduce [[<tvar name=help>Special:MyLanguage/Help:Expansion depth</tvar>|expansion depth]].

<!--T:120-->
For example:
</translate>

* <syntaxhighlight lang="wikitext" inline>{{#switch:{{{1}}} |condition1=branch1 |condition2=branch2 |condition3=branch3 |branch4}}</syntaxhighlight>

<translate>
<!--T:121-->
is equivalent to
</translate>

* <syntaxhighlight lang="wikitext" inline>{{#ifeq:{{{1}}}|condition1 |branch1 |{{#ifeq:{{{1}}}|condition2 |branch2 |{{#ifeq:{{{1}}}|condition3 |branch3 |branch4}}}}}}</syntaxhighlight>

<translate>
<!--T:2458-->
i.e. deep nesting, linear:
</translate>

<syntaxhighlight lang="wikitext">
{{#ifeq:{{{1}}}|condition1
  |<!--then-->branch1
  |<!--else-->{{#ifeq:{{{1}}}|condition2
                |<!--then-->branch2
                |<!--else-->{{#ifeq:{{{1}}}|condition3
                              |<!--then-->branch3
                              |<!--else-->branch4}}}}}}
</syntaxhighlight>

<translate>
<!--T:2459-->
On the other hand, the switch replacement could be complicated/impractical for IFs nested in both branches (shown with alternatives of indentation, indented on both sides), making full symmetrical tree:
</translate>

<syntaxhighlight lang="wikitext">
{{#ifeq:{{{1}}}|condition1
 |<!--then-->branch1t{{
  #ifeq:{{{1}}}|condition2
   |<!--then-->branch1t2t{{#ifeq:{{{1}}}|condition4|<!--then-->branch1t2t4t|<!--else-->branch1t2t4e}}
   |<!--else-->branch1t2e{{#ifeq:{{{1}}}|condition5|<!--then-->branch1t2e5t|<!--else-->branch1t2e5e}}
  }}
 |<!--else-->branch1e{{#ifeq:{{{1}}}|condition3
   |<!--then-->branch1e3t{{#ifeq:{{{1}}}|condition6|branch1e3t6t|branch1e3t6e}}
   |<!--else-->branch1e3e{{
    #ifeq:{{{1}}}|condition7
     |branch1e3e7t
     |branch1e3e7t
    }}
  }}
}}
</syntaxhighlight>

{{Anchor|time|TIME}}

== #time ==
{| class="wikitable plainlinks" style="float: {{dir|{{pagelang}}|left|right}}; font-size: 90%; line-height: normal; margin-{{dir|{{pagelang}}|right|left}}: 1em; max-width: 40%;"
|-
! {{hl1}} | <translate><!--T:122--> Code</translate>
! {{hl1}} | <translate><!--T:123--> Description</translate>
! {{hl1}} | <translate><!--T:124--> Current output <br /><small class="noprint">([<tvar name=fullurl>{{fullurl:Help:Extension:ParserFunctions{{#translation:}}|action=purge}}</tvar> Purge this page's cache] to update)</small></translate>
|-
! {{hl2}} colspan="3"| <translate><!--T:125--> Year</translate>
|-
| style="text-align: center; font-weight: bold;" | <code>Y</code>
| style="font-weight: bold;" | <translate><!--T:126--> 4-digit year.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:Y}}
|-
| style="text-align: center;" | <code>y</code>
| <translate><!--T:127--> 2-digit year.</translate>
| style="text-align: center;" | {{#time:y}}
|-
| style="text-align: center;" | <code>L</code>
| <translate><!--T:128--> 1 if it's a leap year, 0 if not.</translate>
| style="text-align: center;" | {{#time:L}}
|-
| style="text-align: center;" | <code>o</code>{{#tag:ref|<translate><!--T:130--> Requires PHP 5.1.0 and newer and [[rev:45208]].</translate>|group=note}}
| <translate><!--T:129--> ISO-8601 year of the specified week.</translate>{{#tag:ref|<translate><!--T:131--> This has the same value as Y, except that if the ISO week number (W) belongs to the previous or next year, that year is used instead.</translate>|group=note}}
| style="text-align: center;" | {{#time:o}}{{#tag:ref|<translate><!--T:132--> Will output literal ''o'' if note 1 not fulfilled.</translate>|group=note}}
|-
| colspan="3" style="font-size: 88%;" |<references group="note"/>
|-
! {{hl2}} colspan="3"| <translate><!--T:133--> Month</translate>
|-
| style="text-align: center;" | <code>n</code>
| <translate><!--T:134--> Month index, not zero-padded.</translate>
| style="text-align: center;" | {{#time:n}}
|-
| style="text-align: center; font-weight: bold;" | <code>m</code>
| style="font-weight: bold;" | <translate><!--T:135--> Month index, zero-padded.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:m}}
|-
| style="text-align: center;" | <code>M</code>
| <translate><!--T:136--> An abbreviation of the month name, in the site language.</translate>
| style="text-align: center;" | {{#time:M}}
|-
| style="text-align: center;" | <code>F</code>
| <translate><!--T:137--> The full month name in the site language.</translate>
| style="text-align: center;" | {{#time:F}}
|-
| style="text-align: center;" | <code>xg</code>
| <translate><!--T:138--> Output the full month name in the [[w:Genitive case|genitive]] form for site languages that distinguish between genitive and [[w:Nominative case|nominative]] forms. This option is useful for many [[w:Slavic languages|Slavic languages]] like Polish, Russian, Belarusian, Czech, Slovak, Slovene, Ukrainian, etc.</translate>
| style="line-height: 1.6;" | <translate><!--T:139--> For Polish:</translate><br /><code style="font-size: 85%;"><nowiki>{{#time:F Y|June 2010|pl}}</nowiki></code> → {{#time:F Y|June 2010|pl}}<br /><translate><!--T:140--> (nominative)</translate><br /><code style="font-size: 85%;"><nowiki>{{#time:d xg Y|20 June 2010|pl}}</nowiki></code> → {{#time:d xg Y|20 June 2010|pl}}<br /><translate><!--T:141--> (genitive)</translate>
|-
! {{hl2}} colspan="3"| <translate><!--T:144--> Day of the month or the year</translate>
|-
| style="text-align: center;" | <code>j</code>
| <translate><!--T:145--> Day of the month, not zero-padded.</translate>
| style="text-align: center;" | {{#time:j}}
|-
| style="text-align: center; font-weight: bold;" | <code>d</code>
| style="font-weight: bold;" | <translate><!--T:146--> Day of the month, zero-padded.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:d}}
|-
| style="text-align: center;" | <code>z</code>
| <translate><!--T:147--> Day of the year (January&nbsp;1 = 0).<br />{{note}} To get the ISO day of the year add 1.</translate>
| style="text-align: center;" | {{#time:z}}
|-
! {{hl2}} colspan="3"| <translate><!--T:142--> Week and day of the week</translate>
|-
| style="text-align: center;" | <code>W</code>
| <translate><!--T:143--> ISO 8601 week number, zero-padded.</translate>
| style="text-align: center;" | {{#time:W}}
|-
| style="text-align: center;" | <code>N</code>
| <translate><!--T:150--> ISO 8601 day of the week (Monday = 1, Sunday = 7).</translate>
| style="text-align: center;" | {{#time:N}}
|-
| style="text-align: center;" | <code>w</code>
| <translate><!--T:151--> Number of the day of the week (Sunday = 0, Saturday = 6).</translate>
| style="text-align: center;" | {{#time:w}}
|-
| style="text-align: center;" | <code>D</code>
| <translate><!--T:148--> An abbreviation for the day of the week. Rarely internationalized.</translate>
| style="text-align: center;" | {{#time:D}}
|-
| style="text-align: center;" | <code>l</code>
| <translate><!--T:149--> The full weekday name. Rarely internationalized.</translate>
| style="text-align: center;" | {{#time:l}}
|-
!{{hl2}} colspan="3"| <translate><!--T:152--> Hour</translate>
|-
| style="text-align: center;" | <code>a</code>
| <translate><!--T:153--> "am" during the morning (00:00:00 → 11:59:59), "pm" otherwise (12:00:00 → 23:59:59).</translate>
| style="text-align: center;" | {{#time:a}}
|-
| style="text-align: center;" | <code>A</code>
| <translate><!--T:154--> Uppercase version of <tvar name=1><code>a</code></tvar> above.</translate>
| style="text-align: center;" | {{#time:A}}
|-
| style="text-align: center;" | <code>g</code>
| <translate><!--T:155--> Hour in 12-hour format, not zero-padded.</translate>
| style="text-align: center;" | {{#time:g}}
|-
| style="text-align: center;" | <code>h</code>
| <translate><!--T:156--> Hour in 12-hour format, zero-padded.</translate>
| style="text-align: center;" | {{#time:h}}
|-
| style="text-align: center;" | <code>G</code>
| <translate><!--T:157--> Hour in 24-hour format, not zero-padded.</translate>
| style="text-align: center;" | {{#time:G}}
|-
| style="text-align: center; font-weight: bold;" | <code>H</code>
|style="font-weight: bold;"| <translate><!--T:158--> Hour in 24-hour format, zero-padded.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:H}}
|-
!{{hl2}} colspan="3"| <translate><!--T:159--> Minutes and seconds</translate>
|-
| style="text-align: center; font-weight: bold;" | <code>i</code>
| style="font-weight: bold;" | <translate><!--T:160--> Minutes past the hour, zero-padded.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:i}}
|-
| style="text-align: center; font-weight: bold;" | <code>s</code>
| style="font-weight: bold;" | <translate><!--T:161--> Seconds past the minute, zero-padded.</translate>
| style="text-align: center; font-weight: bold;" | {{#time:s}}
|-
| style="text-align: center;" | <code>U</code>
| <translate><!--T:2422--> [[w:Unix time|Unix time]].</translate> <translate><!--T:162--> Seconds since January 1 1970 00:00:00 GMT.</translate>
| style="text-align: center;" | {{#time:U}}
|-
! id="TimeZone" {{hl2}} colspan="3"| <translate><!--T:163--> Timezone (as of [[MediaWiki_1.22/Roadmap|1.22wmf2]])</translate>
|-
| style="text-align: center;" | <code>e</code>
| <translate><!--T:164--> Timezone identifier.</translate>
| style="text-align: center;" | {{#time:e}}
|-
| style="text-align: center;" | <code>I</code>
| <translate><!--T:165--> Whether or not the date is in daylight savings time.</translate>
| style="text-align: center;" | {{#time:I}}
|-
| style="text-align: center;" | <code>O</code>
| <translate><!--T:166--> Difference to Greenwich time (GMT)</translate>
| style="text-align: center;" | {{#time:O}}
|-
| style="text-align: center;" | <code>P</code>
| <translate><!--T:167--> Difference to Greenwich time (GMT), with colon</translate>
| style="text-align: center;" | {{#time:P}}
|-
| style="text-align: center;" | <code>T</code>
| <translate><!--T:168--> Timezone abbreviation.</translate>
| style="text-align: center;" | {{#time:T}}
|-
| style="text-align: center;" | <code>Z</code>
| <translate><!--T:169--> Timezone offset in seconds.</translate>
| style="text-align: center;" | {{#time:Z}}
|-
! {{hl2}} colspan="3"| <translate><!--T:170--> Miscellaneous</translate>
|-
| style="text-align: center;" | <code>t</code>
| <translate><!--T:171--> Number of days in the current month.</translate>
| style="text-align: center;" | {{#time:t}}
|-
| style="text-align: center;" | <code>c</code>
| <translate><!--T:172--> ISO 8601 formatted date, equivalent to <tvar name=iso8601><code>Y-m-d"T"H:i:s+00:00</code></tvar>.</translate>
| style="font-size: 75%;" | {{#time:c}}
|-
| style="text-align: center;" | <code>r</code>
| <translate><!--T:173--> [[rfc:5322|RFC 5322]] formatted date, equivalent to <tvar name=rfc5322><code>D, j M Y H:i:s +0000</code></tvar>, with weekday name and month name not internationalized.</translate>
| style="font-size: 75%;" | {{#time:r}}
|-
! {{hl2}} colspan="3" | <translate><!--T:174--> Non-Gregorian calendars</translate>
|-
! {{hl3}} colspan="3" | <translate><!--T:175--> Islamic</translate>
|-
| style="text-align: center;" | <code>xmj</code>
| <translate><!--T:176--> Day of the month.</translate>
| {{#time:xmj}}
|-
| style="text-align: center;" | <code>xmF</code>
| <translate><!--T:177--> Full month name.</translate>
| {{#time:xmF}}
|-
| style="text-align: center;" | <code>xmn</code>
| <translate><!--T:178--> Month index.</translate>
| {{#time:xmn}}
|-
| style="text-align: center;" | <code>xmY</code>
| <translate><!--T:179--> Full year.</translate>
| {{#time:xmY}}
|-
! {{hl3}} colspan="3" | <translate><!--T:180--> Iranian (Jalaly)</translate>
|-
| style="text-align: center;" | <code>xit</code>
| <translate><!--T:2423--> Number of days in the month.</translate>
| {{#time:xit}}
|-
| style="text-align: center;" | <code>xiz</code>
| <translate><!--T:2424--> Day of the year.</translate>
| {{#time:xiz}}
|-
| style="text-align: center;" | <code>xij</code>
| <translate><!--T:181--> Day of the month.</translate>
| {{#time:xij}}
|-
| style="text-align: center;" | <code>xiF</code>
| <translate><!--T:182--> Full month name.</translate>
| {{#time:xiF}}
|-
| style="text-align: center;" | <code>xin</code>
| <translate><!--T:183--> Month index.</translate>
| {{#time:xin}}
|-
| style="text-align: center;" | <code>xiY</code>
| <translate><!--T:184--> Full year.</translate>
| {{#time:xiY}}
|-
| style="text-align: center;" | <code>xiy</code>
| <translate><!--T:185--> 2-digit year.</translate>
| {{#time:xiy}}
|-
! {{hl3}} colspan="3" | <translate><!--T:186--> Hebrew</translate>
|-
| style="text-align: center;" | <code>xjj</code>
| <translate><!--T:187--> Day of the month.</translate>
| {{#time:xjj}}
|-
| style="text-align: center;" | <code>xjF</code>
| <translate><!--T:188--> Full month name.</translate>
| {{#time:xjF}}
|-
| style="text-align: center;" | <code>xjt</code>
| <translate><!--T:261--> Number of days in month.</translate>
| {{#time:xjt}}
|-
| style="text-align: center;" | <code>xjx</code>
| <translate><!--T:189--> Genitive form of the month name.</translate>
| {{#time:xjx}}
|-
| style="text-align: center;" | <code>xjn</code>
| <translate><!--T:190--> Month number.</translate>
| {{#time:xjn}}
|-
| style="text-align: center;" | <code>xjY</code>
| <translate><!--T:191--> Full year.</translate>
| {{#time:xjY}}
|-
! {{hl3}} colspan="3" | <translate><!--T:192--> Thai solar</translate>
|-
| style="text-align: center;" | <code>xkY</code>
| <translate><!--T:193--> Full year in [[w:Thai solar calendar|Thai solar calendar]]. <br />{{note}} For years before 1941 the dates in Jan-Mar range are not [[w:Thai_solar_calendar#New_year|calculated]] properly.</translate>
| {{#time:xkY}}
|-
! {{hl3}} colspan="3" | <translate><!--T:194--> Minguo/Juche year</translate>
|-
| style="text-align: center;" | <code>xoY</code>
| <translate><!--T:195--> Full year.</translate>
| {{#time:xoY}}
|-
! {{hl3}} colspan="3" | <translate><!--T:196--> Japanese nengo</translate>
|-
| style="text-align: center;" | <code>xtY</code>
| <translate><!--T:197--> Full year.</translate>
| {{#time:xtY}}
|-
! {{hl2}} colspan="3" | <translate><!--T:198--> Flags</translate>
|-
| style="text-align: center;" | <code>xn</code>
| <translate><!--T:199--> Format the next numeric code as a raw ASCII number.</translate>
| <translate><!--T:200--> In the Hindi language, <tvar name=time><syntaxhighlight lang="wikitext" inline>{{#time:H, xnH}}</syntaxhighlight></tvar> produces <tvar name=hindi>०६, 06</tvar>.</translate>
|-
| style="text-align: center;" | <code>xN</code>
| colspan="2" | <translate><!--T:201--> Like <tvar name=1><code>xn</code></tvar>, but as a toggled flag, which endures until the end of the string or until the next appearance of <tvar name=2><code>xN</code></tvar> in the string.</translate>
|-
| style="text-align: center;" | <code>xr</code>
| <translate><!--T:202--> Format the next number as a roman numeral. Only works for numbers up to 10,000<br /><small>(up to 3,000 in pre MediaWiki 1.20)</small>.</translate>
| <syntaxhighlight lang="wikitext" inline>{{#time:xrY}}</syntaxhighlight> → {{#time:xrY}}
|-
| style="text-align: center;" | <code>xh</code>
| <translate><!--T:203--> Format the next number as a Hebrew numeral.</translate>
| <syntaxhighlight lang="wikitext" inline>{{#time:xhY}}</syntaxhighlight> → {{#time:xhY}}
|}

<translate>
<!--T:204-->
This parser function takes a date and/or time (in the Gregorian calendar) and formats it according to the syntax given. A date/time object can be specified; the default is the value of the [[<tvar name=magic>Special:MyLanguage/Help:Magic words#Date and time</tvar>|magic word]] <tvar name=currenttimestamp><syntaxhighlight lang="wikitext" inline>{{CURRENTTIMESTAMP}}</syntaxhighlight></tvar> &ndash; that is, the time the page was last rendered into HTML.
</translate>

:<code><nowiki>{{#time:</nowiki> ''format string'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#time:</nowiki> ''format string'' <nowiki>|</nowiki> ''date/time object'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#time:</nowiki> ''format string'' <nowiki>|</nowiki> ''date/time object'' <nowiki>|</nowiki> ''language code'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#time:</nowiki> ''format string'' <nowiki>|</nowiki> ''date/time object'' <nowiki>|</nowiki> ''language code'' <nowiki>|</nowiki> ''local'' <nowiki>}}</nowiki></code>

<translate><!--T:205--> The list of accepted formatting codes is given in the table to the right.</translate>
<translate><!--T:2677--> Any character in the formatting string that is not recognized is passed through unaltered; this applies also to blank spaces (the system does not need them for interpreting the codes).</translate>
<translate><!--T:2709--> If no character is recognized in the formatting string, and the date/time object is without error, then the formatting string is returned as output.</translate>
<translate><!--T:2678--> There are also two ways to escape characters within the formatting string:</translate>

# <translate><!--T:2679--> A backslash followed by a formatting character is interpreted as a single literal character</translate>
# <translate><!--T:2680--> Characters enclosed in double quotes are considered literal characters, and the quotes are removed.</translate>

<translate>
<!--T:2681-->
In addition, the digraph <tvar name=1><code>xx</code></tvar> is interpreted as a single literal "<tvar name=2>x</tvar>".

<!--T:2650-->
As the list of formatting codes continues to evolve (with the support of new calendars, or of new date fields computed and formatted differently), you should escape all literal characters (not just ASCII letters currently used by formatting codes) that need to be passed through unaltered.
</translate>

<translate><!--T:2651--> Unfortunately, for now, the ASCII single quote is still not recognized as a simple alternative for marking literal text to the currently supported ASCII double quotes (for example, double quotes are mandatory for in other uses like the delimitation of string values in JSON, C, C++...) and backslashes (which have to be escaped as well in string constants used by many languages, including JSON, C, C++, PHP, JavaScript, Lua).</translate>
<translate><!--T:2652--> So you still cannot embed any literal double quote without escaping it with a backslash (or you can use other curly, angular or square quotation marks instead).</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y-m-d }}</syntaxhighlight> → '''{{#time: Y-m-d }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: [[Y]] m d }}</syntaxhighlight> → '''{{#time: [[Y]] m d }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: [[Y (year)]] }}</syntaxhighlight> → '''{{#time: [[Y (year)]] }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: [[Y "(year)"]] }}</syntaxhighlight> → '''{{#time: [[Y "(year)"]] }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: i's" }}</syntaxhighlight> → '''{{#time: i's" }}'''

<translate><!--T:206--> The <code>''date/time object''</code> can be in any format accepted by PHP's <tvar name=strtotime>[https://php.net/function.strtotime strtotime()]</tvar> function.</translate>
<translate><!--T:2660--> Absolute (e.g. <tvar name=1><code>20 December 2000</code></tvar>), relative (e.g. <tvar name=2><code>+20 hours</code></tvar>), and combined times (e.g. <tvar name=3><code>30 July +1 year</code></tvar>) are accepted.</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: r|now}}</syntaxhighlight> → '''{{#time: r|now}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|+2 hours}}</syntaxhighlight> → '''{{#time: r|+2 hours}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|now + 2 hours}}</syntaxhighlight> → '''{{#time: r|now + 2 hours}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|20 December 2000}}</syntaxhighlight> → '''{{#time: r|20 December 2000}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|December 20, 2000}}</syntaxhighlight> → '''{{#time: r|December 20, 2000}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|2000-12-20}}</syntaxhighlight> → '''{{#time: r|2000-12-20}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|2000 December 20}}</syntaxhighlight> &rarr; '''{{#time: r|2000 December 20}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r|last tuesday}}</syntaxhighlight> &rarr; '''{{#time: r|last tuesday}}'''

<translate>
<!--T:207-->
The <code>''language code''</code> in [[w:ISO 639-3|ISO 639-3]] (?) allows the string to be displayed in the chosen language
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time:d F Y|1988-02-28|nl}}</syntaxhighlight> → '''{{#time:d F Y|1988-02-28|nl}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:l|now|uk}}</syntaxhighlight> → '''{{#time:l|now|uk}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:d xg Y|20 June 2010|pl}}</syntaxhighlight> → '''{{#time:d xg Y|20 June 2010|pl}}'''

<translate>
<!--T:2449-->
The <tvar name=1><code>''local''</code></tvar> parameter specifies if the ''date/time object'' refers to the local timezone or to UTC.

<!--T:2450-->
This is a boolean parameters: its value is determined by casting the value of the argument (see the [<tvar name=url>https://php.net/language.types.boolean#language.types.boolean.casting</tvar> official PHP documentation] for details on how string are cast to boolean values).
</translate>

{{note|2=reminder|1=<translate><!--T:2451--> Please note that, if the variable <tvar name=1>{{phpi|$wgLocaltimezone}}</tvar> is set to <tvar name=2><code>UTC</code></tvar>, there is no difference in the output when <tvar name=3><code>local</code></tvar> is set to <tvar name=4><code>true</code></tvar> or <tvar name=5><code>false</code></tvar>.</translate>}}

<translate>
<!--T:2452-->
See the following examples for details:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y F d H:i:s|now|it|0}}</syntaxhighlight> → '''{{#time: Y F d H:i:s|now|it|0}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: Y F d H:i:s|now|it|1}}</syntaxhighlight> → '''{{#time: Y F d H:i:s|now|it|1}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: Y F d H:i:s|+2 hours||0}}</syntaxhighlight> → '''{{#time: Y F d H:i:s|+2 hours||0}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: Y F d H:i:s|+2 hours||1}}</syntaxhighlight> → '''{{#time: Y F d H:i:s|+2 hours||1}}'''

:<syntaxhighlight lang="wikitext" inline>{{#time:c|2019-05-16T17:05:43+02:00|it}}</syntaxhighlight> → '''{{#time:c|2019-05-16T17:05:43+02:00|it}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:c|2019-05-16T17:05:43+02:00|it|0}}</syntaxhighlight> → '''{{#time:c|2019-05-16T17:05:43+02:00|it|0}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:c|2019-05-16T17:05:43+02:00|it|true}}</syntaxhighlight> → '''{{#time:c|2019-05-16T17:05:43+02:00|it|true}}'''

<translate>
<!--T:208-->
If you've calculated a Unix timestamp, you may use it in date calculations by pre-pending an <tvar name=1><code>@</code></tvar> symbol.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: U | now }}</syntaxhighlight> → '''{{#time: U | now}}'''
:<code><nowiki>{{#time: r | @</nowiki>{{#time: U}} <nowiki>}}</nowiki></code> → '''{{#time: r | @{{#time:U}} }}'''

{{Warning|1=
<translate>
<!--T:2435-->
Without the <tvar name=1><code>@</code></tvar> prefix before numeric timestamp values, the result is an error most of the time, or is an unexpected value:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: r | 1970-01-01 00:16:39 }}</syntaxhighlight> → '''{{#time: r | 1970-01-01 00:16:39 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: U | 1970-01-01 00:16:39 }}</syntaxhighlight> → '''{{#time: U | 1970-01-01 00:16:39 }}'''
:<code><nowiki>{{#time: r | @</nowiki>{{#time: U | 1970-01-01 00:16:39 }}<nowiki> }}</nowiki></code> → '''{{#time: r | @{{#time: U | 1970-01-01 00:16:39 }} }}''' <translate><!--T:2436--> (correct)</translate>
:<code><nowiki>{{#time: r | </nowiki>{{#time: U | 1970-01-01 00:16:39 }}<nowiki> }}</nowiki></code> → '''{{#time: r | {{#time: U | 1970-01-01 00:16:39 }} }}''' <translate><!--T:2437--> (unsupported year format)</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: r | 1970-01-01 00:16:40 }}</syntaxhighlight> → '''{{#time: r | 1970-01-01 00:16:40 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: U | 1970-01-01 00:16:40 }}</syntaxhighlight> → '''{{#time: U | 1970-01-01 00:16:40 }}'''
:<code><nowiki>{{#time: r | @</nowiki>{{#time: U | 1970-01-01 00:16:40 }}<nowiki> }}</nowiki></code> → '''{{#time: r | @{{#time: U | 1970-01-01 00:16:40 }} }}''' <translate><!--T:2438--> (correct)</translate>
:<code><nowiki>{{#time: r | </nowiki>{{#time: U | 1970-01-01 00:16:40 }}<nowiki> }}</nowiki></code> → '''{{#time: r | {{#time: U | 1970-01-01 00:16:40 }} }}''' <translate><!--T:2439--> (interpreted as a year with current month and day of the month)</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: r | 1970-01-01 02:46:39 }}</syntaxhighlight> → '''{{#time: r | 1970-01-01 02:46:39 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: U | 1970-01-01 02:46:39 }}</syntaxhighlight> → '''{{#time: U | 1970-01-01 02:46:39 }}'''
:<code><nowiki>{{#time: r | @</nowiki>{{#time: U | 1970-01-01 02:46:39 }}<nowiki> }}</nowiki></code> → '''{{#time: r | @{{#time: U | 1970-01-01 02:46:39 }} }}''' <translate><!--T:2440--> (correct)</translate>
:<code><nowiki>{{#time: r | </nowiki>{{#time: U | 1970-01-01 02:46:39 }}<nowiki> }}</nowiki></code> → '''{{#time: r | {{#time: U | 1970-01-01 02:46:39 }} }}''' <translate><!--T:2441--> (interpreted as a year with current month and day of the month)</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: r | 1970-01-01 02:46:40 }}</syntaxhighlight> → '''{{#time: r | 1970-01-01 02:46:40 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: U | 1970-01-01 02:46:40 }}</syntaxhighlight> → '''{{#time: U | 1970-01-01 02:46:40 }}'''
:<code><nowiki>{{#time: r | @</nowiki>{{#time: U | 1970-01-01 02:46:40 }}<nowiki> }}</nowiki></code> → '''{{#time: r | @{{#time: U | 1970-01-01 02:46:40 }} }}''' <translate><!--T:2442--> (correct)</translate>
:<code><nowiki>{{#time: r | </nowiki>{{#time: U | 1970-01-01 02:46:40 }}<nowiki> }}</nowiki></code> → '''{{#time: r | {{#time: U | 1970-01-01 02:46:40 }} }}''' <translate><!--T:2443--> (unsupported year format)</translate>
}}

{{Warning|1=
<translate>
<!--T:209-->
The range of acceptable input is 1 January 0111 → 31 December 9999. For the years 100 through 110 the output is inconsistent, Y and leap years are like the years 100-110, r, D, l and U are like interpreting these years as 2000-2010.
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 29 Feb 0100 }}</syntaxhighlight> → '''{{#time: d F Y | 29 Feb 0100 }}'''<br /> <translate><!--T:210--> (correct, no leap year), but</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: r | 29 Feb 0100 }}</syntaxhighlight> → '''{{#time: r | 29 Feb 0100 }}''' <translate><!--T:211--> (wrong, even if 100 is interpreted as 2000, because that is a leap year)</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 15 April 10000 }}</syntaxhighlight> → '''{{#time: d F Y | 15 April 10000 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: r | 10000-4-15 }}</syntaxhighlight> → '''{{#time: r | 10000-4-15 }}'''

<translate>
<!--T:212-->
Year numbers 0-99 are interpreted as 2000-2069 and 1970-1999, except when written in 4-digit format with leading zeros:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 1 Jan 6 }}</syntaxhighlight> → '''{{#time: d F Y | 1 Jan 6 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 1 Jan 06 }}</syntaxhighlight> → '''{{#time: d F Y | 1 Jan 06 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 1 Jan 006 }}</syntaxhighlight> → '''{{#time: d F Y | 1 Jan 006 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | 1 Jan 0006 }}</syntaxhighlight> → '''{{#time: d F Y | 1 Jan 0006 }}''' (4-digit format)

<translate>
<!--T:213-->
The weekday is supplied for the years 100-110 and from 1753, for the years 111-1752 the r-output shows "Unknown" and the l-output "<>". As a consequence, the r-output is not accepted as input for these years.
</translate>
}}

<translate>
<!--T:214-->
Full or partial absolute dates can be specified; the function will "fill in" parts of the date that are not specified using the ''current'' values:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y | January 1 }}</syntaxhighlight> → '''{{#time: Y | January 1 }}'''

{{Warning|1=
<translate>
<!--T:215-->
The fill-in feature is not consistent; some parts are filled in using the current values, others are not:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | June }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | June }}''' <translate><!--T:216--> Gives the start of the day, but the current day of the month and the current year.</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | 2003 }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | 2003 }}''' <translate><!--T:217--> Gives the start of the day, but the current day of the year.</translate>

<translate>
<!--T:2646-->
There's exception case of the filled day:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | June 2003 }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | June 2003 }}''' <translate><!--T:2647--> Gives the start of the day and the start of the month.</translate>
}}

<translate>
<!--T:218-->
A four-digit number is always interpreted as a year, never as hours and minutes:<ref>Prior to <tvar name=1>{{rev|86805}}</tvar> in 2011 this was not the case.</ref>
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | 1959 }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | 1959 }}'''

<translate>
<!--T:219-->
A six-digit number is interpreted as hours, minutes and seconds if possible, but otherwise as an error (not, for instance, a year and month):
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | 195909 }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | 195909 }}''' <translate><!--T:220--> Input is treated as a time rather than a year+month code.</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: Y m d H:i:s | 196009 }}</syntaxhighlight> → '''{{#time: Y m d H:i:s | 196009 }}''' <translate><!--T:221--> Although 19:60:09 is not a valid time, 196009 is not interpreted as September 1960.</translate>

<translate>
<!--T:222-->
The function performs a certain amount of date mathematics:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time: d F Y | January 0 2008 }}</syntaxhighlight> → '''{{#time: d F Y | January 0 2008 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F | January 32 }}</syntaxhighlight> → '''{{#time: d F | January 32 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F | February 29 2008 }}</syntaxhighlight> → '''{{#time: d F | February 29 2008 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: d F | February 29 2007 }}</syntaxhighlight> → '''{{#time: d F | February 29 2007 }}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:Y-F|now -1 months}}</syntaxhighlight> → '''{{#time:Y-F|now -1 months}}'''

<translate><!--T:223--> The total length of the format strings of the calls of <tvar name=1><code>#time</code></tvar> is limited to 6000 characters.</translate><ref>
[https://phabricator.wikimedia.org/diffusion/EPFN/browse/master/includes/ParserFunctions.php$31 ParserFunctions.php]
</ref>

<translate>
==== Time Zone issue ==== <!--T:224-->

<!--T:267-->
There is a bug in this #time parser function (more specifically in ''PHP DateTime'') that does not allow the passing-in of ''non-integers'' as relative time zone offsets.  This issue does not apply when using an on-the-hour time zone, such as EDT. For example:
</translate>

* <syntaxhighlight lang="wikitext" inline>{{#time:g:i A | -4 hours }}</syntaxhighlight> &rarr; {{#time:g:i A | -4 hours}}

<translate>
<!--T:225-->
However, India is on a +5.5 hours time offset from UTC, and thus using its time zone will not normally allow the correct calculation of a relative time zone offset. Here's what happens:
</translate>

* <syntaxhighlight lang="wikitext" inline>{{#time:g:i A | +5.5 hours }}</syntaxhighlight> &rarr; {{#time:g:i A | +5.5 hours }}

<translate>
<!--T:226-->
To workaround this issue, simply convert the time into minutes or seconds, like this:
</translate>

* <syntaxhighlight lang="wikitext" inline>{{#time:g:i A | +330 minutes }}</syntaxhighlight> &rarr; {{#time:g:i A | +330 minutes }}
* <syntaxhighlight lang="wikitext" inline>{{#time:g:i A | +19800 seconds }}</syntaxhighlight> &rarr;  {{#time:g:i A | +19800 seconds }}

<translate>
<!--T:227-->
(Tim Starling, the developer of this function, provided the exact syntax for this solution.)

==== <tvar name=1>#time</tvar> format like in signatures ==== <!--T:2721-->
</translate>
<translate><!--T:2722--> Sometimes it is useful to construct a timestamp, which looks like the automatic timestamp generated by [[<tvar name=1>Special:MyLanguage/Help:Signatures</tvar>|signatures]] in discussions on talk pages.</translate>
<translate><!--T:2723--> On an English-language wiki, it can be created with:</translate>

* <syntaxhighlight lang="wikitext" inline>{{#timel:H:i, j xg Y (e)|+330 minutes}}</syntaxhighlight> &rarr; {{#timel:H:i, j xg Y (e)|+330 minutes}}

== #timel ==
<translate>
<!--T:228-->
This function is identical to <tvar name=time><code><nowiki>{{#time: ... }}</nowiki></code></tvar> with the <tvar name=2><code>local</code></tvar> parameter set to <tvar name=3>{{phpi|true}}</tvar>, so it always uses the local time of the wiki (as set in <tvar name=1>{{ll|Manual:$wgLocaltimezone|$wgLocaltimezone}}</tvar>).

<!--T:2453-->
Syntax of the function is:
</translate>

:<code><nowiki>{{#timel:</nowiki> ''format string'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#timel:</nowiki> ''format string'' <nowiki>|</nowiki> ''date/time object'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#timel:</nowiki> ''format string'' <nowiki>|</nowiki> ''date/time object'' <nowiki>|</nowiki> ''language code'' <nowiki>}}</nowiki></code>

{{note|2=reminder|1=<translate><!--T:2454--> Please note that, if the variable <tvar name=1>{{phpi|$wgLocaltimezone}}</tvar> is set to <tvar name=2><code>UTC</code></tvar>, there is no difference in the output when <tvar name=3><code>local</code></tvar> is set to <tvar name=4>{{phpi|true}}</tvar> or <tvar name=5>{{phpi|false}}</tvar></translate>}}

[[File:Time-Timel.png|thumb|<translate><!--T:2628--> Example of the use of #time and #timel parser functions from a server where the timezone is not UTC</translate>]]

<translate>
<!--T:2557-->
For instance, see the following examples:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#time:c|now|it}}</syntaxhighlight> → '''{{#time:c|now|it}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:c|now|it|0}}</syntaxhighlight> → '''{{#time:c|now|it|0}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time:c|now|it|1}}</syntaxhighlight> → '''{{#time:c|now|it|1}}'''
:<syntaxhighlight lang="wikitext" inline>{{#timel:c|now|it}}</syntaxhighlight> → '''{{#timel:c|now|it}}'''

[[File:Maldiskusjon Sommertid–Wikipedia.jpg|thumb|<translate><!--T:2425--> Warning Example from <tvar name=no-wp>https://no.wikipedia.org/wiki/Maldiskusjon:Sommertid</tvar></translate>]]

{{Warning|1=
<translate>
<!--T:2426-->
Be aware that U for both time and timel will return the same number of seconds since 1970-01-01 00:00:00 UTC on Wikipedias with different timezones than UTC (formerly known as GMT)
</translate>

:<code>U</code> <translate><!--T:2427--> Unix time. Seconds since January 1 1970 00:00:00 GMT.</translate>
:<code>Z</code> <translate><!--T:2428--> Timezone offset in seconds.</translate>
:<syntaxhighlight lang="wikitext" inline>{{#time: U}}</syntaxhighlight> → '''{{#time: U}}'''
:<syntaxhighlight lang="wikitext" inline>{{#timel: U}}</syntaxhighlight> → '''{{#timel: U}}'''
:<syntaxhighlight lang="wikitext" inline>{{#time: Z}}</syntaxhighlight> → '''{{#time: Z}}'''
:<syntaxhighlight lang="wikitext" inline>{{#timel: Z}}</syntaxhighlight> → '''{{#timel: Z}}'''
}}

{{-}}

{{Anchor|timef}}

== #timef ==
<translate><!--T:2748--> This function formats a date using a standard format for the selected language, as defined in <tvar name=1><code>[https://codesearch.wmcloud.org/core/?q=dateFormats&files=languages/messages $dateFormats]</code></tvar> (see <tvar name=2>{{Phab|T223772}}</tvar>).</translate>

: {{tmpl|0=<code><nowiki>{{#timef:</nowiki> $1 <nowiki>}}</nowiki></code>
|1=<translate><!--T:2754--> ''date/time object''</translate>
}}
: {{tmpl|0=<code><nowiki>{{#timef:</nowiki> $1 <nowiki>|</nowiki> $2 <nowiki>}}</nowiki></code>
|1=<translate><!--T:2755--> ''date/time object''</translate>
|2=<translate><!--T:2756--> ''format type''</translate>
}}
: {{tmpl|0=<code><nowiki>{{#timef:</nowiki> $1 <nowiki>|</nowiki> $2 <nowiki>|</nowiki> $3 <nowiki>}}</nowiki></code>
|1=<translate><!--T:2757--> ''date/time object''</translate>
|2=<translate><!--T:2758--> ''format type''</translate>
|3=<translate><!--T:2759--> ''language code''</translate>
}}

<translate>
<!--T:2749-->
The format of the ''date/time object'' is the same as for <tvar name=1>[[#time|#time]]</tvar>. If it is empty, the time when the page was rendered is used.

<!--T:2750-->
The ''format type'' may be one of:
</translate>

; <code>time</code> : <translate><!--T:2751--> Only the time is shown.</translate>
; <code>date</code> : <translate><!--T:2752--> Only the date is shown.</translate>
; <code>both</code> : <translate><!--T:2753--> Both the time and date are shown.</translate>
; <code>pretty</code> : <translate><!--T:2760--> Only the date is shown, using an abbreviated format which does not include the year.</translate> <translate><!--T:2761--> Not all languages support this; if it is not supported, the "date" format is used.</translate>

<translate>
<!--T:2762-->
If the ''format type'' is not specified, both the time and date will be show, as if <tvar name=1><code>both</code></tvar> were specified.

<!--T:2763-->
If the ''language code'' is not specified, the page's content language is used.

<!--T:2764-->
Using <tvar name=1><code>#timef</code></tvar> instead of <tvar name=2><code>#time</code></tvar> allows templates to more easily support multiple languages, since different languages have different ways to format dates.

<!--T:2765-->
In English, the order of the day and month is controlled by <tvar name=1>{{wg|AmericanDates}}</tvar>.

<!--T:2766-->
Examples:
</translate>

:<syntaxhighlight lang="wikitext" inline>{{#timef:now|both|en}}</syntaxhighlight> → ''' {{#timef:now|both|en}} '''
:<syntaxhighlight lang="wikitext" inline>{{#timef:now|both|ja}}</syntaxhighlight> → ''' {{#timef:now|both|ja}} '''
:<syntaxhighlight lang="wikitext" inline>{{#timef:now|pretty|en}}</syntaxhighlight> → ''' {{#timef:now|pretty|en}} '''
:<syntaxhighlight lang="wikitext" inline>{{#timef:now|pretty|pl}}</syntaxhighlight> → ''' {{#timef:now|pretty|pl}} '''
:<syntaxhighlight lang="wikitext" inline>{{#timef:|time}}</syntaxhighlight> → ''' {{#timef:|time}} '''

{{Anchor|timefl}}

== #timefl ==

<translate><!--T:2767--> This function is the same as <tvar name=1>[[#timef|#timef]]</tvar> except that it uses the local timezone of the wiki as configured in <tvar name=2>{{ll|Manual:$wgLocaltimezone|$wgLocaltimezone}}</tvar>.</translate>

:<code><nowiki>{{#timefl:</nowiki> ''date/time object'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#timefl:</nowiki> ''date/time object'' <nowiki>|</nowiki> ''format type'' <nowiki>}}</nowiki></code>
:<code><nowiki>{{#timefl:</nowiki> ''date/time object'' <nowiki>|</nowiki> ''format type'' <nowiki>|</nowiki> ''language code'' <nowiki>}}</nowiki></code>

{{Anchor|titleparts|Titleparts}}
== #titleparts ==
<translate>
<!--T:229-->
This function separates a page title into segments based on slashes, then returns some of those segments as output.
</translate>

: {{tmpl|0=<code class="mw-highlight">{{((}}#titleparts: ''$1'' {{!}} ''$2'' {{!}} ''$3'' {{))}}</code>
|1=<translate><!--T:2375--> pagename</translate>
|2=<translate><!--T:2656--> number of segments to return</translate>
|3=<translate><!--T:2657--> segment to start at</translate>
}}

<translate>
<!--T:230-->
If the ''number of segments to return'' parameter is not specified, it defaults to "0", which returns all the segments from the ''segment to start at'' to the end (included). If the ''segment to start at'' parameter is not specified or is "0", it defaults to "1":
</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{highlight|{{ns:1}}:Foo/bar/baz/quok}} <nowiki>}}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{highlight|{{ns:1}}:Foo}}/bar/baz/quok <nowiki>| 1 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | 1 }}''' <translate><!--T:2708--> See also <tvar name=1><nowiki>{{</nowiki>{{ll|Help:Magic_words#Page_names|ROOTPAGENAME}}<nowiki>}}</nowiki></tvar>.</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{highlight|{{ns:1}}:Foo/bar}}/baz/quok <nowiki>| 2 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | 2 }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/{{highlight|bar/baz}}/quok <nowiki>| 2 | 2 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | 2 | 2 }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/{{highlight|baz/quok}} <nowiki>| 2 | 3 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | 2 | 3 }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/{{highlight|bar/baz/quok}} <nowiki>| 3 | 2 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | 3 | 2 }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/{{highlight|bar/baz/quok}} <nowiki>| | 2 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | | 2 }}'''
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/baz/quok <nowiki>| | 5 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | | 5 }} '''

<translate>
<!--T:231-->
Negative values are accepted for both values. Negative values for the ''number of segments to return'' parameter effectively 'strips' segments from the end of the string. Negative values for the ''first segment to return'' translates to "start with this segment counting from the right":
</translate>

: <code><nowiki>{{#titleparts:</nowiki> {{highlight|{{ns:1}}:Foo/bar/baz}}/quok <nowiki>| -1 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | -1 }}''' <translate><!--T:232--> Strips one segment from the end of the string.</translate> <translate><!--T:2460--> See also <tvar name=1>{{ll|Help:Magic_words#Page_names|2=BASEPAGENAME}}</tvar>.</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/baz/quok <nowiki>| -4 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | -4 }} ''' <translate><!--T:233--> Strips all 4 segments from the end of the string</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/baz/quok <nowiki>| -5 }}</nowiki></code> → '''{{#titleparts: {{ns:1}}:Foo/bar/baz/quok | -5 }} ''' <translate><!--T:234--> Strips 5 segments from the end of the string (more than exist)</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/baz/{{highlight|quok}} <nowiki>| | -1 }}</nowiki></code> → ''' {{#titleparts: {{ns:1}}:Foo/bar/baz/quok | | -1 }}''' <translate><!--T:235--> Returns last segment.</translate> <translate><!--T:2461--> See also <tvar name=1>{{ll|Help:Magic_words#Page_names|2=SUBPAGENAME}}</tvar>.</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/{{highlight|bar/baz}}/quok <nowiki>| -1 | 2 }}</nowiki></code> → ''' {{#titleparts: {{ns:1}}:Foo/bar/baz/quok | -1 | 2 }}''' <translate><!--T:236--> Strips one segment from the end of the string, then returns the second segment and beyond</translate>
: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bar/{{highlight|baz}}/quok <nowiki>| -1 | -2 }}</nowiki></code> → ''' {{#titleparts: {{ns:1}}:Foo/bar/baz/quok | -1 | -2 }}''' <translate><!--T:237--> Start copying at the second last element; strip one segment from the end of the string</translate>

<translate>
<!--T:2376-->
Before processing, the ''pagename'' parameter is HTML-decoded: if it contains some standard HTML character entities, they will be converted to plain characters (internally encoded with UTF-8, i.e. the same encoding as in the MediaWiki source page using this parser function).
</translate>

: <translate><!--T:2377--> For example, any occurrence of <tvar name=q1><code>&amp;quot;</code></tvar>, <tvar name=q2><code>&amp;#34;</code></tvar>, or <tvar name=q3><code>&amp;#x22;</code></tvar> in ''pagename'' will be replaced by <tvar name=q><code>"</code></tvar>.</translate>
: <translate><!--T:2378--> No other conversion from HTML to plain text is performed, so HTML tags are left intact at this initial step even if they are invalid in page titles.</translate>

{{tip| 1=
<translate>
<!--T:2379-->
Some magic keywords or parser functions of MediaWiki (such as <tvar name=PAGENAME><code><nowiki>{{</nowiki>{{ll|Help:Magic words#Page names|PAGENAME}}<nowiki>}}</nowiki></code></tvar> and similar) are known to return strings that are needlessly HTML-encoded, even if their own input parameter was not HTML-encoded:

<!--T:2380-->
The titleparts parser function can then be used as a workaround, to convert these returned strings so that they can be processed correctly by some other parser functions also taking a page name in parameter (such as <tvar name=PAGESINCAT><code><nowiki>{{</nowiki>{{ll|Help:Magic words#PAGESINCAT|PAGESINCAT:}}<nowiki>}}</nowiki></code></tvar>) but which are still not working properly with HTML-encoded input strings.
</translate>

This may not be a suitable workaround for you if a pagename {{ll|Manual:$wgCapitalLinks|is allowed to start with a lowercase character}}: the titleparts parser function will always capitalize the first letter.

<translate>
<!--T:2381-->
For example, if the current page is [[commons:Category:Côte-d'Or|Category:Côte-d'Or]], then:
</translate>

* <translate><!--T:2382--> <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{#ifeq: {{FULLPAGENAME}} | Category:Côte-d'Or | 1 | 0 }}</syntaxhighlight></tvar>, and <tvar name=code2><syntaxhighlight lang="wikitext" inline>{{#ifeq: {{FULLPAGENAME}} | Category:Côte-d&apos;Or | 1 | 0 }}</syntaxhighlight></tvar> are both returning <tvar name=actual><code>1</code></tvar>; (the #ifeq parser function does perform the HTML-decoding of its input parameters).</translate>
* <translate><!--T:2383--> <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{#switch: {{FULLPAGENAME}} | Category:Côte-d'Or = 1 | #default = 0 }}</syntaxhighlight></tvar>, and <tvar name=code2><syntaxhighlight lang="wikitext" inline>{{#switch: {{FULLPAGENAME}} | Category:Côte-d&apos;Or = 1 | #default = 0 }}</syntaxhighlight></tvar> are both returning <tvar name=actual><code>1</code></tvar>; (the #switch parser function does perform the HTML-decoding of its input parameters).</translate>
* <translate><!--T:2384--> <tvar name=code0><syntaxhighlight lang="wikitext" inline>{{#ifexist: {{FULLPAGENAME}} | 1 | 0 }}</syntaxhighlight></tvar>, <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{#ifexist: Category:Côte-d'Or | 1 | 0 }}</syntaxhighlight></tvar>, or even <tvar name=code2><syntaxhighlight lang="wikitext" inline>{{#ifexist: Category:Côte-d&apos;Or | 1 | 0 }}</syntaxhighlight></tvar> will all return <tvar name=actual><code>1</code></tvar> if that category page exists (the #ifexist parser function does perform the HTML-decoding of its input parameters);</translate>
* <translate><!--T:2385--> <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: Côte-d'Or }}</syntaxhighlight></tvar> will return a non-zero number, if that category contains pages or subcategories, '''but''':</translate>
* <translate><!--T:2386--> <tvar name=code0><syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{CURRENTPAGENAME}} }}</syntaxhighlight></tvar>, may still '''unconditionally''' return 0, just like:</translate>
* <syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{PAGENAME:Category:Côte-d'Or}} }}</syntaxhighlight>
* <syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{PAGENAME:Category:Côte-d&apos;Or}} }}</syntaxhighlight>

<translate>
<!--T:2389-->
The reason of this unexpected behavior is that, with the current versions of MediaWiki, there are two caveats:
</translate>

* <translate><!--T:2390--> <tvar name=code0><syntaxhighlight lang="wikitext" inline>{{FULLPAGENAME}}</syntaxhighlight></tvar>, or even <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{FULLPAGENAME:Côte-d'Or}}</syntaxhighlight></tvar> ''may'' return the actually HTML-encoded string <tvar name=actual><code>Category:Côte-d&apos;Or</code></tvar> and not the expected <tvar name=expected><code>Category:Côte-d'Or</code></tvar>, and that:</translate>
* <translate><!--T:2391--> <tvar name=code1><syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: Côte-d&apos;Or }}</syntaxhighlight></tvar> '''unconditionally''' returns 0 (the PAGESINCAT magic keyword does not perform any HTML-decoding of its input parameter).</translate>

<translate>
<!--T:2392-->
The simple workaround using titleparts (which will continue to work if the two caveats are fixed in a later version of MediaWiki) is:
</translate>

* <syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{#titleparts: {{CURRENTPAGENAME}} }} }}</syntaxhighlight>
* <syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{#titleparts: {{PAGENAME:Category:Côte-d'Or}} }} }}</syntaxhighlight>
* <translate><!--T:2395--> <tvar name=code0><syntaxhighlight lang="wikitext" inline>{{PAGESINCAT: {{#titleparts: {{PAGENAME:Category:Côte-d&apos;Or}} }} }}</syntaxhighlight></tvar>, that all return the actual number of pages in the same category.</translate>
}}

<translate>
<!--T:2396-->
Then the decoded ''pagename'' is canonicalized into a standard page title supported by MediaWiki, as much as possible:
</translate>

# <translate><!--T:2371--> All underscores are automatically replaced with spaces:</translate>
#: <code><nowiki>{{#titleparts:</nowiki> {{ns:1}}:Foo/bah_boo|1|2{{))}}</code>  → '''{{#titleparts: {{ns:1}}:Foo/bah_boo|1|2}}''' <translate><!--T:2372--> Not <tvar name=1>bah_boo</tvar>, despite the underscore in the original.</translate>
# <translate><!--T:238--> The string is split a maximum of 25 times; further slashes are ignored and the 25th element will contain the rest of the string.</translate> <translate><!--T:2671--> The string is also limited to 255 characters, as it is treated as a [[<tvar name=1>Special:MyLanguage/Manual:Page table#Schema summary</tvar>|page title]]:</translate>
#: <code><nowiki>{{#titleparts:</nowiki> a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/{{highlight|y/z/aa/bb/cc/dd/ee}} <nowiki>| 1 | 25 }}</nowiki></code> → '''{{#titleparts: 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/aa/bb/cc/dd/ee | 1 | 25 }}'''
#: <translate><!--T:239--> If for whatever reason you needed to push this function to its limit, although very unlikely, it is possible to bypass the 25 split limit by nesting function calls:</translate>
#: <code><nowiki>{{#titleparts: {{#titleparts:</nowiki> 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/{{highlight|z}}/aa/bb/cc/dd/ee<nowiki>| 1 | 25 }} | 1 | 2}}</nowiki></code> → '''{{#titleparts: {{#titleparts: 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/aa/bb/cc/dd/ee | 1 | 25 }} | 1 | 2}}'''
# <translate><!--T:2398--> Finally the first substring is capitalized according to the capitalization settings of the local wiki (if that substring also starts by a local namespace name, that namespace name is also normalized).</translate>
#: <code><nowiki>{{#titleparts:</nowiki> {{highlight|talk:a}}/b/c <nowiki>}}</nowiki></code> → '''{{#titleparts: talk:a/b/c }}'''

{{tip| 1=
<translate>
<!--T:240-->
You can use #titleparts as a small "string parser and converter", but consider that it returns the first substring capitalized:
</translate>

: <code><nowiki>{{#titleparts:</nowiki> {{highlight|one}}/two/three/four<nowiki>|1|1 }}</nowiki></code> → '''{{#titleparts: one/two/three/four|1|1 }}'''
: <code><nowiki>{{#titleparts:</nowiki> one/{{highlight|two}}/three/four<nowiki>|1|2 }}</nowiki></code> → '''{{#titleparts: one/two/three/four|1|2 }}'''

<translate>
<!--T:241-->
If lower case is needed, use lc: function to control output:
</translate>

: <code><nowiki>{{lc: {{#titleparts:</nowiki> {{highlight|one}}/two/three/four<nowiki>|1|1 }} }}</nowiki></code> → '''{{lc: {{#titleparts:one/two/three/four|1|1 }}}}'''

<translate>
<!--T:242-->
You can prepend a 'dummy' slash at the beginning of the string to get the correct first substring capitalization (uppercase or lowercase). Use <tvar name=2><code>2</code></tvar> instead of <tvar name=p1><code>1</code></tvar> for ''first segment to return'':
</translate>

: <code><nowiki>{{#titleparts:</nowiki> '''/'''{{highlight|one}}/two/three/four<nowiki>|1|</nowiki>'''<u>2</u>'''<nowiki> }}</nowiki></code> → '''{{#titleparts: /one/two/three/four|1|2 }}'''
: <code><nowiki>{{#titleparts:</nowiki> '''/'''{{highlight|One}}/two/three/four<nowiki>|1|</nowiki>'''<u>2</u>'''<nowiki> }}</nowiki></code> → '''{{#titleparts: /One/two/three/four|1|2 }}'''
}}

{{Warning|1=
<translate><!--T:243--> Certain characters that are {{<tvar name=1>ll|Manual:Page title#Naming restrictions</tvar>|illegal in a page title}} will cause #titleparts to not parse the string:</translate>

: <syntaxhighlight lang="wikitext" inline>{{#titleparts: {one/two} | 1 | 1 }}</syntaxhighlight> → '''{{#titleparts: {one/two} | 1 | 1 }}'''. <translate><!--T:244--> Does not produce the expected:</translate> '''{one'''
: <syntaxhighlight lang="wikitext" inline>{{#titleparts: [[page]]/123 | 1 | 2 }}</syntaxhighlight> → '''{{#titleparts: [[page]]/123 | 1 | 2 }}'''. <translate><!--T:2399--> Does not work because brackets are illegal in page titles and this parser function does not process links embedded in its input ''pagename'' parameter, even when they use the MediaWiki syntax, or any other HTML or MediaWiki tags.</translate>
: <syntaxhighlight lang="wikitext" inline>{{#titleparts: red/#00FF00/blue | 1 | 3 }}</syntaxhighlight> → "{{#titleparts:  red/#00FF00/blue| 1 | 3 }}". <translate><!--T:2410--> Does not work because "#" is also illegal in page titles.</translate>
}}

{{Warning|1=
<translate>
<!--T:2429-->
If any part of the title is just "<tvar name=1><code>.</code></tvar>" or "<tvar name=2><code>..</code></tvar>", #titleparts will not parse the string:
</translate>

: <syntaxhighlight lang="wikitext" inline>{{#titleparts: one/./three | 1 | 1 }}</syntaxhighlight> → '''{{#titleparts: one/./three | 1 | 1 }}'''. <translate><!--T:2430--> The whole string is returned.  It does not produce the expected:</translate> '''one'''
}}

{{Warning|1=
<translate><!--T:245--> This function does not degrade gracefully if the input exceeds 255 bytes in UTF-8. If the input string is 256 bytes or more, the whole string is returned.</translate>
}}

<translate>
== String functions == <!--T:2468-->
</translate>
{{Main|Extension:ParserFunctions/String functions}}

<translate><!--T:2696--> The ParserFunctions extension optionally defines various string functions if <tvar name=2><code>$wgPFEnableStringFunctions</code></tvar> is set to <tvar name=3>{{phpi|true}}</tvar>:</translate>

* <code>#len</code>
* <code>#pos</code>
* <code>#rpos</code>
* <code>#sub</code>
* <code>#count</code>
* <code>#replace</code>
* <code>#explode</code>
* <code>#urldecode</code>

<translate><!--T:2702--> See the dedicated subpage for documentation, and <tvar name=2>{{ll|Manual:Performing string operations with parser functions}}</tvar> for examples.</translate>

{{Warning
|1=<translate><!--T:2697--> In 2013, it was decided that '''these functions will <em>never</em> be enabled on any Wikimedia wiki''', because they are inefficient when used on a large scale (see <tvar name=1>[[phab:T8455]]</tvar> for some history).</translate> '''<translate><!--T:2698--> These functions do <strong>NOT</strong> work on Wikimedia wikis!</translate>'''<br/><br/><translate><!--T:2699--> If you are here to write something on a Wikimedia project, you are looking for something else: if your home wiki has string functions, it probably uses <tvar name=1>{{ll|Extension:Scribunto|Lua}}</tvar>.</translate> <translate><!--T:2700--> For example, the English Wikipedia uses [[w:Module:String|Module:String]], which does some of the same things with wildly different syntax.</translate> <translate><!--T:2701--> There are also individual [[w:Template:String-handling templates|String-handling templates]].</translate>
}}

<translate><!--T:2703--> Here is a short overview of <tvar name=1>[[Module:String]]</tvar> functions:</translate>

* [[w:Module:String#len|&#35;len]] <translate><!--T:2704--> (length of string)</translate>: <code><nowiki>{{</nowiki>[[:w:Wikipedia:Lua|#invoke]]<nowiki>:String|len|target_string}}</nowiki></code>
* [[w:Module:String#sub|&#35;sub]] <translate><!--T:2705--> (substring)</translate>: <code><nowiki>{{#invoke:String|sub|target_string|start_index|end_index}}</nowiki></code>
* [[w:Module:String#match|&#35;match]]: <code><nowiki>{{#invoke:String|match|source_string|pattern_string|start_index|match_number|plain_flag|nomatch_output}}</nowiki></code>
* [[w:Module:String#pos|&#35;pos]] <translate><!--T:2706--> (position of target)</translate>: <code><nowiki>{{#invoke:String|pos|target_string|index_value}}</nowiki></code>
* [[w:Module:String#find|&#35;find]]: <code><nowiki>{{#invoke:String|find|source_string|target_string|start_index|plain_flag}}</nowiki></code>
* [[w:Module:String#replace (gsub)|&#35;replace]]: <code><nowiki>{{#invoke:String|replace|source_str|pattern_string|replace_string|replacement_count|plain_flag}}</nowiki></code>
* [[w:Module:String#rep|&#35;rep]] <translate><!--T:2707--> (repeat)</translate>: <code><nowiki>{{#invoke:String|rep|source|count}}</nowiki></code>
* [[w:Module:String#escapePattern|&#35;escapePattern]]: <code><nowiki>{{#invoke:String|escapePattern|pattern_string}}</nowiki></code>
* [[w:Module:String#count|&#35;count]]: <code><nowiki>{{#invoke:String|count|source_str|pattern_string|plain_flag}}</nowiki></code>
* [[w:Module:String#join|&#35;join]]: <code><nowiki>{{#invoke:String|join|separator|string1|string2|...}}</nowiki></code>

<translate>
== General points == <!--T:269-->

=== Substitution === <!--T:270-->

<!--T:271-->
Parser functions can be [[<tvar name=help>Special:MyLanguage/Help:Substitution</tvar>|substituted]] by prefixing the hash character with <tvar name=subst><code>'''subst:'''</code></tvar>:
</translate>

:<code><nowiki>{{subst:#ifexist:</nowiki> {{FULLPAGENAME}} <nowiki>| [[</nowiki>{{FULLPAGENAME}}<nowiki>]] | </nowiki>{{FULLPAGENAME}}<nowiki> }}</nowiki></code> → <translate><!--T:247--> the code <tvar name=1>'''<code><nowiki>[[</nowiki>{{FULLPAGENAME}}<nowiki>]]</nowiki></code>'''</tvar> will be inserted in the wikitext since the page <tvar name=2>{{FULLPAGENAME}}</tvar> exists.</translate>

{{Warning|1=
<translate>
<!--T:248-->
The results of substituted parser functions are undefined if the expressions contain ''un''substituted volatile code such as {{<tvar name=magic>ll|Help:Magic words#Variables</tvar>|variables}} or other parser functions. For consistent results, all the volatile code in the expression to be evaluated must be substituted. See [[w:Help:Substitution|Help:Substitution]].</translate>}}

<translate>
<!--T:2411-->
Substitution does not work within <tvar name=code1>{{ll|Extension:Cite#Substitution and embedded parser functions|{{tag|ref|open}}&hellip;{{tag|ref|close}} }}</tvar>; you can use <tvar name=code2><code>{{((}}subst:#tag:ref{{!}}</code>&hellip;<code>{{))}}</code></tvar> for this purpose.

=== Redirects === <!--T:262-->

<!--T:272-->
Especially <tvar name=time>[[#.23time|<nowiki>{{</nowiki><code>#time:</code>&hellip;&vert;<code>now-</code>&hellip;<nowiki>}}</nowiki>]]</tvar> could be handy in {{<tvar name=redirects>ll|Help:Redirects</tvar>|redirects}} to pages including dates, but this does not work.

=== Escaping pipe characters === <!--T:249-->

==== In tables ==== <!--T:2742-->
</translate>
<translate><!--T:250--> Parser functions will mangle {{<tvar name=help-tables>ll|Help:Tables</tvar>|wikitable}} syntax and pipe characters (<tvar name=pipe><code><nowiki>|</nowiki></code></tvar>), treating all the raw pipe characters as parameter dividers.</translate>
<translate><!--T:2664--> To avoid this, most wikis used a template <tvar name=template>[[{{ns:10}}:!]]</tvar> with its contents only a raw pipe character (<tvar name=pipe><code><nowiki>|</nowiki></code></tvar>), since MW 1.24 a {{<tvar name=help-magic>ll|Help:Magic words#Other|<code><nowiki>{{!}}</nowiki></code></tvar> magic word}} replaced this kludge.</translate>
<translate><!--T:2665--> This 'hides' the pipe from the MediaWiki parser, ensuring that it is not considered until after all the templates and variables on a page have been expanded.</translate>
<translate><!--T:2666--> It will then be interpreted as a table row or column separator.</translate>
<translate><!--T:2667--> Alternatively, raw HTML table syntax can be used, although this is less intuitive and more error-prone.</translate>

<translate>
<!--T:251-->
You can also escape the pipe character <tvar name=1>&vert;</tvar> for display as a plain, uninterpreted character using an HTML entity: <tvar name=entity><code>&amp;#124;</code></tvar> or <tvar name=2><code>&amp;vert;</code></tvar> .
</translate>

{| class="wikitable"
! <translate><!--T:252--> Description</translate>
! <translate><!--T:253--> You type</translate>
! <translate><!--T:254--> You get</translate>
|-
| <translate><!--T:255--> Escaping pipe character as table row/column separator</translate>
| <pre>
{{!}}
</pre>
| style="text-align:center;" | &#124;
|-
| <translate><!--T:256--> Escaping pipe character as a plain character</translate>
| <pre>
&amp;#124;
</pre>
| style="text-align:center;" | &#124;
|}

<translate>
==== In template calls ==== <!--T:2743-->

<!--T:2744-->
The same pipe protection applies as for the following example:
</translate>

<syntaxhighlight lang="wikitext">
{{Documentation
|content=
... text before ...<code>subpage-name=sandbox3</code> is equivalent to <code>sandbox link=../sandbox3 | sandbox name=sandbox3</code>.
... text after ...
}}
</syntaxhighlight>

<translate>
<!--T:2745-->
We observe that <tvar name=1>''text after''</tvar> is not displayed when the pipe <tvar name=2>|</tvar> just before <tvar name=3>''sandbox name=''</tvar> is present since <tvar name=4>''|sandbox name=''</tvar> is considered erroneously to be a parameter of template <tvar name=5>''Documentation''</tvar> at the same level as <tvar name=6>''|content=''</tvar> is.

=== Stripping whitespace === <!--T:257-->

<!--T:258-->
Whitespace, including newlines, tabs, and spaces, is stripped from the beginning and end of all the parameters of these parser functions. If this is not desirable, comparison of strings can be done after putting them in quotation marks.
</translate>

:<code><nowiki>{{#ifeq: foo&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; | &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;foo | equal | not equal }}</nowiki></code> → '''{{#ifeq: foo            |            foo | equal | not equal }}'''
:<code><nowiki>{{#ifeq: "foo&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;" | "&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;foo" | equal | not equal }}</nowiki></code> → '''{{#ifeq: "foo           " | "           foo" | equal | not equal }}'''

<translate><!--T:259--> To prevent the trimming of then and else parts, see <tvar name=1>[[m:Template:If]]</tvar>.</translate>
<translate><!--T:2774--> Some people achieve this by using <tvar name=2><nowiki>&lt;</nowiki>{{ll|Wiki markup#Nowiki|nowiki}}<nowiki>&gt;</nowiki> <nowiki>&lt;</nowiki>/nowiki<nowiki>&gt;</nowiki></tvar> instead of spaces.</translate>

:<syntaxhighlight lang="wikitext" inline>foo{{#if:|| bar }}foo</syntaxhighlight> → '''foo{{#if:|| bar }}foo'''
:<syntaxhighlight lang="xml" inline>foo{{#if:||<nowiki /> bar <nowiki />}}foo</syntaxhighlight> → '''foo{{#if:||<nowiki /> bar <nowiki />}}foo'''

<translate>
<!--T:313-->
However, this method can be used to render a '''single''' whitespace character only, since the parser squeezes multiple whitespace characters in a row into one.
</translate>

:{|
|<syntaxhighlight lang="html"><span style="white-space: pre;">foo{{#if:||<nowiki/>      bar      <nowiki/>}}foo</span></syntaxhighlight>|| → ||'''<span style="white-space: pre;">foo{{#if:||<nowiki/>      bar      <nowiki/>}}foo</span>'''
|}

<translate>
<!--T:315-->
In this example, the <tvar name=1><code>white-space: pre</code></tvar> style is used to force the whitespace to be preserved by the browser, but even with it the spaces are not shown.  This happens because the spaces are stripped by the software, before being sent to the browser.

<!--T:316-->
It is possible to workaround this behavior replacing whitespaces with <tvar name=1><code>&amp;#32;</code></tvar> (''breakable space'') or <tvar name=2><code>&amp;nbsp;</code></tvar> (''non-breakable space''), since they are not modified by the software:
</translate>

:<syntaxhighlight lang="html" inline><span style="white-space: pre;">foo{{#if:||&#32;&#32;&#32;bar&#32;&#32;&#32;}}foo</span></syntaxhighlight> → '''<span style="white-space: pre;">foo{{#if:||&#32;&#32;&#32;bar&#32;&#32;&#32;}}foo</span>'''
:<syntaxhighlight lang="html" inline>foo{{#if:||&nbsp;&nbsp;&nbsp;bar&nbsp;&nbsp;&nbsp;}}foo</syntaxhighlight> → '''foo{{#if:||&nbsp;&nbsp;&nbsp;bar&nbsp;&nbsp;&nbsp;}}foo'''

<translate><!--T:2684--> Beware that not all parameters are created equal.</translate>
<translate><!--T:2685--> In ParserFunctions, whitespace at the beginning and end is always stripped.</translate>
<translate><!--T:2686--> In {{<tvar name=1>ll|Help:Templates</tvar>|templates}}, whitespace at the beginning and end is stripped for named parameters and named unnamed parameters but ''not'' from unnamed parameters:</translate>

:<code>foo{{tlx|1x|content{{=}} bar }}foo</code> → '''foo{{1x|content= bar }}foo'''
:<code>foo{{tlx|1x|1{{=}} bar }}foo</code> → '''foo{{1x|1= bar }}foo'''
:<code>foo{{tlx|1x|&nbsp;bar&nbsp;}}foo</code> → '''foo{{1x| bar }}foo'''

<translate>
== Other parser functions == <!--T:2724-->

=== Case conversion functions === <!--T:2725-->
</translate>
* <translate><!--T:2726--> '''Lowercase:'''</translate> {{xpdoc|lc: AbC|}}
* <translate><!--T:2727--> '''Uppercase:'''</translate> {{xpdoc|uc: AbC|}}
* <translate><!--T:2728--> '''Lowercase first character:'''</translate> {{xpdoc|lcfirst: AbC}}
* <translate><!--T:2729--> '''Uppercase first character:'''</translate> {{xpdoc|ucfirst: abc}}

<translate>
=== Encoding functions === <!--T:2730-->
</translate>
* <translate><!--T:2731--> '''URL encoding:'''</translate>
<syntaxhighlight lang="wikitext">
"{{urlencode: AbC
dEf ghi}}"
</syntaxhighlight>
<translate><!--T:2732--> renders as</translate>
----
"{{urlencode: AbC
dEf ghi}}"
----
<translate>
<!--T:2733-->
So inner new lines convert into %0A, and inner spaces convert into +.

=== Anchor encoding === <!--T:2734-->
</translate>
<syntaxhighlight lang="wikitext">
{{anchorencode: AbC dEf ghi}}
</syntaxhighlight>
<translate><!--T:2735--> renders as</translate>
---- 
{{anchorencode: AbC dEf ghi}}
----

<translate>
== See also == <!--T:260-->
</translate>
* {{ll|Help:Parser functions in templates}}
* {{ll|Parser extension tags}}
* {{ll|Parser function hooks}} – <translate><!--T:2687--> an (incomplete) list of parser functions added by core and extensions.</translate>
* {{ll|Manual:Newlines and spaces}}
* {{ll|Manual:Converting between ParserFunctions syntax and TeX syntax}}
* {{ll|Help:Magic words}} – <translate><!--T:2746--> in particular for number formatting and padding</translate>
* {{ll|Manual:Template limits#Expensive parser function calls}}
<translate>
<!--T:2412-->
* <tvar name=1>[[Module:String]]</tvar> obsoleting <tvar name=2>{{ll|Extension:StringFunctions}}</tvar>
</translate>
* <translate><!--T:2688--> Parser functions for Wikibase (the extensions that enables Wikidata):</translate> [[d:Special:MyLanguage/Wikidata:How to use data on Wikimedia projects#Parser function|<translate><!--T:2738--> How to use data on Wikimedia projects#Parser function</translate>]]

<translate>
== References == <!--T:2431-->
</translate>
<references />
{{Parser functions nav}}

[[Category:Extension help{{#translation:}}|ParserFunctions]]
[[Category:Magic words{{#translation:}}]]