Help:Extension:ParserFunctions/sv

From MediaWiki.org
Jump to: navigation, search
This page is a translated version of the page Help:Extension:ParserFunctions and the translation is 44% complete.

Other languages:
български • ‎বাংলা • ‎Deutsch • ‎Zazaki • ‎English • ‎British English • ‎español • ‎français • ‎italiano • ‎日本語 • ‎Lëtzebuergesch • ‎मराठी • ‎polski • ‎پښتو • ‎português • ‎português do Brasil • ‎русский • ‎svenska • ‎தமிழ் • ‎ไทย • ‎Türkçe • ‎中文

Programtillägget ParserFunctions ger tio extra parserfunktioner som kompleterar de magiska orden som följer med MediaWiki. Grundsyntaxen på alla perserfunktionerna på den här sidan ser ut som följer:

{{#funktionens-namn: argument 1 | argument 2 | argument 3 ... }}

#expr

Typ Operator
Gruppering (parenteser) ( )
Siffror 1234.5   e (2.718)   pi (3.142)
binär operatör e   unär +,-
Unär not ceil trunc floor abs exp ln sin cos tan acos asin atan
Binär ^
* / div mod
+ -
Avrundning round
Logisk = != <> > < >= <=
and
or

This function evaluates a mathematical expression and returns the calculated value.

{{#expr: uttryck }}

Alla tillgängliga operatorer finns i tabellen till höger. Se Help:Calculation för mer detaljerad information om varje operator. Noggrannheten och formatet på resultatet beror på operativsystemet och språkinställningarna i MediaWiki-installationen.

När man använder boolesk algebra, ersätts noll med false och alla nollskilda värden, positiva och negativa, ersätts med true.

{{#expr: 1 and -1 }}1
{{#expr: 1 and 0 }}0

Ett tomt input-uttryck returnerar en tom sträng. Ogiltiga uttryck returnerar en av följande felmeddelanden, som kan fångas med #iferror:-funktionen:

{{#expr: }}
{{#expr: 1+ }}Expression error: Missing operand for +.
{{#expr: 1 foo 2 }}Expression error: Unrecognized word "foo".

Den här funktionen testar om en sträng är tom eller inte. Den svarar med false om strängen är tom eller bara innehåller mellanslag och radbrytningar.

{{#expr: +1 }}1
{{#expr: -1 }}-1
{{#expr: + 1 }}1
{{#expr: - 1 }}-1

Note, if using the output of magic words, you must raw format them in order to remove commas, and translate the numerals. For example {{NUMBEROFUSERS}} results in {{NUMBEROFUSERS}}, where we want {{NUMBEROFUSERS}} which can be obtained by {{NUMBEROFUSERS}}. This is especially important in some languages, where numerals are translated. For example, in Bengali, {{NUMBEROFUSERS}} produces ৩০,০৬১.

{{#expr:{{NUMBEROFUSERS}}+100}}Expression error: Unrecognized punctuation character "�".
{{#expr:{{formatnum:{{NUMBEROFUSERS}}|R}}+100}}1873552


Varning Varning: Operatorn mod ger felaktiga resultat för en del värden i det andra argumentet:
{{#expr: 123 mod (2^64-1)}}Division by zero. (producerar en tom sträng; borde vara 123)
If you want to do calculations based on dates (If current date and time is after some other date and time), first convert the time to number of seconds after January 1, 1970 using {{#time: U }}, then you can simply add and subtract dates.

Avrundning

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.

To round up or down use unary ceil or floor respectively.

Testfall Resultat Avrundningsmetod
{{#expr: 1/3 round 5 }} 0.33333 Final digit is < 5, so no apparent rounding occurs
{{#expr: 1/6 round 5 }} 0.16667 Final digit is >= 5, so it is rounded up
{{#expr: 8.99999/9 round 5 }} 1 Again, the result is rounded up on the last digit, which results in additional rounding
{{#expr: 1234.5678 round -2 }} 1200 Rounded to nearest 100 because negative values round to the left of the decimal point
{{#expr: 1234.5678 round 2 }} 1234.57 Rounded to nearest 100th since positive values round to the right of the decimal point
{{#expr: 1234.5678 round 2.3 }} 1234.57 Decimals in the rounding index make no difference in the rounded result
{{#expr: trunc 1234.5678 }} 1234 Decimal portion truncated (chopped off)
Avrundar till det närmaste heltalet
{{#expr: 1/3 round 0 }} 0 Ned till det "närmaste" heltalet, som är noll
{{#expr: 1/2 round 0 }} 1 Upp till det närmaste heltalet, som är ett
{{#expr: 3/4 round 0 }} 1 Upp till det närmaste heltalet, som är ett
{{#expr: -1/3 round 0 }} -0 Upp till det närmaste heltalet, som är noll
Varning Varning: The negative sign on zero could cause equality mismatches
{{#expr: -1/2 round 0 }} -1 Ned till det närmaste heltalet, som är minus ett
{{#expr: -3/4 round 0 }} -1 Ned till det närmaste heltalet, som är minus ett
Avrundar upp eller ned med ceil och floor
{{#expr: ceil(1/3) }} 1 Upp till nästa större heltal, som är ett
{{#expr: floor(1/3) }} 0 Ned till nästa mindre heltal, som är noll
{{#expr: ceil(-1/3) }} -0 Upp till nästa större heltal, som är noll
Varning Varning: Minustecknet vid nollan kan orsaka felaktigheter vid jämförelse
{{#expr: floor(-1/3) }} -1 Ned till det närmaste mindre heltalet, som är minus noll
{{#expr: ceil 1/3 }} 0.33333333333333 Inte avrundad eftersom 1 redan är ett heltal
Varning Varning: Tolkas som (ceil 1)/3, inte ceil(1/3) som du kanske förväntade dig

#if

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.

{{#if: teststräng | värde om teststrängen inte är tom | värde om teststrängen är tom (eller endast ett mellanslag) }}
{{#if: första parametern | andra parametern | tredje parametern }}

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.

{{#if: | yes | no}}no
{{#if: string | yes | no}}yes
{{#if:      | yes | no}}no
{{#if:


| yes | no}}
no

The test string is always interpreted as pure text, so mathematical expressions are not evaluated:

{{#if: 1==2 | yes | no }}yes
{{#if: 0 | yes | no }}yes

Either or both the return values may be omitted:

{{#if: foo | yes }}yes
{{#if: | yes }}
{{#if: foo | | no}}

The function may be nested. To do so, nest the inner #if function in its full form in place of the third parameter of the enclosing #if function. A depth of at least seven levels of nesting is possible, although that may depend on the wiki or a memory limit.

  • Example: {{#if: test string | value if test string is not empty | {{#if: test string | value if test string is not empty | value if test string is empty (or only white space) }} }}

See Help:Parser functions in templates for more examples of this parser function.

#ifeq

This parser function compares two strings and determines whether they are identical.

{{#ifeq: string 1 | string 2 | value if identical | value if different }}

If both strings are valid numerical values, the strings are compared numerically:

{{#ifeq: 01 | 1 | equal | not equal}}equal
{{#ifeq: 0 | -0 | equal | not equal}}equal
{{#ifeq: 1e3 | 1000 | equal | not equal}}equal
{{#ifeq: {{#expr:10^3}} | 1000 | equal | not equal}}equal

Otherwise the comparison is made as text; this comparison is case sensitive:

{{#ifeq: foo | bar | equal | not equal}}not equal
{{#ifeq: foo | Foo | equal | not equal}}not equal
{{#ifeq: "01" | "1" | equal | not equal}}not equal  (compare to similar example above, without the quotes)
{{#ifeq: 10^3 | 1000 | equal | not equal}}not equal  (compare to similar example above, with #expr)

As a practical example, given an existing template Template:Size used to set standard short and long sizes, defined as:

{{#ifeq: {{{1|}}} | short | 20 | 40}}

the following ensue:

{{size|short}}20
{{size|20}}40
{{size}}40
Varning Varning: Numerical comparisons with #ifeq and #switch are not equivalent to comparisons with #expr:
{{#ifeq: 12345678901234567 | 12345678901234568 | equal | not equal}}not equal
{{#switch: 12345678901234567 | 12345678901234568 = equal | not equal}}not equal

because PHP compares two numbers of type integer, whereas:

{{#ifexpr: 12345678901234567 = 12345678901234568 | equal | not equal}}equal

because MediaWiki converts literal numbers in expressions to type float, which, for large integers like these, involves rounding.

Varning Varning: Tags and parser functions inside parser tags (such as <nowiki>) are temporarily replaced by a unique code. This affects comparisons:
{{#ifeq: <nowiki>foo</nowiki> | <nowiki>foo</nowiki> | equal | not equal}}not equal
{{#ifeq: <math>foo</math> | <math>foo</math> | equal | not equal}}not equal
{{#ifeq: {{#tag:math|foo}} | {{#tag:math|foo}} | equal | not equal}}not equal
{{#ifeq: [[foo]] | [[foo]] | equal | not equal}}equal

If the strings to be compared are given as equal calls to the same 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.

Varning Varning:

Be careful when comparing against the current page title using the page name magic words. These magic words convert special characters into numeric HTML entities. This may result in misleading results. For example, if you are on a page titled "L'Aquila"...

{{#ifeq: L'Aquila | {{FULLPAGENAME}} | equal | not equal}}not equal

To work around this, apply the magic word to both parameters:

{{#ifeq: {{FULLPAGENAME: L'Aquila}} | {{FULLPAGENAME}} | equal | not equal}}equal

#iferror

This function takes an input string and returns one of two results; the function evaluates to true if the input string contains an HTML object with class="error", as generated by other parser functions such as #expr, #time and #rel2abs, template errors such as loops and recursions, and other "failsoft" parser errors.

{{#iferror: test string | value if error | value if correct }}

One or both of the return strings can be omitted. If the correct string is omitted, the test string is returned if it is not erroneous. If the error string is also omitted, an empty string is returned on an error:

{{#iferror: {{#expr: 1 + 2 }} | error | correct }}correct
{{#iferror: {{#expr: 1 + X }} | error | correct }}error
{{#iferror: {{#expr: 1 + 2 }} | error }}3
{{#iferror: {{#expr: 1 + X }} | error }}error
{{#iferror: {{#expr: 1 + 2 }} }}3
{{#iferror: {{#expr: 1 + X }} }}
{{#iferror: <strong class="error">a</strong> | error | correct }}error

#ifexpr

This function evaluates a mathematical expression and returns one of two strings depending on the boolean value of the result:

{{#ifexpr: expression | value if true | value if false }}

The expression input is evaluated exactly as for #expr above, with the same operators being available. The output is then evaluated as a boolean expression.

An empty input expression evaluates to false:

{{#ifexpr: | yes | no}}no

As mentioned above, zero evaluates to false and any nonzero value evaluates to true, so this function is equivalent to one using #ifeq and #expr only:

{{#ifeq: {{#expr: expression }} | 0 | value if false | value if true }}

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 value if true).

{{#ifexpr: = | yes | no }}Expression error: Unexpected = operator.

jämfört med

{{#ifeq: {{#expr: = }} | 0 | no | yes }}yes

Either or both the return values may be omitted; no output is given when the appropriate branch is left empty:

{{#ifexpr: 1 > 0 | yes }}yes
{{#ifexpr: 1 < 0 | yes }}
{{#ifexpr: 0 = 0 | yes }}yes
{{#ifexpr: 1 > 0 | | no}}
{{#ifexpr: 1 < 0 | | no}}no
{{#ifexpr: 1 > 0 }}

#ifexist

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.

{{#ifexist: page title | value if exists | value if doesn't exist }}

The function evaluates to true if the page exists, whether it contains content, is visibly blank (contains meta-data such as category links or magic words, but no visible content), is blank, or is a redirect. Only pages that are redlinked evaluate to false, including if the page used to exist but has been deleted.

{{#ifexist: Help:Extension:ParserFunctions/sv | exists | doesn't exist }}exists
{{#ifexist: XXHelp:Extension:ParserFunctions/svXX | exists | doesn't exist }}doesn't exist

The function evaluates to true for system messages that have been customized, and for special pages that are defined by the software.

{{#ifexist: Special:Watchlist | exists | doesn't exist }}exists
{{#ifexist: Special:CheckUser | exists | doesn't exist }}exists (because the CheckUser extension is installed on this wiki)
{{#ifexist: MediaWiki:Copyright | exists | doesn't exist }}exists (eftersom MediaWiki:Copyright har anpassats)

If a page checks a target using #ifexist:, then that page will appear in the Special:WhatLinksHere list for the target page. So if the code {{#ifexist:Foo}} were included live on this page (Help:Extension:ParserFunctions/sv), Special:WhatLinksHere/Foo will list Help:Extension:ParserFunctions/sv.

On wikis using a shared media repository, #ifexist: can be used to check if a file has been uploaded to the repository, but not to the wiki itself:

{{#ifexist: File:Example.png | exists | doesn't exist }}doesn't exist
{{#ifexist: Image:Example.png | exists | doesn't exist }}doesn't exist
{{#ifexist: Media:Example.png | exists | doesn't exist }}exists

If a local description page has been created for the file, the result is exists for all of the above.

#ifexist: fungerar inte med interwikilänkar.

ifexist limits

#ifexist: is considered an "expensive parser function"; only a limited number of which can be included on any one page (including functions inside transcluded templates). When this limit is exceeded, any further #ifexist: functions automatically return false, whether the target page exists or not, and the page is categorized into Category:Pages with too many expensive parser function calls. The name of the tracking category may vary depending on the content language of your wiki.

For some use cases it is possible to emulate the ifexist effect with css, by using the selectors a.new (to select links to unexisting pages) or a:not(.new) (to select links to existing pages). Furthermore, since the number of expensive parser functions that can be used on a single page is controlled by $wgExpensiveParserFunctionLimit, one can also increase the limit in LocalSettings.php if needed.

#rel2abs

This function converts a relative file path into an absolute filepath.

{{#rel2abs: path }}
{{#rel2abs: path | base path }}

Within the path input, the following syntax is valid:

  • . → the current level
  • .. → "go up one level"
  • /foo → "go down one level into the subdirectory /foo"

If the base path is not specified, the full page name of the page will be used instead:

{{#rel2abs: /quok | Help:Foo/bar/baz }}Help:Foo/bar/baz/quok
{{#rel2abs: ./quok | Help:Foo/bar/baz }}Help:Foo/bar/baz/quok
{{#rel2abs: ../quok | Help:Foo/bar/baz }}Help:Foo/bar/quok
{{#rel2abs: ../. | Help:Foo/bar/baz }}Help:Foo/bar

Invalid syntax, such as /. or /./, is ignored. Since no more than two consecutive full stops are permitted, sequences such as these can be used to separate successive statements:

{{#rel2abs: ../quok/. | Help:Foo/bar/baz }}Help:Foo/bar/quok
{{#rel2abs: ../../quok | Help:Foo/bar/baz }}Help:Foo/quok
{{#rel2abs: ../../../quok | Help:Foo/bar/baz }}quok
{{#rel2abs: ../../../../quok | Help:Foo/bar/baz }}Error: Invalid depth in path: "Help:Foo/bar/baz/../../../../quok" (tried to access a node above the root node).

#switch

This function compares one input value against several test cases, returning an associated string if a match is found.

{{#switch: comparison string
 | case = result
 | case = result
 | ...
 | case = result
 | default result
}}

Exempel:

{{#switch: baz | foo = Foo | baz = Baz | Bar }}Baz
{{#switch: foo | foo = Foo | baz = Baz | Bar }}Foo
{{#switch: zzz | foo = Foo | baz = Baz | Bar }}Bar

#switch allows an editor [to add information in one template and this information will be visible in several other templates which all have different formatting.][clarification needed][examples needed]

Standard

Standardresultatet returneras om ingen målsträng överensstämmer med jämförelsesträngen:

{{#switch: test | foo = Foo | baz = Baz | Bar }}Bar

In this syntax, the default result must be the last parameter and must not contain a raw equals sign.

{{#switch: test | Bar | foo = Foo | baz = Baz }} →
{{#switch: test | foo = Foo | baz = Baz | B=ar }} →

Alternatively, the default result may be explicitly declared with a case string of "#default".

{{#switch: comparison string
 | case = result
 | case = result
 | ...
 | case = result
 | #default = default result
}}

Default results declared in this way may be placed anywhere within the function:

{{#switch: test | foo = Foo | #default = Bar | baz = Baz }}Bar

If the default parameter is omitted and no match is made, no result is returned:

{{#switch: test | foo = Foo | baz = Baz }}

Grouping results

It is possible to have 'fall through' values, where several case strings return the same result string. This minimizes duplication.

{{#switch: comparison string
 | case1 = result1
 | case2 
 | case3 
 | case4 = result2
 | case5 = result3
 | case6 
 | case7 = result4
 | #default = default result
}}

Here cases 2, 3 and 4 all return result2; cases 6 and 7 both return result4

Comparison behavior

As with #ifeq, 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:

{{#switch: 0 + 1 | 1 = one | 2 = two | three}} → three
{{#switch: {{#expr: 0 + 1}} | 1 = one | 2 = two | three}} → one
{{#switch: a | a = A | b = B | C}} → A
{{#switch: A | a = A | b = B | C}} → C

A case string may be empty:

{{#switch: | = Nothing | foo = Foo | Something }}Nothing

Once a match is found, subsequent cases are ignored:

{{#switch: b | f = Foo | b = Bar | b = Baz | }}Bar
Varning Varning: Numerical comparisons with #switch and #ifeq are not equivalent with comparisons in expressions (see also above):
{{#switch: 12345678901234567 | 12345678901234568 = A | B}} → B
{{#ifexpr: 12345678901234567 = 12345678901234568 | A | B}} → A

Raw equal signs

"Case" strings cannot contain raw equals signs. To work around this, create a template {{=}} containing a single equals sign: =.

Exempel:

{{#switch: 1=2
 | 1=2 = raw
 | 1<nowiki>=</nowiki>2 = nowiki
 | 1&#61;2 = html
 | 1{{=}}2 = template
 | default
}}html

Note Note: For a simple real life example of the use of this function, check Template:NBA color. Two complex examples can be found at Template:Extension and Template:BOTREQ.

Replacing #ifeq

#switch can be used to reduce expansion depth.

Till exempel:

  • {{#switch:{{{1}}} |condition1=branch1 |condition2=branch2 |condition3=branch3 |branch4}}

is equivalent to

  • {{#ifeq:{{{1}}}|condition1 |branch1 |{{#ifeq:{{{1}}}|condition2 |branch2 |{{#ifeq:{{{1}}}|condition3 |branch3 |branch4}}}}}}
(i.e. {{#ifeq:{{{1}}}|condition1 |<!--then-->branch1 |<!--else-->{{#ifeq:{{{1}}}|condition2 |<!--then-->branch2 |<!--else-->{{#ifeq:{{{1}}}|condition3 |<!--then-->branch3 |<!--else-->branch4}}}}}})

#time

Kod Beskrivning Aktuell utmatning
(Rensa denna sidas cache för att uppdatera)
År
Y Fyrsiffrigt årtal. 2014
y Tvåsiffrigt årtal. 14
L 1 om det är en skottår, annars 0. 0
o ¹ Den valda veckans år i formatet ISO-8601 ² 2014 ³

¹ Kräver PHP 5.1.0 och nyare samt rev:45208.
² 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.
³ Will output literal o if ¹ not fulfilled.

Månad
n Månadsindex, utan en nolla. 10
m Månadsindex, med en nolla. 10
M En förkortning av månadens namn på webbplatsens språk. okt
F Månadens fullständiga namn på webbplatsens språk. oktober
xg Output the full month name in the genitive form for site languages that distinguish between genitive and nominative forms. This option is useful for many Slavic languages like Polish, Russian, Belarusian, Czech, Slovak, Slovene, Ukrainian, etc. För polska:

(nominative)

{{#time:d F Y|20 June 2010|pl}} → 20 czerwiec 2010

(genitive)

{{#time:d xg Y|20 June 2010|pl}} → 20 czerwca 2010

Vecka
W Veckans nummer i formatet ISO 8601 med en nolla. 43
Dag
j Månadens dag, utan en nolla. 25
d Månadens dag, med en nolla. 25
z Day of the year (January 1 = 0).
Note Note: To get the ISO day of the year add 1.
297
D En förkortning för veckans dag. Översätts sällan. lör
l Veckodagens fullständiga namn. Översätts sällan. lördag
N Veckans dag i formatet ISO 8601 (Måndag = 1, söndag = 7). 6
w Number of the day of the week (Sunday = 0, Saturday = 6). 6
Timme
a "am" under morgonen (00:00:00 → 11:59:59), annars "pm" (12:00:00 → 23:59:59). am
A Versal version av a ovan. AM
g Timme i 12-timmarsformat, utan en nolla. 8
h Timme i 12-timmarsformat, med en nolla. 08
G Timme i 24-timmarsformat, utan en nolla. 8
H Timme i 24-timmarsformat, med en nolla. 08
Minuter och sekunder
i Minutes past the hour, zero-padded. 18
s Seconds past the minute, zero-padded. 01
U Sekunder sedan 1 januari 1970 00:00:00 GMT. 1414225081
Tidzon (från och med 1.22wmf2)
e Tidzonsidentiferare. UTC
I Whether or not the date is in daylight savings time. 0
O Difference to Greenwich time (GMT) +0000
P Difference to Greenwich time (GMT), with colon +00:00
T Timezone abbreviation. UTC
Z Timezone offset in seconds. 0
Övrigt
t Antal dagar i den aktuella månaden. 31
c ISO 8601 formatted date, equivalent to Y-m-dTH:i:s+00:00. 2014-10-25T08:18:01+00:00
r RFC 5322 formatted date, equivalent to D, j M Y H:i:s +0000, with weekday name and month name not internationalized. Sat, 25 Oct 2014 08:18:01 +0000
Icke-gregorianska kalendrar
Islamisk
xmj Månadens dag. 1
xmF Månadens fullständiga namn. Muharram
xmn Månadsindex. 1
xmY Fullständigt år. 1436
Iranian (Jalaly)
xij Månadens dag. 3
xiF Månadens fullständiga namn. Aban
xin Månadsindex. 8
xiY Fullständigt år. 1393
xiy Tvåsiffrigt årtal. 93
Hebreisk
xjj Månadens dag. 1
xjF Månadens fullständiga namn. Cheshvan
xjt Antal dagar i månaden. 29
xjx Genitive form of the month name. Cheshvan
xjn Månadens nummer. 2
xjY Fullständigt år. 5775
Thailändsk kalender
xkY Full year in Thai solar calendar.
Note Note: For years before 1941 the dates in Jan-Mar range are not calculated properly.
2557
Minguo/Juche year
xoY Fullständigt år. 103
Japanska perioder
xtY Fullständigt år. 平成26
Flaggor
xn Format the next numeric code as a raw ASCII number. På hindi kommer {{#time:H, xnH}} visas som ०६, 06.
xN Like xn, but as a toggled flag, which endures until the end of the string or until the next appearance of xN in the string.
xr Format the next number as a roman numeral. Only works for numbers up to 10,000
(upto 3,000 in pre MediaWiki 1.20).
{{#time:xrY}} → MMXIV
xh Format the next number as a Hebrew numeral. {{#time:xhY}} → ב'י"ד

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 magic word {{CURRENTTIMESTAMP}} – that is, the time the page was last rendered into HTML.

{{#time: format string }}
{{#time: format string | date/time object }}
{{#time: format string | date/time object | language code }}

The list of accepted formatting codes is given in the table to the right. 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). There are also two ways to escape characters within the formatting string:

  1. A backslash followed by a formatting character is interpreted as a single literal character
  2. Characters enclosed in double quotes are considered literal characters, and the quotes are removed.

In addition, the digraph xx is interpreted as a single literal "x".

{{#time: Y-m-d }}2014-10-25
{{#time: [[Y]] m d }}2014 10 25
{{#time: [[Y (year)]] }}2014 (14UTCamSat, 25 Oct 2014 08:18:01 +0000)
{{#time: [[Y "(year)"]] }}2014 (year)
{{#time: i's" }}18'01"

The date/time object can be in any format accepted by PHP's strtotime() function. Both absolute (eg 20 December 2000) and relative (eg +20 hours) times are accepted.

{{#time: r|now}}Sat, 25 Oct 2014 08:18:02 +0000
{{#time: r|+2 hours}}Sat, 25 Oct 2014 10:18:02 +0000
{{#time: r|now + 2 hours}}Sat, 25 Oct 2014 10:18:02 +0000

The language code in ISO 639-3 (?) allows the string to be displayed in the chosen language {{#time:d F Y|1988-02-28|nl}}28 februari 1988
{{#time:l|now|uk}}субота
{{#time:d xg Y|20 June 2010|pl}}20 czerwca 2010

If you've calculated a Unix timestamp, you may use it in date calculations by pre-pending an @ symbol.

{{#time: U | now }}1414225082
{{#time: r|@1414225081}}Sat, 25 Oct 2014 08:18:01 +0000

Varning Varning: 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.

{{#time: d F Y | 29 Feb 0100 }}01 mars 0100
(correct, no leap year), but
{{#time: r | 29 Feb 0100 }}Tue, 01 Mar 0100 00:00:00 +0000 (wrong, even if 100 is interpreted as 2000, because that is a leap year)
{{#time: d F Y | 15 April 10000 }}Error: Invalid time.
{{#time: r | 10000-4-15 }}Sat, 15 Apr 2000 10:00:00 +0000

Year numbers 0-99 are interpreted as 2000-2069 and 1970-1999, even when written with leading zeros:
{{#time: d F Y | 1 Jan 0069 }}01 januari 0069

{{#time: d F Y | 1 Jan 0070 }}01 januari 0070

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.

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:

{{#time: Y | January 1 }}2014
Varning Varning: The fill-in feature is not consistent; some parts are filled in using the current values, others are not:

{{#time: Y m d H:i:s | June }}2014 06 25 00:00:00 Gives the start of the day, but the current day of the month and the current year. {{#time: Y m d H:i:s | 2003 }}2003 10 25 00:00:00 Gives the start of the day, but the current day of the year.

With MediaWiki r86805 - Code Review, a four-digit number is always interpreted as a year, never as hours and minutes:
{{#time: Y m d H:i:s | 1959 }}1959 10 25 00:00:00

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):
{{#time: Y m d H:i:s | 195909 }}2014 10 25 19:59:09 Input is treated as a time rather than a year+month code.
{{#time: Y m d H:i:s | 196009 }}Error: Invalid time. Although 19:60:09 is not a valid time, 196009 is not interpreted as September 1960.

The function performs a certain amount of date mathematics:

{{#time: d F Y | January 0 2008 }}31 december 2007
{{#time: d F | January 32 }}Error: Invalid time.
{{#time: d F | February 29 2008 }}29 februari
{{#time: d F | February 29 2007 }}01 mars
{{#time:Y-F|now -1 months}}2014-september

The total length of the format strings of the calls of #time is limited to 6000 characters [1].

Time Zone issue

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:

  • {{#time:g:i A | -4 hours }} ==> 4:18 AM

However, Venezuela is on a -4.5 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:

  • {{#time:g:i A | -4.5 hours }} ==> 5:18 PM

To workaround this issue, simply convert the time into minutes or seconds, like this:

  • {{#time:g:i A | -270 minutes }} ==> 3:48 AM
  • {{#time:g:i A | -16200 seconds }} ==> 3:48 AM

(Tim Starling, the developer of this function, provided the exact syntax for this solution.)


#timel

This function is identical to {{#time: ... }}, except that it uses the local time of the wiki (as set in $wgLocaltimezone) when no date is given.

{{#time: Y-m-d }}2014-10-25
{{#timel: Y-m-d }}2014-10-25
{{#time: Y F d h:i:s}}2014 oktober 25 08:18:01
{{#timel: Y F d h:i:s}}2014 oktober 25 08:18:01

#titleparts

This function separates a pagetitle into segments based on slashes, then returns some of those segments as output.

{{#titleparts: pagename | number of segments to return | first segment to return }}

If the number of segments parameter is not specified, it defaults to "0", which returns all the segments from first segment (included). If the first segment parameter is not specified or is "0", it defaults to "1":

{{#titleparts: Talk:Foo/bar/baz/quok }}Talk:Foo/bar/baz/quok
{{#titleparts: Talk:Foo/bar/baz/quok | 1 }}Talk:Foo
{{#titleparts: Talk:Foo/bar/baz/quok | 2 }}Talk:Foo/bar
{{#titleparts: Talk:Foo/bar/baz/quok | 2 | 2 }}bar/baz
{{#titleparts: Talk:Foo/bar/baz/quok | | 2 }}bar/baz/quok

Negative values are accepted for both values. Negative values for number of segments effectively 'strips' segments from the end of the string. Negative values for first segment translates to "start with this segment counting from the right":

{{#titleparts: Talk:Foo/bar/baz/quok | -1 }}Talk:Foo/bar/baz Strips one segment from the end of the string. See also {{BASEPAGENAME}}.
{{#titleparts: Talk:Foo/bar/baz/quok | -4 }}Strips all 4 segments from the end of the string
{{#titleparts: Talk:Foo/bar/baz/quok | -5 }}Strips 5 segments from the end of the string (more than exist)
{{#titleparts: Talk:Foo/bar/baz/quok | | -1 }}quok Returns last segment. See also {{SUBPAGENAME}}.
{{#titleparts: Talk:Foo/bar/baz/quok | -1 | 2 }}bar/baz Strips one segment from the end of the string, then returns the second segment and beyond
{{#titleparts: Talk:Foo/bar/baz/quok | -1 | -2 }}baz Start copying at the second last element; strip one segment from the end of the string

The string is split a maximum of 25 times; further slashes are ignored and the 25th element will contain the rest of the string. The string is also limited to 255 characters, as it is treated as a page title:

{{#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 }}y/z/aa/bb/cc/dd/ee

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:

{{#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}}z

Varning Varning: You can use #titleparts as a small "string parser & converter", but consider that it returns the first substring capitalized.

{{#titleparts: one/two/three/four|1|1 }}One
{{#titleparts: one/two/three/four|1|2 }}two
  • If lower case is needed, use lc: function to control output.
{{lc: {{#titleparts: one/two/three/four|1|1 }} }}one
  • You can prepend a 'dummy' slash at the beginning of the string to get the correct first substring capitalization (uppercase or lowercase). Use |2 instead of |1 for first segment to return.
{{#titleparts: /one/two/three/four|1|2 }}one
{{#titleparts: /One/two/three/four|1|2 }}One

Varning Varning: Certain characters that are illegal in a page title will cause #titleparts to not parse the string.

{{#titleparts: {one/two} | 1 | 1 }}{one/two}. Does not produce the expected: {one

Varning Varning:

{{#titleparts: [[page]]/123 | 1 | 2 }}page/123

Varning Varning: This function does not degrade gracefully if your input exceeds 255 characters. If the inputted string is 256 characters long or more, this function will simply toss the string back at you.

Strings parsing

An extension is available for additional string parsing, see Extension:StringFunctions.

General points

Substitution

Parser functions can be substituted by prefixing the hash character with subst::

{{subst:#ifexist: Help:Extension:ParserFunctions/sv | [[Help:Extension:ParserFunctions/sv]] | Help:Extension:ParserFunctions/sv }} → the code [[Help:Extension:ParserFunctions/sv]] will be inserted in the wikitext since the page Help:Extension:ParserFunctions/sv exists.
Varning Varning: The results of substituted parser functions are undefined if the expressions contain unsubstituted volatile code such as variables or other parser functions. For consistent results, all the volatile code in the expression to be evaluated must be substituted. See Help:Substitution.

Omdirigeringar

Speciellt {{#time:…|now-…}} skulle vara behändigt i omdirigeringar till sidor som inkluderar datum, men detta fungerar inte.

Especially {{#time:…|now-…}} could be handy in redirects to pages including dates, but this does not work.

Vertikalstreck i tabeller

Parser functions will mangle wikitable syntax and pipe characters (|), treating all the raw pipe characters as parameter dividers. To avoid this, most wikis create the template Template:! with its contents only a raw pipe character (|) or use the {{!}} magic word which is available since MW 1.24. 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. It will then be interpreted as a table row or column separator. Alternatively, raw HTML table syntax can be used, although this is less intuitive and more error-prone.

You can also escape the pipe character for display as a plain, uninterpreted character using an HTML entity: &#124; .

Beskrivning Du skriver Du får
Vertikalstreck som rad-/kolumnavskiljare i tabeller (de flesta wikis)
{{!}}
|
Vertikalstreck som ett vanligt tecken
&#124;
|


Blankstegsignorering

Blanksteg, inklusive radbrytningar, tabbavstånd och mellanrum ignoreras från början och slutet på alla parametrar för dessa tolkningsfunktioner. Om detta inte önskas kan man jämföra strängar genom att omfatta dem med citationstecken.

{{#ifeq: foo           |           foo | equal | not equal }}equal
{{#ifeq: "foo          " | "          foo" | equal | not equal }}not equal

For preventing trimming then- and else-parts, see m:Template:If. Some people use also <nowiki> </nowiki> instead of spaces.

See also