Help:Extension:ParserFunctions

From MediaWiki.org
Jump to: navigation, search
Other languages:
български 1% • ‎বাংলা 0% • ‎Deutsch 14% • ‎Zazaki 0% • ‎English 100% • ‎British English 98% • ‎español 5% • ‎français 8% • ‎日本語 52% • ‎Lëtzebuergesch 7% • ‎मराठी 1% • ‎polski 14% • ‎پښتو 8% • ‎português 5% • ‎português do Brasil 23% • ‎русский 7% • ‎svenska 37% • ‎ไทย 0% • ‎Türkçe 8%

The ParserFunctions extension provides eleven additional parser functions to supplement the "magic words", which are already present in MediaWiki. All the parser functions provided by this extension take the form:

{{#関数名: 引数 1 | 引数 2 | 引数 3 ... }}

#expr

種類 演算子
グループ化 (括弧) ( )
数値 1234.5   e (2.718)   pi (3.142)
二項演算子 e   単項 +,-
単項 not ceil trunc floor abs exp ln sin cos tan acos asin atan
二項 ^
* / div mod
+ -
丸め round
論理 = != <> > < >= <=
and
or

この関数は、数式を計算してその計算結果を返します。

{{#expr: 数式 }}

The available operators are listed to the right, in order of precedence. See Help:Calculation for more details of the function of each operator. 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.

When evaluating using boolean algebra, zero evaluates to false and any nonzero value, positive or negative, evaluates to true:

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

空の入力式は空の文字列を返します。無効な式はいくつかのエラー メッセージのうちのいずれかを返します。このエラーは #iferror 関数を使用して捕捉できます:

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

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:

{{#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 1,484,739, where we want 1484739 which can be obtained by {{formatnum:{{NUMBEROFUSERS}}|R}}. 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}}1484839


警告 警告: 演算子 mod は、第 2 引数の値によっては誤った結果を返す場合があります:
{{#expr: 123 mod (2^64-1)}}Division by zero. (空の文字列を返しますが、123 を返すべきです)

丸め

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.

切り上げや切り捨てを行なうには、それぞれ単項の ceilfloor を使用してください。

結果 丸めの手法
{{#expr: 1/3 round 5 }} 0.33333 最終桁が 5 未満であるため、切り上げはされません
{{#expr: 1/6 round 5 }} 0.16667 最終桁が 5 以上であるため、切り上げされます
{{#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 round の右の値の小数点以下は、丸めの結果には影響しません
{{#expr: trunc 1234.5678 }} 1234 小数点以下の切り捨て
最も近い整数への丸め
{{#expr: 1/3 round 0 }} 0 最も近い整数への切り捨てで 0
{{#expr: 1/2 round 0 }} 1 最も近い整数への切り上げで 1
{{#expr: 3/4 round 0 }} 1 最も近い整数への切り上げで 1
{{#expr: -1/3 round 0 }} -0 最も近い整数への切り上げで 0
警告 警告: 0 にマイナスを付けると、比較で一致しなくなります
{{#expr: -1/2 round 0 }} -1 最も近い整数への切り捨てで -1
{{#expr: -3/4 round 0 }} -1 最も近い整数への切り捨てで -1
ceillfloor による切り上げや切り捨て
{{#expr: ceil(1/3) }} 1 Up to the next larger integer, which is one
{{#expr: floor(1/3) }} 0 Down to the next smaller integer, which is zero
{{#expr: ceil(-1/3) }} -0 Up to the next larger integer, which is zero
警告 警告: 0 にマイナスを付けると、比較で一致しなくなります
{{#expr: floor(-1/3) }} -1 Down to the next smaller integer, which is negative one
{{#expr: ceil 1/3 }} 0.33333333333333 1 が既に整数であるため、丸めは行なわれない
警告 警告: あなたの予想とは異なり、ceil(1/3) ではなく (ceil 1)/3 と解釈される

#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: テスト文字列 | テスト文字列が空ではない場合の値 | テスト文字列が空 (または空白のみ) の場合の値 }}
{{#if: パラメーター 1 | パラメーター 2 | パラメーター 3 }}

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) }} }}

このパーサー関数の他の例については Help:Parser functions in templates/ja を参照してください。

#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
警告 警告: 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.

警告 警告: Content inside parser tags (such as <nowiki>) is 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.

#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.

comparing

{{#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 }}
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.

#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/ja | exists | doesn't exist }}exists
{{#ifexist: XXHelp:Extension:ParserFunctions/jaXX | 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 (because MediaWiki:Copyright has been customized)

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/ja), Special:WhatLinksHere/Foo will list Help:Extension:ParserFunctions/ja.

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: does not work with interwiki links.

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

この関数は相対ファイル パスを絶対ファイル パスに変換します。

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

path 部分では、以下の構文を使用できます:

  • . → 現在の階層
  • .. → 「1 つ上の階層に移動」
  • /foo → 「下位ディレクトリ /foo に 1 階層移動」

base path を指定していない場合は、ページの完全なページ名が代わりに使用されます:

{{#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
}}

例:

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

#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.

Default

The default result is returned if no case string matches the comparison string:

{{#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
警告 警告: 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: =.

例:

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

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.

例えば:

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

は、以下と同等です

  • {{#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

コード 説明 現在の出力
(更新するにはこのページのキャッシュを破棄してください)
Y 4 桁の年 2014
y 2 桁の年 14
L 閏年の場合 1、平年の場合 0 0
o ¹ 指定した週の ISO-8601 年 ² 2014 ³

¹ PHP 5.1.0 以降および rev:45208 が必要
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.

n 月番号、ゼロ補完なし 4
m 月番号、ゼロ補完あり 04
M サイトの言語での月名の省略形 4月
F サイトの言語での完全な月名 4月
xg Output the full month name in the genitive form for site languages that distinguish between genitive and nominative forms. ポーランド語の例:

(主格)

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

(所有格)

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

W ISO 8601 週番号、ゼロ補完あり 17
日と曜日
j 日、ゼロ補完なし 25
d 日、ゼロ補完あり 25
z Day of the year (January 1 = 0).
Note Note: To get the ISO day of the year add 1.
114
D An abbreviation for the day of the week. Rarely internationalized.
l The full weekday name. Rarely internationalized. 金曜日
N ISO 8601 曜日番号 (月曜日 = 1、日曜日 = 7) 5
w 曜日番号 (日曜日 = 0、土曜日 = 6) 5
a 午前 (00:00:00 ~ 11:59:59) の場合「am」、午後 (12:00:00 ~ 23:59:59) の場合「pm」 am
A 上の a の大文字版 AM
g 時、12 時間制、ゼロ補完なし 6
h 時、12 時間制、ゼロ補完あり 06
G 時、24 時間制、ゼロ補完なし 6
H 時、24 時間制、ゼロ補完あり 06
分と秒
i 分、ゼロ補完あり 41
s 秒、ゼロ補完あり 41
U 1970年1月1日 00:00:00 GMT からの経過秒数 1398408101
タイムゾーン (1.22wmf2 時点)
e タイムゾーン識別子 UTC
I 日時がサマータイムかどうか 0
O グリニッジ標準時 (GMT) との差 +0000
P グリニッジ標準時 (GMT) との差、コロン付き +00:00
T タイムゾーンの省略形 UTC
Z タイムゾーンのオフセット (秒) 0
その他
t 30
c ISO 8601 形式の日付、Y-m-dTH:i:s+00:00 と同等 2014-04-25T06:41:41+00:00
r RFC 5322 形式の日時、D, j M Y H:i:s +0000 と同等。曜日と月名は国際化されない。 Fri, 25 Apr 2014 06:41:41 +0000
グレゴリオ暦以外
イスラム暦
xmj 24
xmF 完全な月名 ジュマーダーッ=サーニー
xmn 月番号 6
xmY 完全な年 1435
イラン暦 (Jalaly)
xij 5
xiF 完全な月名 オルディーベヘシュト
xin 月番号 2
xiY 完全な年 1393
xiy 2 桁の年 93
ユダヤ暦
xjj 25
xjF 完全な月名 ニサン
xjt 月の日数 30
xjx 月名の所有格 ニサン
xjn 月番号 7
xjY 完全な年 5774
タイ太陽暦
xkY 完全な年 2557
民国紀元/主体暦
xoY 完全な年 103
日本の年号
xtY 完全な年 平成26
フラグ
xn 後続の数値をそのまま半角英数字に整形する。 ヒンディー語では、{{#time:H, xnH}} は ०६, 06 を返す。
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 後続の数値をローマ数字に整形する。10,000 までの数値で動作する。
(MediaWiki 1.20 より前では 3,000 まで)
{{#time:xrY}} → MMXIV
xh 後続の数値をヘブライ数字に整形する。 {{#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 }}2014-04-25
{{#time: [[Y]] m d }}2014 04 25
{{#time: [[Y (year)]] }}2014 (14UTCamFri, 25 Apr 2014 06:41:41 +0000)
{{#time: [[Y "(year)"]] }}2014 (year)
{{#time: i's" }}41'41"

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}}Fri, 25 Apr 2014 06:41:41 +0000
{{#time: r|+2 hours}}Fri, 25 Apr 2014 08:41:41 +0000
{{#time: r|now + 2 hours}}Fri, 25 Apr 2014 08:41:41 +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}}п'ятниця
{{#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 }}1398408101
{{#time: r|@1398408101}}Fri, 25 Apr 2014 06:41:41 +0000

警告 警告: 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 3月 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 1月 0069

{{#time: d F Y | 1 Jan 0070 }}01 1月 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
警告 警告: 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 04 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 04 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 04 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 12月 2007
{{#time: d F | January 32 }}Error: Invalid time.
{{#time: d F | February 29 2008 }}29 2月
{{#time: d F | February 29 2007 }}01 3月
{{#time:Y-F|now -1 months}}2014-3月

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 }} ==> 2:41 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 }} ==> 3:41 PM

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

  • {{#time:g:i A | -270 minutes }} ==> 2:11 AM
  • {{#time:g:i A | -16200 seconds }} ==> 2:11 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 }}2014-04-25
{{#timel: Y-m-d }}2014-04-25
{{#time: Y F d h:i:s}}2014 4月 25 06:41:41
{{#timel: Y F d h:i:s}}2014 4月 25 06:41:41

#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

警告 警告: 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

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

{{#titleparts: {one/two} | 1 | 1 }}{one/two}. 以下の値にはなりません: {one

警告 警告:

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

警告 警告: 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.

General points

Substitution

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

{{subst:#ifexist: Help:Extension:ParserFunctions/ja | [[Help:Extension:ParserFunctions/ja]] | Help:Extension:ParserFunctions/ja }} → the code [[Help:Extension:ParserFunctions/ja]] will be inserted in the wikitext since the page Help:Extension:ParserFunctions/ja exists.
警告 警告: 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.

Redirects

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

表内でのパイプ記号のエスケープ

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 (|). 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.

パイプ記号は、プレーンな (解釈されない) 文字としてエスケープすることもできます。これには HTML エンティティ &#124; を使用します。

説明 入力内容 結果
パイプ記号を表の行/列区切りとしてエスケープ (大部分のウィキ)
{{!}}
|
パイプ記号をプレーンな文字としてエスケープ
&#124;
|


空白類の除去

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.

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

then 部分および else 部分での空白の除去を回避する方法は、m:Template:If を参照してください。

See also

言語: English  • български • বাংলা • Deutsch • Zazaki • español • français • 日本語 • मराठी • polski • português • português do Brasil • русский • svenska • ไทย • Türkçe
This page is a translated version of a page Help:Extension:ParserFunctions and the translation is 52% complete.