Template:Lang/doc

The purpose of this template is to indicate, via a code, that a span of text belongs to a particular language. It often makes no visible changes to the text, but can prompt web browsers to use a more appropriate font, or screen readers to use a particular kind of pronunciation; and so on. The, below, provides more information.

Syntax and usage


The ISO 639 code is usually a two- or three-letter abbreviation, in lowercase, of the language's name. French, for example, has the code "fr":

produces:
 * She said: "Je suis française."

By default, this template will place articles into the relevant subcategory of. To suppress this – e.g. when using {{lang}} within a wikilink or the title parameter of a citation – add the parameter true. If {{lang}} is used to identify a language read from right to left, e.g. Arabic, the template rtl-lang may be used afterward to ensure the standard (English) left-to-right formatting is resumed. It also supports the nocat parameter.

It is better to incorporate only the foreign-language text (linked if appropriate) inside the template, so that extraneous markup that is not from the non-English language in question does not receive incorrect metadata for that language. For example: This includes italicization as foreign; English-language quotation marks around titles of work in language that use other quotation character glyphs; italicization of titles in languages that do not use that convention; and emphasis that is not found (in one style or another) in the original non-English text; among other cases. If in doubt, put such markup outside the.
 * Use
 * Not

There are also versions of this template for specific languages, intended for use the first time one of those languages appears in an article, that print the language's name and, for languages using the Latin alphabet, italicize the content automatically. For example, and  yield, respectively, "español" and "русский язык" (without the quotation marks).

Language subtags can be used to indicate a regional variant of and/or a particular script used for a language. According to the World Wide Web Consortium, "the golden rule when creating language tags is to keep the tag as short as possible", so such subtags should only be added if there is good reason to use them. ISO 639‑1 is preferred over ISO 639‑2 and ISO 639‑3.

Italics
By default, outputs plain text, while  templates (e.g. lang-de) output text in italics. If, with the latter, plain text is required (e.g. for proper names), noitalic may be used. For example,

produces:
 * ... the border town of Ventimiglia (Vintimille)

Indicating writing script
If necessary, add the ISO 15924 code to indicate the script.

For example, Russian is usually written in the Cyrillic script, therefore the  script code is superfluous and the language code will be   instead of. However, when that text is transliterated the  code (transliteration) should be used because it isn't the default script for Russian:. Example:

which is the same as
 * Moscow (Russian: Москва́, Moskva)
 * Moscow (Москва́, Moskva)

Results in your browser:
 * Moscow (Москва́, Moskva)

Instead of using  the   can be used, but if you specified a font and a size for the target language in your custom CSS (see Help:User style), the transliteration will appear in that specified style. To specify a transliteration scheme, for example, you are using the ISO 9 transliteration of Cyrillic, use :
 * Moscow (Москва́; ISO 9: Moskva)

The IANA maintains a list specifying when the script tag should be suppressed. In some cases, the script must be always specified, like Tajik which can be equally written in Arabic, Latin or Cyrillic alphabets: Which results in your browser: Note the use of rtl-lang instead of lang when using the Arabic script (see ).
 * Tajik (تاجیکی, tojikī, тоҷикӣ)
 * Tajik (تاجیکی, tojikī, тоҷикӣ)

Undetermined language
The template is not only used to specify the language of foreign words, but also to specify a single symbol or character in a specific script, for example when discussing about a specific writing system. Many times the character / symbol is used in several languages, but when the article refers to the grapheme itself the ISO 639‑2 and ISO 639‑3 language code  for Undetermined language should be used. For example:

The 字 Han character has 6 strokes.
 * The 字 Han character has 6 strokes.

Han characters are used in Chinese, Japanese, sometimes Korean, and formerly Vietnamese, and in this case the character is not used for any specific language. Note that the script code used is, which specifies generic Han characters (Hanzi, Kanji, Hanja).

Compare script The Han character has 6 strokes.
 * The Han character has 6 strokes.

Indicating regional variant
In some cases, maybe it will be needed to add ISO 3166-1 alpha-2 country codes (specific usage of that country). Of course the three codes can appear in the same tag, for example the code  will be used for Chinese text written with Traditional Han characters, containing words or expressions specific to Taiwan. is for simplified. Examples:


 * 臺灣

Results in your browser:
 * 臺灣

Writing direction
rtl-lang is used to wrap strings of right-to-left text (from languages like Arabic or Hebrew) embedded in left-to-right text, and resets text direction to left-to-right at its conclusion. For whole paragraphs of right-to-left text, use rtl-para. The  prefix is not needed in the  form, as this is implied by the language: there is no, for instance: just use.

Rationale

 * Web browsers can use the information in such multilingual support templates to choose an appropriate font.
 * This is great for CJK where a character can be given its language-specific shape but will fall back to another form if no appropriate font is found or if the preferred font lacks that character, for example because the language does not make use of that character: see and these comparison table and screen photo.
 * For accessibility – screen readers need language info to speak text in the correct language – and to satisfy Wikipedia accessibility guidelines.
 * For spell checkers and grammar checkers.
 * To help browsers choose appropriate quotation marks, and make decisions about hyphenation, ligatures, and spacing.
 * Users can apply styles to languages in their style sheets (useful for editors).
 * Search engines can use this information when indexing text.
 * Facilitates better data-scraping, parsing and re-use.
 * Useful for application developers who re-publish Wikipedia (also see ).
 * Useful for research or compiling statistics about language use in Wikipedia.

Applying styles
You can apply CSS styles in your user style sheet. Registered users can put styles into User:XXX/monobook.css, where XXX is the user name.

These examples may work in Internet Explorer 8; prior versions do not support attribute selectors. Alternatively, try another browser such as Firefox.

Example: to apply a font to Russian-language text:

span[lang|=ru] { font-family: fonteskaya; }

Example: to apply a colour to text marked with any language:

span[lang] { color: green; }

Do not use quotation marks in your user style sheet; wikitext would screw them up. They are recommended in CSS, but not required other than for font families containing generic-family keywords ('inherit', 'serif', 'sans-serif', 'monospace', 'fantasy', and 'cursive'). See here for details.

TemplateData
{ "description": "Indicate that a given span of text belongs to a particular language. Use by browsers which use different styles for different languages.", "params": { "1": {   "label": "Language code", "description": "A language tag or an ISO 639 language code, e.g. fr (the code for French).", "type": "string", "required": true }, "2": {    "label": "Text", "description": "The text e.g. Je suis française.", "type": "string", "required": true } }}

Further information

 * W3C
 * Language tags in HTML and XML – Overview
 * Internationalization Best Practices: Specifying Language in XHTML & HTML Content – W3C Working Draft 21 July 2006
 * Understanding the New Language Tags
 * FAQ: Styling using the lang attribute
 * IANA
 * IANA Language Subtag Registry
 * Tags for Identifying Languages (RFC 4646)
 * Matching of Language Tags (RFC 4647)
 * Language tags (obsolete per RFC 4646)
 * Mozilla Firefox
 * Bug with script selection