Skip to content

Latest commit

 

History

History
125 lines (94 loc) · 9.37 KB

README.md

File metadata and controls

125 lines (94 loc) · 9.37 KB

intl-segmenter elements

A custom element for Intl.Segmenter

intl-segmenter elements can be used to separate a string with language-sensitive sematic segmentation.

A particularly useful case is a page title used in a hero area of a webpage. Usually this kind of title is used in a graphical way, and often in languages like Chinese and Japanese, because they are written in characters without spaces to separate words, they could result in awkward line breaks. For example, the Chinese sentence, “欢迎来到我的网站！” (Welcome to my website), has the following semantic segments:

  • “欢迎” (“Welcome”)
  • “来到” (“to”)
  • “我的” (“my”)
  • “网站” (“website”)
  • “！” (exclamation mark)

But it could be broken into:

欢迎来到我
的网站！

(The word “我的” is broken into two lines)

The ideal result should be:

欢迎来到
我的网站！

The <intl-segmenter-segment> element takes its own textContent, segments it, and wraps each segment in a <span part="segment"> element in its shadow DOM. For the above case, the <intl-segmenter> element should have its option-granularity attribute set to word, and the <span> elements in the Shadow DOM will have to wordlike as their parts, and non-wordlike segments are not wrapped in a <span>, this makes sure that punctuation segments will stay in the appropriate line between line breaks (e.g. a comma won’t be at the beginning of a line and an opening quotation mark won’t be at the beginning of a line).

Example

Segmenting a string into words

<style>.title-content::part(wordlike) { white-space: nowrap; }</style>
<intl-segmenter locales="zh-Hans" option-granularity="word">
  <h1>
    <intl-segmenter-segment class="title-content">
      欢迎来到我的网站！
    </intl-segmenter-segment>
  </h1>
</intl-segmenter>

<intl-segmenter> (HTMLIntlSegmenterElement)

Learn more about Intl.Segmenter()

Attributes

Name Type Default Required? Description Property
locales string | undefined undefined locales
locales-from string | undefined undefined localesFrom
option-granularity Intl.SegmenterOptions['granularity'] | undefined undefined optionGranularity
option-localematcher Intl.RelativeTimeFormatLocaleMatcher | undefined undefined optionLocaleMatcher

Properties

Name Type Default Required? Read only? Description Attribute
intlObject Intl.Segmenter undefined Yes
localeList DOMTokenList undefined Yes
locales string | undefined undefined locales
localesFrom string | undefined undefined locales-from
localesFromElements HTMLIntlLocaleElement[] undefined Yes
optionGranularity Intl.SegmenterOptions['granularity'] | undefined undefined option-granularity
optionLocaleMatcher Intl.RelativeTimeFormatLocaleMatcher | undefined undefined option-localematcher

Methods

Name Return Description
resolvedOptions() Intl.ResolvedSegmenterOptions Learn more about Intl.Segmenter.prototype.resolvedOptions()

<intl-segmenter-segment> (HTMLIntlSegmenterSegmentElement)

Learn more about Intl.Segmenter.prototype.segment

Attributes

Name Type Default Required? Description Property
provider string | undefined undefined provider

Properties

Name Type Default Required? Read only? Description Attribute
input string undefined Yes A read only reference to the input argument of Intl.Segmenter’s segment().
provider string | undefined undefined provider
providerElement HTMLIntlSegmenterElement | undefined undefined Yes
value Intl.Segments | undefined undefined Yes

Slots

Name Description
The slot should only contain plain text, its textContent is used as the input argument of Intl.Segmenter’s segment(). Any HTML elements in the slot are ignored.

CSS Parts

Name Description
segment The <span> element that contains the text of a segment. Not that if the provider element’s option-granularity is word, for a non-word-like segment, the text is not wrapped in a <span> element. This is to avoid characters like punctuation marks being inappropriately wrapped to the next line, e.g. a comma at the beginning of a line, or a opening quotation mark at the end of a line.
value The <span> element that contains the segmented HTML result.
wordlike The <span> element that contains the text of a segment which isWordLike is true.