HTML

Web documents are written using HTML
The current version is v4.01
XHTML is an XML version of HTML
XHTML applies stricter rules for syntax than HTML
XHTML Transitional

A typpical XHTML document is


<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html 
     PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  <head>
    <title>Virtual Library</title>
  </head>
  <body>
    <p>Moved to <a href="http://example.org/">example.org</a>.</p>
  </body>
</html>

HTML charset

HTML allows use of the Unicode character set, but does not mandate that Unicode must be used - any IETF charset can be used
The charset for a document should be specified by the server (see HTTP) in the Content-Type tag

The document itself can pass this information to the server by a META tag


      <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">

How can the server read the document to find out its charset? It is assumed that the document can be read as though it was an ASCII document (e.g. ASCII, ISO8859 or UTF-8), at least upto this META tag

Character references

Almost any Unicode characters can be used (depending on the charset)
Some characters are reserved e.g. "<" and ">"
For these, an "escape" notation is used
- Special phrases such as "<"
- Numeric escapes such as "&#D", "&#xH"

Text normalization

If you want to compare two documents for e.g. equality, they may need to be normalized
Unicode characters need to be transformed into normal form
In addition, &- escape sequences need to be transformed into "real" characters
Tags in HTML can be in any case, so they need to be normalised to e.g. lower-case
Normalization can be done early - at the server, or late - at the user-agent (browser)
The Character Model for the World Wide Web 1.0: Normalization specifies that early normalization should be applied on the server side for efficiency
The Unicode Normalization "C" is specified, and if the document is HTML or XML, the character escapes should be replaced first

Language tags

The language can specified in tags using ISO 3066 country/language values e.g. to say that the document is in British English
```
      <HTML lang="en-GB">
      
```

Other HTML tags can also use the language specification e.g.


En janvier, toutes les boutiques de Londres affichent des panneaux 
<span lang="en-GB">SALE</span>, mais en fait ces magasins sont bien propres!

A stylesheet using lang can be recognised by some browsers
```
      :lang(en) { color: red; }
      
```
will colour the English text red

http://jan.newmarch.name/i18n/lang.html

Browser language specification

According to the W3C FAQ: Setting language preferences in a browser

Internet Explorer: Tools > Internet Options > General > Languages
Firefox: Tools > Options > Advanced > Languages
Mozilla / Netscape Navigator: Edit > Preferences > Navigator > Languages
Opera: File > Preferences > Languages
IE Mac: Edit > Preferences > Web Browser > Language/Fonts
Safari: (Set via the Mac OS X system preferences) System preferences > International > Language
Galeon: Settings > Preferences > Rendering > Languages
Lynx: O (Options) > Headers Transferred to Remote Servers > Preferred document language
Amaya: Edit > Preferences > Browsing > List of preferred languages

Right-to-left

Look to the right for the next item!
This is text in a <p dir="rtl"> tag
Read the next item from right-to-left
‮This is text surrounded by Unicode RTL chars‬
Read the next item from right-to-left, over on the right
‮This is text surrounded by Unicode RTL chars‬ in a <p dir="rtl"> tag
This is text in a <bdo dir="rtl"> tag

XML

The XML charset is UTF-8 unless something specifies otherwise

CSS2 stylesheets

Use @charset "ISO-8859-1";
The "@charset" must be the first characters in the file.

Quotes

HTML provides the <qo> element to replace quotation marks.

Style-sheets allows to control the type of quote according to the language.


body { display: block;        margin: 15px 15px 15px 15px;}
p { display: block;          margin-bottom: 10px;}
qo { display: inline;}
qo:before { content: open-quote; }
qo:after { content: close-quote; }
/* Locale-specific marks: */
[lang|='en'] > * { /* English */     quotes: "\201C" "\201D"; }
[lang|='fr'] > * { /*Fr(guillemets)*/ quotes:"\AB\A0" "\A0\BB"}
[lang|='fi'] > * { /*Finnish (same dir)*/ quotes:"\201D" "\201D"}
[lang|='de'] > * { /* German */       quotes: "\201E" "\201C" }
[lang|='ja'] > * { /* Japanese */     quotes: "\300C" "\300D" }
[lang|='nl'] > * { /* Dutch */         quotes: "\2018" "\2019" }
[lang|='pl'] > * { /* Polish */        quotes: "\201E" "\201D"

Numbered Lists

The browser supplies the numbers in numbered lists

To allow for i18n, you can specify the list-type in a style sheet


<style>
.list_heb  { list-style-type: hebrew }
.list_geo { list-style-type: georgian }
.list_arm { list-style-type: armenian }
.list_cjk { list-style-type: cjk-ideographic }
</style>

List numbered in Hebrew: <ol class="list_heb"> <li>Item 1</li><li>Item 2</li><li>Item 3</li><li>Item 4</li> </ol>
1. Item 1
2. Item 2
3. Item 3
4. Item 4
List numbered in Georgian: <ol class="list_geo"> <li>Item 1</li><li>Item 2</li><li>Item 3</li><li>Item 4</li> </ol>
1. Item 1
2. Item 2
3. Item 3
4. Item 4
List numbered in Armenian: <ol class="list_arm"> <li>Item 1</li><li>Item 2</li><li>Item 3</li><li>Item 4</li> </ol>
1. Item 1
2. Item 2
3. Item 3
4. Item 4
List numbered in Han character (cjk-ideographic): <ol class="list_cjk"> <li>Item 1</li><li>Item 2</li><li>Item 3</li><li>Item 4</li> </ol>
1. Item 1
2. Item 3
3. Item 4

JavaScript

JavaScript has virtually no support for i18n
It will support Unicode characters in strings, and you can use the "\uXXXX" notation
The Date object has a toLocaleString() method
<script language="JavaScript"> document.write("hello\u2145\u00A5\u04D6<br>") date = new Date() document.write(date.toLocaleString()) </script>
If you want to print e.g. money formats, have localised strings etc, the server has to prepare the strings - JavaScript doesn't have classes to manipulate money, strings, resource bundles, etc
```
decimalSeparator = "."
moneySymbol = "\u20A0" // euro

dollars = 10
cents = 40
moneyStr = "" + dollars + decimalSeparator + cents + moneySymbol
      
```

Jan Newmarch <jan@newmarch.name>

Last modified: Mon Sep 4 15:10:38 EST 2006