Text and Internationalization support in Opera Presto 2.7

Unicode character set support in Opera

Opera can work with all the characters in the Unicode specification.

• All text communicated to Opera from the network is converted into Unicode.
• In order for Opera to render Unicode characters, the needed glyphs have to be available in the fonts on your system. This might be a problem for older Windows systems. For information on available fonts, see Unicode fonts for Windows computers.
• Updated Unicode character data tables from Unicode v5.0.0 to v5.1.0.
• Uniblocks table now supports ranges outside Unicode plane 0. This is needed to do proper font-switching of characters outside the Unicode plane 0.

Opera implements the following writing system related functionality improvements:

• font-switching: needed in order to display characters that the current font does not include
• line-breaking: needed in order to break scripts written without spaces, such as Chinese, Japanese, and Korean
• CJK: improved line height and underlining in Chinese, Japanese, and Korean
• KDDI emojis: improved support for KDDI emojis and special characters
• Multistyle: improved default fonts for non-western Web pages

Opera relies on the operating system to perform:

• character shaping: contextual glyph selection, ligature forming, character stacking, combining character support, etc.

Opera Presto includes support for Unicode 5.2 character properties (class, casing, bidirectionality, mirroring, normalization) from 5.0.

• See UAX#14 linebreaking
• See UAX#29 text segmentation

Legacy encoding support

Although Opera works with the Unicode character set and its character encodings of UTF-16 and UTF-8, most text on the Internet is encoded in legacy encodings, for instance:

• ISO 8859-1
• Windows-1251
• Shift_JIS (MIME name)
• EUC-KR

Opera handles this by detecting the character encoding used, and converting it to UTF-16. The user has three options for how to handle these pages.

• Auto-detect: in this mode Opera will attempt to detect the encoding used by the page
  • If the transport protocol provides an encoding name, that is used
  • If not, Opera will look at the page for a charset declaration
  • If this is missing, Opera will attempt to auto-detect the encoding, using the domain name to see if the script is a CJK script, and if so which one
  • Opera can also auto-detect UTF-8
• Writing script auto-detect: In this mode the user can tell that this is a Japanese or Chinese page, but that the encoding is unknown. Opera will then analyze the text in the page to determine which encoding is used.
• Encoding override: In this mode the user selects an encoding. This encoding will be used by Opera, regardless of what the page and transport protocol claims is the encoding for the page.

Big5-HKSCS support for the HKSCS-2008 encoding standard has been updated.

• See iana.org: Character sets
• See ogcio.gov: hkscs 2008 summary

Support for bidirectional text

Opera supports bidirectional text as described in Unicode, HTML, and CSS.