A lightweight UTF8-aware String class meant for use with Ruby 1.8.x

The scanning code is implemented in C (would love a Java version as well, if any of you want to help with that).

At the moment, this gem is tested on 1.8.7, 1.9.2 and Rubinius 1.2.1dev - it may work on others but ymmv.

String::UTF8 Example

The String::UTF8#[] API isn’t fully supported yet, and may never support regular expressions. I’ll update this readme as I make progress.

# String (on 1.8)
str = "你好世界"
str.length # 12
str[0] # 228
str.chars.to_a.inspect # ["\344", "\275", "\240", "\345", "\245", "\275", "\344", "\270", "\226", "\347", "\225", "\214"]

# String::UTF8 (on 1.8)
utf8_str = "你好世界".as_utf8
utf8_str.length # 4
utf8.chars.to_a.inspect # => ["你", "好", "世", "界"]

utf8[0]        # => "你"
utf8_str[0, 3] # => "你好世"
utf8_str[0..3] # => "你好世界"

StringScanner::UTF8 Example

The full StringScanner API hasn’t been made UTF8-aware yet. I may try and finish it later, but the getch method is all I need at the moment.

# StringScanner (on 1.8)
scanner = StringScanner.new("你好世界")
scanner.getch # => "\344"

# StringScanner::UTF8
require 'utf8/string_scanner'

# NOTE: we could have used scanner.as_utf8 instead
utf8_scanner = StringScanner::UTF8.new("你好世界")
utf8_scanner.getch # => "你"
utf8_scanner.getch # => "好"
utf8_scanner.getch # => "世"
utf8_scanner.reset
utf8_scanner.getch # => "你"

Disclaimer

I’m only planning on making the methods I need UTF8-aware, if you need others I welcome pull requests :)