Unicode testing - r3n/rebol-wiki GitHub Wiki

This is a private page for early R3 Unicode testers only. This page will be deleted after completion of early test phase.

First prototype release: 25-Feb-2008

Current release: core004

Current test distribution contains only the kernel ("core") of R3.

It includes: all datatypes, native functions, mezz functions, compression, checksums and hashes, PARSE, help strings, and system object.

Not included: networking, CGI, graphics, view.

Current Zip download size for R3 is: 141 KB.

Notes:

1. Scripts must be UTF-8 encoded.
2. You may need to enable the correct fonts for the CMD console.
3. Some internal debug messages may still be printed. You can ignore them.

• Console - can you type any char into Rebol as a string or as a word?
• Rebol internals - there are more than 100 different parts of Rebol changed to allow Unicode. So, for example, create a string and try inserting Unicode chars.
• OS command line - e.g. can you pass any file name to Rebol?
• OS function API - things like WHAT-DIR and system/options, do those look correct?
• OS file system API - can you open, read, and write a file using unicoded chars?

• Item: UTF-16 formats.
• Plan: Will be supported via refinements.
• Item: IO is binary only. For now, use TO-STRING READ %file.txt
• Plan: Will be supported via refinements.
• Item: Collation/sort and search.
• Plan: We will provide some level of collation and search. But, for full localization, users will need to load special tables.
• Item: Surrogates.
• Plan: no plan at this time.