Tcl/Tk 8.1b1 Release Announcement December 11, 1998 We are pleased to announce the 8.1b1 releases of the Tcl scripting language and the Tk toolkit. This release has a number of important improvements over the previous release, but this is still beta quality. Where to get the new releases: ------------------------------ Tcl 8.1b1 and Tk 8.1b1 are available by public FTP from ftp.scriptics.com in the directories /pub/tcl/tcl8_1 and /pub/tcl/mac. Within a few days they will also appear at numerous mirror sites around the Internet. The easiest way to retrieve these releases is via the Scriptics Web page for the 8.1 releases: http://www.scriptics.com/software/8.1.html (The Macintosh releases are not yet ready, but we expect these to be built relatively soon.) For additional information: --------------------------- For complete information about all the Tcl/Tk resources available on the Internet, visit the Tcl Resource Center at Scriptics: http://www.scriptics.com/resource/ Summary of changes: ------------------- Below is a summary of the most important changes in the 8.1b1 releases. For a complete list of all changes, see the "changes" files in the Tcl and Tk source distributions. 1. Added a message catalog facility to Tcl. This adds several commands in the "msgcat" package. Thanks to Mark Harrison for contributing the initial implementation. 2. Added the "encoding" command that facilitate translations of strings between different character encodings. 3. Added "string totitle" command to convert strings to capitalize the first character of a string and lowercase all of the other characters. 4. "regexp" now accepts three new switches ("-line", "-lineanchor", and "-linestop") that control how regular expressions treat line breaks. See the regexp manual entry for more details. 5. "scan" now supports XPG3 position specifiers and the "%n" conversion character. See the "scan" manual entry for more details. 6. Changed the Tcl_EvalObjv interface to remove the string and length arguments. 7. Various bug fixes, including: - Various regular expression performance bug fixes supplied by Henry Spencer - Fixed various thread related race conditions. - scan and format correctly handle Unicode characters - regexp and string perly handle case folding according to the Unicode character tables. - The regular expression character classification syntax now includes Unicode characters in the supported classes. - The Tcl memory allocator now returns 8-byte aligned chunks of memory which improves performance on Windows and avoids crashes on other platforms. - Fixed a number of memory overflow and leak bugs. - Fixed various bugs related to line feed translation. - Tk_UnderlineChars did not handle UTF strings properly so underline indices were in bytes instead of characters. - Fixed menus and titles so they properly display Unicode characters under Windows. - Fixed several small bugs in the menu display code. - The Windows clipboard was not correctly traslating multibyte characters.