This is a very simple extension that defines only two conditions: "unicode {X} is supported" and "unicode {X} is supported exactly". It is employed with code like this: When play begins: if unicode 913 is supported, say "Your interpreter supports Greek!" Because Z-Code interpreters that predate version 1.0 will crash if asked to print or even check the availability of a given Unicode character, this extension will immediately return "not supported" without checking if it detects that the story file is running on an interpreter that predates 1.0. Also, only Glulx is capable of distinguishing between "supported" and "supported exactly."
Version 2 of Unicode Interrogation by Michael Martin begins here. "Allows an author to verify that an interpreter can support specific Unicode effects." "With Glulx support code contributed by Matthew Slattery" [Version 2: Adds Glulx support. Thanks to Matthew Slattery for the code.] Use authorial modesty. Section Z (for Z-Machine only) To decide whether unicode (X - a number) is supported exactly/--: (- (unitest({X}) == 1) -) Include (- [ unitest x i; ! If we don't support at least standard 1.0, @check_unicode will crash the terp! ! Thus, we check version in $32 first. It's a little inefficient to check this in ! every iteration, but sometimes life is hard. if (0->$32 == 0) return 0; @check_unicode x -> i; return i & 1; ]; -) Section G (for Glulx only) [This code comes from a post by Matthew Slattery.] To decide whether unicode (X - a number) is supported exactly: (- (glk_gestalt(gestalt_CharOutput, {X}) == gestalt_CharOutput_ExactPrint) -) To decide whether unicode (X - a number) is supported: (- (glk_gestalt(gestalt_CharOutput, {X}) ~= gestalt_CharOutput_CannotPrint) -) Unicode Interrogation ends here.
Include Unicode Interrogation by Michael Martin.