> ... should all characters in the class [:space:] be > treated as a token seperator in shells/languages, or > just the ASCII SPACE? If it seems pertinent to you, the C language standard sets this precedent [1]: "The source file is decomposed into preprocessing tokens and sequences of white-space characters (including comments)." Although the verbiage in the standard annoyingly leaps a few extra gymnastics to support some obsolete anti-ascii character sets, basically, all six of the traditional [:space:] characters [\t\n\v\f\r ] are treated the same---except that the two-character sequence "\\\n" (backslash followed by newline) serves to join two lines of text before preprocessing. However, the NO-BREAK SPACE is not one of the six characters, nor is it a member of [:space:], nor is it acknowledged by the iswspace(3) function. Personally, if I were using your interpreter, I would not want it quietly to accept a NO-BREAK SPACE as a token separator. I would prefer it to warn me that I had some weird character lurking in my script. -- Thaddeus H. Black 508 Nellie's Cave Road Blacksburg, Virginia 24060, USA +1 540 961 0920, t@b-tk.org [1] ISO/IEC 9899:1999 5.1.1.2.3.
Attachment:
pgprBpaF_L4ii.pgp
Description: PGP signature