> I tried [[:space:]] and that did not work but [:space:]? did. > > What is the difference? I'm unclear on that even after reading this > thread.
[[:space:]] is what you should use if you use the usual regexp-aware tools like awk, perl, grep, lex and so on. For those tools the outer []-s mean "any of the characters inside" and the [:space:] is just a shorthand to all characters that look like space. The [:space:] is only meaningful inside an other pair of []-s. If you use it without the second enclosing[]-s, it the regexp evaluator takes it at face value, that is "any one of the ':', 's', 'p', 'a', 'c', 'e', ':' characters". That is, "abcde" matches [:space:] but does not match [[:space:]], but " " matches [[:space:]] but won't match [:space:]. OOo/SO treat regexp a different way. For OO [:space:] on its own means what [[:space:]] is for the rest: any one of the space-like characters. Except that OOs implementation, apart from being incompatible with the rest, also has a minor bug. It does not recognise [:space:] on its own (standing for one character), only as a part of a higher level regexp construct. In [:space:]+ the + says "one or more of whatever is before me" and as the 'before me' part OO processes [:space:] fine. Same with ([:space:]) or an expression involvoing [:space:]. I do believe it is indeed a genuine bug. Zoltan --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
