Java replaceAll失败，源字符串中有美元符号

Question

Say I have the following code

String test = "$abc<>";
test = test.replaceAll("[^A-Za-z0-9./,#-' ]", "");

test is now "$abc".

Why does it keep the dollar sign?

Answer 1

Your list of characters to preserve includes #-' , which is a range from Unicode U+0023 (the # symbol) to U+0027 (the ' symbol), including $ (U+0024).

If you meant #-' to be interpreted as a list of three characters, just escape it:

test = test.replaceAll("[^A-Za-z0-9./,#\\-' ]", "");

or put it at the end of the list:

test = test.replaceAll("[^A-Za-z0-9./,#' -]", "");

Answer 2

Because you must put the - as the last character in your character class.

Try

test.replaceAll("[^A-Za-z0-9./,#' -]", "");

It'll work :)

See also In a java regex, how can I get a character class eg [az] to match a - minus sign?

and the Javadoc for Pattern (Ctrl-F "Character classes")

Note that a different set of metacharacters are in effect inside a character class than outside a character class. For instance, the regular expression . loses its special meaning inside a character class, while the expression - becomes a range forming metacharacter.