如何使用正则表达式来匹配不包含特殊字符（＆，\\，<，>，|，）的字符串，除非它们以反斜杠进行

Question

Right now I am using [^ \\\\&<>|\\t\\n]+ which will match any string that contains characters that are not a space, \\, &, <, >, |, \\t, \\n. What I want to do is also allow you to escape any of these special characters so that (for example) \\< or \\& would still allow my entire string to be matched.

Should match:

abcdefghijk abcdef\\&hdehud\\<jdow\\\\

Should not match:

abcdefhfh&kdjeid abcdjedje\\idwjdj

I found this pattern ([^\\[]|(?<=\\\\)\\[)+ which does the same thing for just the "[" character. I couldn't figure out how to extend this to apply to any additional characters.

Any idea how I can make the exception for characters preceded by a backslash?

If it makes any difference, I'm using this in Flex and C++ to tokenize a string for a shell. I believe I need to use negative look-behinds but I don't know how to do that with multiple characters.

Answer 1

You are already most of the way to the answer:

You are using the negated set [^ \\\\&<>|\\t\\n] to specifiy which characters may not be present, so all you have to do is then use the same set without the negation preceded by a \\ to escape the character. That gets you this \\\\[ \\\\&<>|\\t\\n] which can be read as "a \\ followed by any one of the items in the set" now combine the two and you get ([^ \\\\&<>|\\t\\n]|\\\\[ \\\\&<>|\\t\\n])+ .

To break it down:

One or more of: [^ \\\\&<>|\\t\\n] or \\\\[ \\\\&<>|\\t\\n]

Answer 2

As usual, using a regular expression here is overkill. This is a simple text search:

const std::string target = "\\&<>|";
std::string iter = str.find_first_of(target);
while (iter != str.end()) {
    if (*iter != '\\')
        found_bad_character(*iter);
    iter = str.find_first_of(target, std::next(iter));
}