跳过某些字符之间的正则表达式匹配

Question

In the following program I'm trying to prefix any text starting with "_" with the word "app":

 String s = "C_NAME=\"usco _f2 re_col\" &&_f1=\"ot\" && _f1 = \"fd\"|stats count(_f1)|fields _f1, f3|filter match(app_f1,\"get\")|extract field=\".*?\\[\\d+\\s+(?<_LogLevel>\\w+).*";

        StringBuilder s1 = new StringBuilder(s);
        java.util.regex.Pattern p = java.util.regex.Pattern.compile(".*?((?<!\\w)_\\w+).*?",
                Pattern.UNICODE_CHARACTER_CLASS);
        Matcher m = p.matcher(s);
        while (m.find()) {
            String fieldname = m.group(1);
            s1.replace(m.start(1), m.end(1), "app" + fieldname);
                m.reset(s1);

        }
        System.out.println(s1.toString());

But, I've one more clause to take care: the above regex shouldn't append the text "app" to the text starting with "_" if the text is located in between a pair of double quotes. In this particular example the text "_f2" shouldn't get prefixed. Any clues on how to achieve this functionality?

Update:

---

The following regex workd fine in my case:

Pattern p = Pattern.compile(".*?(?:\"[^\"]*\"|((?<!\\w)_\\w+)).*?",
                Pattern.UNICODE_CHARACTER_CLASS);

Answer 1

Since (*SKIP)(*FAIL) is not implemented in Java , you'll need to come up with some logic on your own:

What_I_want_to_avoid|(What_I_want_to_match)

So, in your case:

"[^"]*"|(\b_\w+)
# left: anything between two double quotes
# right: a word boundary, an underscore and at least one word character

Match everything unwanted in the left branch and if group 1 exists, do a replacement here.
See a demo on regex101.com .