Question

This question is a derivation from another question in this forum which I thought was quite simple, but in the end found out a difficult one.

OP asked to replace any text xyz from the all substrings <tagname>xyz</tagname> with some NEW TEXT. I did the following:

String str="<tagname>bgerh</tagname>sdfghuhjg<tagname>bgew</tagname>rwesdgrhtf<tagname>bfgh</tagname>";

System.out.println(str.replaceAll("(?<=(<tagname>)).*(?=(</tagname>))","NEW TEXT"));

The output I got:

<tagname>NEW TEXT</tagname>

which is obviously not the desired one.

So, I would like to know if the regex checking happen from both the ends of a string, and I will bbe delighted with a solution to the example.
Thanks in advance.

Answer 1

.* is a greedy quantifier, meaning it will match everything it can. Because your sample string begins with <tagname> and ends with </tagname>, everything in between will be matched by .*.

To fix this, you can instead use a reluctant quantifier, and it will only grab as little as it can. The reluctant qualifier looks like this: .*?, and thus the entire expression would look like this:

"(?<=<tagname>).*?(?=</tagname>)"