Regex to match multiline string start with x, ends with y and contains z but not x in the middle

Question

Better explain with an example.

This is text:

<li>hello 
THE WORDS
</li>

<li> cruel </li>

<li> world THE WORDS </li>

I want to find strings start with <li> and ends with </li> and contains THE WORDS. I am expecting to only match with <li> hello THE WORDS </li> and <li> world THE WORDS </li>.

What I tried: (?s)<li>.*?(THE WORDS).*?</li>

With this, second match is <li> cruel </li> <li> world THE WORDS </li>.

I am using Sublime Text.

Answer 1

EDIT: For the new requirements, use this regex:

(?s)<li>(?:(?!</li>).)*?THE WORDS.*?</li>

Explanation

• (?s) activates DOTALL mode, allowing the dot to match across lines
• <li> matches literal chars
• (?:(?!</li>).) asserts that what follows is not </li and matches one character
• The *? quantifier makes us lazily match up to....
• THE WORDS
• • The *? quantifier makes us lazily match up to....
• literal </li>

Original Answer (different requirements):

Use this simple regex (tested in Sublime):

x+[^xy]*z[^xy]*y+

See matches in the regex demo.

Explanation

• x+ matches one or more x chars
• [^xy]* matches any chars that are neither an x nor a y
• z matches the z we want (ensuring there is at least one)
• [^xy]* matches any chars that are neither an x nor a y
• y+ matches one or more y chars

Answer 2

The below command would match only the xx hello zz yy, xx world zz yy strings and capture only zz from both the strings.

<li>(?:(?!<\/li>).)*(THE WORDS).*?<\/li>

DEMO