I am working on an NLP project that requires me to remove computer code from a piece of text. The code is encased between the tags <pre><code> and </code></pre>. Now I could do a simple regex match, but I want to generalize this function so that it can remove text between any two specified strings, even when they are nested.
For example, if I have a string:

string = "<pre><code> Should be deleted <pre><code> also deleted </code></pre> delete this too </code></pre> don't delete <pre><code> delete </code></pre> "

then I am expecting the output to be the string " don't delete ".

If I do re.sub('(<pre><code>)(.*)(</code></pre>)', '', string), it gives me an empty string.

I know how to do this for single characters, like removing string between curly braces. For example, if my string was string = "{ Should be deleted { also deleted } delete this too } don't delete { delete } ", then doing this gives me the desired output:

a = '{'
b = '}'
regexp = a +  '[^' +a + '^' + b + ']*' + b

while a in string or b in string:
  string = re.sub(regexp, '', string)

regexp evaluates to '{[^{^}]*}' here. The while loop is necessary because solutions like re.sub(r'{[^{^}]*}', '', string) doesn’t work for nested cases.

I tried to apply the same logic as the single characters case by doing re.sub('(<pre><code>)[^(</code></pre>)^(<pre><code>)](</code></pre>)', '', string) but it produces the output: "<pre><code> Should be deleted <pre><code> also deleted </code></pre> delete this too </code></pre> don't delete <pre><code> delete </code></pre> " which means nothing is matched.