Regex positive lookahead is too greedy

Asked Aug 07 '17 at 11:21

Active Aug 07 '17 at 11:23

Viewed 35 times

I am attempting to remove the <span class="notranslate"></span> of a string using a regex. Basically it's because Google Translate API needs that kind of wrapping if some string should not be translated, which is what I need.

I have that regex to do this (it needs to handle new lines, that's why [\S\s] is used instead of .) :

<span class="notranslate">[\S\s]+(?=<\/span>)<\/span>

However when testing it, it is way too greedy because for that input for example:

<span class="notranslate">%1</span> per il seguente motivo: <span class="notranslate">\n</span> <span class="notranslate">%2!d!</span>

That regex gets the whole string. I was a little confused because many answers (including "How to match "anything up until this sequence of characters" in a regular expression?") seems to indicate that ? as a non greedy thing.

I need to have it find every span wrap with the notranslate class, not all of them if there are many.

edited Aug 07 '17 at 11:23

asked Aug 07 '17 at 11:21

Micaël Félix

2,697
5
34
46

1

Remove the `(?=<\/span>)` lookahead and replace `+` with `+?`. – Wiktor Stribiżew Aug 07 '17 at 11:23
Thanks, in fact it was just missing one ? after the + – Micaël Félix Aug 07 '17 at 11:25
1

Your lookahead is redundant. `a` = `(?=a)a`. – Wiktor Stribiżew Aug 07 '17 at 11:27
Sorry, I didn't understand that last comment? – Micaël Félix Aug 07 '17 at 14:40
1

Remove the `(?=<\/span>)` lookahead, it does nothing. Itmakes no asense. – Wiktor Stribiżew Aug 07 '17 at 14:53
Oh ok I got it, thanks. ```[\S\s]+?<\/span>``` – Micaël Félix Aug 08 '17 at 14:42

Regex positive lookahead is too greedy

0 Answers0