How to overcome preg_match 'offset is higher than subject length'?

Asked Jul 02 '18 at 12:54

Active Jul 02 '18 at 13:03

Viewed 80 times

The pattern ~(\w+-?)+[0-9]+(?=\.html)~ returns false on www.test.com/this/works/12345678901234567/2009/08/2009819107554384978.html

But it returns 1 on www.test.com/this/works/12345678901234567/2009/08/200981910755438497.html (one character less).

The PHP documentation reads:

'Returns FALSE if offset is higher than subject length.'

How so?

edited Jul 02 '18 at 13:03

asked Jul 02 '18 at 12:54

simpatico

1

It is due to the catastrophic backtracking. Use [`'~\w+(?:-\w+)*[0-9]+(?=\.html)~'`](https://regex101.com/r/xsapiG/1). – Wiktor Stribiżew Jul 02 '18 at 12:56
It's still not clear to me if this is a fixable syntax issue (if so how), or if I have to re-think the regex. – simpatico Jul 02 '18 at 13:07
Although it has issues on greediness, I can't reproduce https://3v4l.org/DSW4Z – revo Jul 02 '18 at 13:09
Check it out here: https://3v4l.org/WDtjtA – simpatico Jul 02 '18 at 13:12
Now you changed the input. If you read second duplicate marked question, you will find some useful info. – revo Jul 02 '18 at 13:14
To find the error you could use `preg_last_error()` as in `var_dump(preg_last_error() === PREG_BACKTRACK_LIMIT_ERROR);` which returns `true`. – revo Jul 02 '18 at 13:18

0 Answers0