Skip match if in specific context

Question

The regex pattern is required to match all instances of a specified word, that does not need to match only individual words but also in-word content.

E.g. searching for media match the instance in mediator.

There are few exceptions. If the word content is within a url or a font-family declaration the match should SKIP. This is the code i came up so far, but i'am missing something as it skips all.

(?:font-family:|https?:\/\/)[^\s\'";}]*(*SKIP)(*FAIL)(media)

The above can be tested at Regex101.

Please include a sample of text from which you are trying to extract the matches. — Tim Biegeleisen, Oct 25 '19 at 10:43
You need a `|` after SKIP FAIL and perhaps also match media on the left side to make sure it is part of the font-family or https? `(?:font-family:\s*|https?:\/\/)[^\s'";}]*\bmedia\b(*SKIP)(*FAIL)|\bmedia\b` See https://regex101.com/r/HCZ473/1 but I am not sure this will cover all the ways you might specify `font-family:` — The fourth bird, Oct 25 '19 at 10:47
@Thefourthbird thanks so much for your help, obviously i omited the | operator. I come up with this regex https://regex101.com/r/5CX3Ea/3/ but it still catch the "font-family: media" — kole23, Oct 25 '19 at 10:50
@kole23 Did you try this pattern? https://regex101.com/r/jAjmUK/1 There are 1+ spaces after font-family: — The fourth bird, Oct 25 '19 at 10:54
Looks like you want to [skip the stuff inside braces](https://regex101.com/r/5CX3Ea/4). — bobble bubble, Oct 25 '19 at 11:21

The fourth bird · Accepted Answer · 2019-10-25T11:04:32.113

You could add a pipe after (*SKIP)(*FAIL) and there is a space after font-family:

Note that you don't have to escape the \' in the character class and if you change the delimiter to something else than / you also don't have to escape those. You can also omit the capturing group around media.

(?:font-family:\h*|https?:\/\/)[^\s'";}]*(*SKIP)(*FAIL)|\bmedia\b

(?: Non capturing group
- font-family:\h* Match font:family: and 0+ horizontal whitespace chars
- | Or
- https?:\/\/ Match http and optional s
) Close non capturing group
[^\s'";}]* Match 0+ times any char except '"; or a whitespace char
(*SKIP)(*FAIL) Consume and avoid using SKIP and FAIL
| Or
\bmedia\b Match media between word boundaries

Regex demo

Skip match if in specific context

1 Answers1