If I have a string that is this
<f = x1106f='0'>something
and The values inside <> can change How would I use regular expressions to isolate "something" and replace the tag?
EDIT:
<(.*?)> Pattern worked
-
A lot of people are going to tell you to not parse HTML yourself. You almost always want to use something robust like Nokogiri.Max– Max2013-09-20 03:39:40 +00:00Commented Sep 20, 2013 at 3:39
-
@Max it's not really "parsing" but merely extracting a value from a specific tag.Nir Alfasi– Nir Alfasi2013-09-20 03:48:26 +00:00Commented Sep 20, 2013 at 3:48
-
@alfasin with questions like this it's hard to tell if this is an isolated string extraction problem where regexs are appropriate, or if they've already torn apart a larger document with regexs and string methods and they're stuck on this one part.Max– Max2013-09-20 16:21:13 +00:00Commented Sep 20, 2013 at 16:21
Add a comment
|
2 Answers
What you need is
string =~ />([^<]+)/
and the something will be captured in $1.
Comments
Use the following regex to capture "something":
(?<=>)(.*?)(?=<)
assuming that after "something" there's a closing tag.
Link to fiddle
6 Comments
Nir Alfasi
Downvoting without commenting is lame!
I didn't downvote, but for the sake of efficiency, the lookbehind and lookahead are unnecessary here with your capture group. Your regexp is looking for anything that follows a
> and precedes a </. However, this regexp would provide the same match: >(.*)<\/. Also, the .* would be overly greedy - a .*? is safer - or even better [^<]* (e.g. if the string were <f = x1106f='0'>something</f><f = asdf>somethingelse</f>, the match would be something</f><f = asdf>somethingelse).Nir Alfasi
@ChicagoRedSox thanks for your comment - I changed the regex (and fiddle) to support the scenario you showed. The reason I'm using lookaround is tin order to have a group-match which can be referred and used later on in the code.
Nir Alfasi
@ChicagoRedSox I wasn't sure but that's what I understood from "use regular expressions to isolate "something" and replace the tag?"
|