Extract URLs from String (Ruby) (Regex and link shortened)

Question

I heard that URI::extract() only returns links with a :, however since I am grabbing a tweet, and it does not contain a :, I believe I would have to use a regular expression. I need to check for a "swoo.sh/whatever" link, and store it to a variable. However, how could I look for the first (which it returns automatically apparently), "swoo.sh/whatever" link, in regards to that I have to maintain everything after the /. For example, if the tweet says

Lorem ipsum lorem ipsum swoo.sh/12xfsW lorem ipsum

How would I grab the swoo.sh link, and all the different things that come directly after the /?

I would assume such links are clickable on twitter, which means the original HTML would have the actual URI in it, making this task trivial. Are you sure you can't use a different API/scraper to get the actual HTML content of the tweet? — Max
– Max, Commented May 9, 2018 at 13:55

Tim Biegeleisen · Accepted Answer · 2018-05-09 01:52:56Z

1

Here is one approach using match:

match = /(\w+\.\w+\/\w+)/.match("Lorem ipsum lorem ipsum swoo.sh/12xfsW lorem ipsum")
if match
    puts match[1]
else
    puts "no match"
end

Demo

If you also need the simultaneous ability to capture full URLs, then my answer would have to be updated. This only answers your immediate question.

answered May 9, 2018 at 1:52

Tim Biegeleisen

526k32 gold badges324 silver badges399 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

evan Over a year ago

Thank you. You answering my immediate question was all that was needed!

Max · Accepted Answer · 2018-05-09 13:52:22Z

We can use the fact that URIs can't contain spaces and Ruby has URI::Generic which will parse almost anything that looks URI-ish. Then we just need to filter out non-web-URIs, which I do by assuming that every web URI has to start with something like foo.bar

require 'uri'
require 'pathname'

tweet.
  split.
  map { |s| URI.parse(s) rescue nil }.
  select { |u| u && (u.hostname || Pathname(u.path).each_filename.first =~ /\w\.\w/) }

Example output

tweet = 'foo . < google.com bar swoosh.sh/blah?q=bar http://google.com/bar'
# the above returns
# [#<URI::Generic google.com>, #<URI::Generic swoosh.sh/blah?q=bar>, #<URI::HTTP http://google.com/bar>]

This can't really work in general because of ambiguity. "car.net" looks like a shortened link, but in context it could be "my neighbor threw a baseball through my window so i yanked the hubcabs off his car.net gain!!!", where it's clearly just a missing space.

Collectives™ on Stack Overflow

Extract URLs from String (Ruby) (Regex and link shortened)

2 Answers 2

Demo

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related