There have been some reports of linking text using a regular expression. The most popular is this post .
However, my specification is a bit more complicated:
describe TextFormatter do def l(input) TextFormatter.gsub_links!(input){|link| "!!#{link}!!"} end it "should detect simple links" do l("http://www.cnn.com").should == "!!http://www.cnn.com!!" end it "should detect multi links" do l("http://www.cnn.com http://boats.com?help.asp").should == "!!http://www.cnn.com!! !!http://boats.com?help.asp!!" end it "should compensate for parans properly" do l("(http://this.is?hello_world)").should == "(!!http://this.is?hello_world!!)" end it "should ignore existing links" do s = "<A HREF='http://sam.com'> http://sam.com </A>" l(s.dup).should == s end it "should allow parans" do l("http://sam.com.au?(red)").should == "!!http://sam.com.au?(red)!!" end end
Any ideas on how to implement a hairy Regex:
Here I am still (it did not pass 2 tests):
def gsub_links!(input) regex = /https?\:\/\/[\-\w+&@#\/%?=~\(\)\|!:,.;]*[\-\w+&@#\/%=~_\(\)|]/ input.gsub!(regex) { |link| yield link } end
ruby regex linkify
Sam saffron
source share