You want to find URLs in a larger body of text. URLs may or may not be enclosed in punctuation that is part of the larger body of text rather than part of the URL. You want to give users the option to place URLs between quotation marks, so they can explicitly indicate whether punctuation, or even spaces, should be part of the URL.

(?:(?:https?|ftp|file)://|(www|ftp).)[-A-Z0-9+&@#/%?=~_|$!:,.;]*
                                        [-A-Z0-9+&@#/%=~_|$]
|"(?:(?:https?|ftp|file)://|(www|ftp).)[^"
]+"
|'(?:(?:https?|ftp|file)://|(www|ftp).)[^'
]+'

The previous recipe explains the issue of mixing URLs with English text, and how to differentiate between English punctuation and URL characters. Though the solution to the previous recipe is a very useful one that gets it right most of the time, no regex will get it right all of the time.

If your regex will be used on text to be written in the future, you can provide a way for your users to quote their URLs. The solution we present allows a pair of single quotes or a pair of double quotes to be placed around the URL. When a URL is quoted, it must start with one of several schemes: ‹https?|ftp|file› or one of two subdomains ‹www|ftp›. After the scheme or subdomain, the regex allows the URL to include any character, except for line breaks, and the delimiting quote.

The regular expression as a whole is split into three alternatives. The first alternative is the regex from the previous recipe, which matches an unquoted URL, trying to differentiate between English punctuation and URL characters. The second alternative matches a double-quoted URL. The third alternative matches a single-quoted URL. We use two alternatives rather than a single alternative with a capturing group around the opening quote and a backreference for the closing quote, because we cannot use a backreference inside the negated character class that excludes the quote character from the URL.

We chose to use single and double quotes because that’s how URLs commonly appear in HTML and XHTML files. Quoting URLs this way is natural to people who work on the Web, but you can easily edit the regex to allow different pairs of characters to delimit URLs.

Techniques used in the regular expressions in this recipe are discussed in Chapter 2. Recipe 2.2 explains how to match nonprinting characters. Recipe 2.3 explains character classes. Recipe 2.6 explains word boundaries. Recipe 2.8 explains alternation. Recipe 2.9 explains grouping. Recipe 2.12 explains repetition.