LinkDetector

A Java utility that lets you detect where in plain text a hyperlink begins and ends.

Correctly identifying the start and end location of a link in text can be tricky, especially when those links either include or are surrounded by parenthesis, followed by a comma, and so on:

• https://example.org/foo, instead of https://example.org/foo (from a text like ... on https://example.org/foo, where ...).
• https://example.org/foo) instead of https://example.org/foo (from a text like ... the webpage (https://example.org/foo) where ... ).

These often lead to browsers being opened to invalid URLs, causing end-users to see 404 pages or other errors.

This project simplifies parsing the correct start and end of links in text, which helps avoid such issues.

Usage

The distributable is available through the Maven central repository. You can then define this project to be a dependency of your project, like so:

<dependency>
    <groupId>nl.goodbytes.util</groupId>
    <artifactId>linkdetector</artifactId>
    <version>1.0.0</version> <!-- Please remember to check if this is the latest, as this example could be outdated. -->
</dependency>

To use the utility in your code, invoke the parse method of the LinkDetector class, as shown below. This will split up the text in fragments (returned in a list). For each fragment, a start and end index is provided, and defines if it does or does not represent a link.

final String input = "Please find more information in the corresponding page on "
    + "Wikipedia (https://en.wikipedia.org/wiki/Ambiguity_(disambiguation)). Let me "
    + "know if you have questions!";

final List<Fragment> fragments = LinkDetector.parse(input);

for (final Fragment fragment : fragments)
{
    System.out.println("Fragment starting at index " + fragment.startIndex()
        + ", ending at index " + fragment.endIndex() + " (exclusive) "
        + (fragment.isLink() ? "is" : "is not") + " a link:");
    System.out.println("\t" + fragment);
    System.out.println();
}

The example code above generates the following output:

Fragment starting at index 0, ending at index 69 (exclusive) is not a link:
	Please find more information in the corresponding page on Wikipedia (

Fragment starting at index 69, ending at index 125 (exclusive) is a link:
	https://en.wikipedia.org/wiki/Ambiguity_(disambiguation)

Fragment starting at index 125, ending at index 162 (exclusive) is not a link:
	). Let me know if you have questions!

Build / Compilation

This project should be compatible with any version of Java that is not ancient. It should be compatible with Java 1.4, but to circumvent some issues with modern build tooling, its project descriptor defines 1.8.

The project can be built using standard Maven invocations, like this:

mvn clean package

The project does not use any external dependencies (although for testing, the JUnit library is added to the test scope of the build process).

Attribution

This is but a simple Java wrapper around a regular expression that was provided by Wiktor Kwapisiewicz.