Splitting hairs^H^H^H^H^H strings with Java

Split Personality

Offtopic: where does ^H come from? (since I too found it only recently) – from the source of all wisdom – Wikipedia :-p

Pressing the backspace key on a computer terminal would generate the ASCII code 08, BS or Backspace, which would delete the preceding character. That control code could also be accessed by pressing Control-H, as H is the eighth letter of the Latin alphabet. Terminals which did not have the backspace code mapped to the function of moving the cursor backwards and deleting the preceding character would display the symbols ^H (caret, H — see Caret notation) when the backspace key was pressed. This sequence is still used humorously for epanorthosis by computer literates, denoting the deletion of a pretended blunder, much like overstriking.

Now back to our topic: how to split a String in Java into parts? Here are a couple of options:

  • Use a StringTokenizer – one thing which surprises Java novices is that the second parameter in the constructor is a list of separators not one big separator. So if you specify ", ", this will split at either commas or spaces, not at commas followed by spaces.
  • Use String.split – this can also surprise novices since – even though it accepts a String parameter – the String will be compiled into regular expression (Pattern). While there are several warning signs (like the documentation – because everyone reads those :-p – or the parameter name), it is easy to ignore and one can get surprising results (like splitting the String "aaa.bbb.ccc" using the pattern "." and getting an empty String).
  • Precompiling a Pattern and using the split method on it. This has the advantage of expressing intent more clearly than the previous method and also being more efficient if one needs to split multiple strings using the same Pattern. However this still can be overkill if you just want to split using a single character (or even a list of characters).
  • Using the StringUtils.split method from the Apache Commons library (which can be used in commercial projects!) or even roll your own.

When considering these alternatives you should consider expressiveness first and performance second (also, don’t optimize until a profiler tells you to!). Luckily StringUtils is a clear choice from both points of view, unless you need something more complex than one character separators. Here are some performance numbers:

  • Custom implementation (Apache Commons should be in the same ballpark) – 1
  • StringTokenizer – 2.1
  • String.split – 3.46
  • Pattern.split – 2.87

These numbers shouldn’t be taken as an absolute and they can depend on many things (for example if a profiler is or isn’t attached, the version of the JVM, etc), but they should point you in the right direction.

Update: a very important note: while these solutions are mostly drop-in replaceable, be aware that there are some differences between them around the corner cases (ie. how they handle null’s, what if the string ends in a delimiter, what if two delimiters follow each other, etc). If and when you decide to swap one for the other, be sure to unittest your piece of code with the types of data you expect it to handle.

So there you have it, more than you ever wanted to know about splitting Strings 🙂 Hopefully this helps somebody.


Leave a Reply

Your email address will not be published. Required fields are marked *