Much like the frequent and painful abuse of String in Java, like:
String s = “”
for (int i=0; i=0; ) {} //Note that i does not go negative inside the loop.

for (var i = 0, len = document.getElementById(‘t’).getElementsByTagName(‘td’).length; i

for (int i=0; i=0; i--) {} // comparing with 0 is faster but it leads to slightly not very clean code and the difference is not big  <br />&nbsp;

and sure most of the time iterators and enhanced for loops are being used anyway

Well, “document.getElementById(‘t’).getElementsByTagName(‘td’).length” is quite an expensive operation, particularly in JavaScript. It’s about as expensive an operation as you’re likely to find on the end of a for loop! Remember that even looking up an identifier in JavaScript needs to consult a hash table. So, the above expression would have to:

* Lookup “document” in the global identifiers.
* Lookup “getElementById” in the document object’s dictionary.
* Execute the “getElementById” function, passing ‘t’ as an argument.
* Look throughout the document for an element with the id ‘t’. This is rather expensive, depending on how the browser indexes element IDs (could be cheap).
* Check to make sure the returned object is not null; if so throw an exception.
* Lookup “getElementsByTagName” on the returned object’s dictionary.
* Execute the “getElementsByTagName” function, passing ‘td’ as an argument.
* The killer: Search the entire document tree, doing a string comparison against _every_ element’s tag name, and doing a list insert for all successful elements. The browser could intelligently hash tag names but I don’t think they do.
* Finally lookup “length” in the returned list object’s dictionary.
* And execute the “length” function, which cheaply returns the length as an integer.

So it is rather a good example then, of the perfect expression which deserves to be thrown with great vigour out of the loop body!

Good simple article.I liked it and am also not aware of this till now having almost 3.7 yrs experience.

Thanks
Prashant