|
|
|
| Authored by: Ed L. on Friday, April 27 2012 @ 02:49 PM EDT |
This (operator overloading) is a very useful and powerful technique
in object oriented programming.
But not in
Java.
:-)
---
Real Programmers mangle their own memory. [ Reply to This | Parent | # ]
|
| |
| Authored by: kuroshima on Friday, April 27 2012 @ 03:00 PM EDT |
Java does not allowing overloaded operators. There is one
overloaded operator in Java, and it's '+', that can be
applied to String objects only (and it's actually a bad
practice to do so, because of the way String objects work*)
to concatenate them. The correct way to handle objects is
with methods. I actually think that not using overloaded
operators leads to cleaner, but more verbose code.
*Strings in Java are strange. Once a String object is
created, it is immutable (it can not be changed), meaning
that to concatenate two String objects, you need to create a
new String object that contains both character strings from
both String objects. Now take a simple program that loops
over a file, reads a line and adds the line to a String
object. If you use +, you end with a memory usage that is
geometric with the number of lines in the file (You don't
explicitly free memory in Java, you let the garbage
collector clean unreferenced objects for you. Helps prevent
memory leaks, but if you don't code well, your program will
stop from time to time while the garbage collector tries to
clear some memory), in that the first line will be read fine
and dandy, then you end up creating a new String that
contains the first line and the second, one that contains
the first, second and third,.... Java has StringBuilder and
StringBuffer to avoid this.
See http://chaoticjava.com/posts/stringbuilder-vs-string/
for a better treatment of this.[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Friday, April 27 2012 @ 09:28 PM EDT |
Java's implementation of max() is NOT a trivial a>=b?a:b. Besides the
overloading for several types, the float and double versions have extra code to
support NaN and signed zeros (nastiness from IEEE754).[ Reply to This | Parent | # ]
|
|
|
|
|
