Comments on: Lesson learned: copying bytes from an OutputStream to an InputStream in Weblogic 8 http://www.mooreds.com/wordpress/archives/277 Tue, 07 Feb 2012 14:11:43 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Carl Smotricz http://www.mooreds.com/wordpress/archives/277/comment-page-1#comment-2320 Carl Smotricz Sun, 16 Nov 2008 13:26:47 +0000 http://www.mooreds.com/wordpress/?p=277#comment-2320 oops... code in comment mangled by HTML. One more try:<code> while (true) {   int count = bis.read(data);   if (count < 0) break;   bos.write(data, 0, count); } </code> oops… code in comment mangled by HTML. One more try:
while (true) {
  int count = bis.read(data);
  if (count < 0) break;
  bos.write(data, 0, count);
}

]]> By: Carl Smotricz http://www.mooreds.com/wordpress/archives/277/comment-page-1#comment-2319 Carl Smotricz Sun, 16 Nov 2008 13:23:52 +0000 http://www.mooreds.com/wordpress/?p=277#comment-2319 This post is over 3 years old, but I thought I'd respond anyway and clear up the problem. In the code fragment above, Dan ignores the data count from bis.read() . Depending on the size of the text, the last call to read() will almost never read exactly 1024 bytes. My preferred solution is:<code> while (true) { int count = bis.read(data); if (count </code> The solutions quoted from the cookbook and tutorial have a performance problem: The read() and write() methods, as well as the underlying operating system calls, are repeated for each byte. With megabytes of data, we'd be talking a lot of unnecessary overhead. Buffering is good! This post is over 3 years old, but I thought I’d respond anyway and clear up the problem.

In the code fragment above, Dan ignores the data count from bis.read() . Depending on the size of the text, the last call to read() will almost never read exactly 1024 bytes. My preferred solution is:
while (true) {
int count = bis.read(data);
if (count

The solutions quoted from the cookbook and tutorial have a performance problem: The read() and write() methods, as well as the underlying operating system calls, are repeated for each byte. With megabytes of data, we’d be talking a lot of unnecessary overhead. Buffering is good!

]]>