在查看字节缓冲输出流的源码时候发现这段代码看不懂，谁能解释一下啊

1. /**
   
2. * Writes <code>len</code> bytes from the specified byte array
   
3. * starting at offset <code>off</code> to this buffered output stream.
   
4. *
   
5. * <p> Ordinarily this method stores bytes from the given array into this
   
6. * stream's buffer, flushing the buffer to the underlying output stream as
   
7. * needed. If the requested length is at least as large as this stream's
   
8. * buffer, however, then this method will flush the buffer and write the
   
9. * bytes directly to the underlying output stream. Thus redundant
   
10. * <code>BufferedOutputStream</code>s will not copy data unnecessarily.
    
11. *
    
12. * @param b the data.
    
13. * @param off the start offset in the data.
    
14. * @param len the number of bytes to write.
    
15. * @exception IOException if an I/O error occurs.
    
16. */
    
17. public synchronized void write(byte b[], int off, int len) throws IOException {
    
18. if (len >= buf.length) {
    
19. /* If the request length exceeds the size of the output buffer,
    
20. flush the output buffer and then write the data directly.
    
21. In this way buffered streams will cascade harmlessly. */
    
22. flushBuffer();
    
23. out.write(b, off, len);
    
24. return;
    
25. }
    
26. if (len > buf.length - count) {
    
27. flushBuffer();
    
28. }
    
29. System.arraycopy(b, off, buf, count, len);
    
30. count += len;
    
31. }

复制代码

首先在写字节数组的时候又调用了写字节数组的方法，这样不会无限递归吗
下面第二个if我怎么感觉不会执行到呢count默认值是0啊，如果第一个if不满足第二个if应该也不会满足条件的啊。
还有下面的几行麻烦代价解释一下

代码看着太费劲，，，，

第一个if意思是说：当传入的len大于等于buf.length时，就将数据刷新并且写到相应的缓存中去。第二个if是说:当len大于buf.length - count时就向缓存刷新一次。count从后面的代码看，应该是用来计算数据的长度的。