..




FULL PRODUCT VERSION :
java version "1.4.0-beta3"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.0-beta3-b84)
Java HotSpot(TM) Client VM (build 1.4.0-beta3-b84, mixed mode)

FULL OPERATING SYSTEM VERSION :
SunOS aloysius 5.8 Generic_108528-05 sun4u sparc
SUNW,Ultra-5_10



A DESCRIPTION OF THE PROBLEM :
I have a sourcepath /a/path.  The topmost source package
is atg.  It contains packages foo and bar.  If I pass
subpackages atg.foo:atg.bar, everything builds properly.

However, if I pass subpackages just atg, apparently nothing
happens (nothing appears in verbose output), but cpu use
is ~90% indefinitely until javadoc is killed.

This bug can be reproduced always.

CUSTOMER WORKAROUND :
Don't use -subpackages.  Continue to use script-assembled
list of packages to hand to javadoc.
(Review ID: 138439) 
======================================================================

If you fix all the compiler-like error messages that appear before the
hang, that should make the problem go away.  Specifically, fix the 
"package ... does not exist" and "cannot resolve symbol" errors.

 xxxxx@xxxxx  2002-01-29

Received a test case from submitter
 xxxxx@xxxxx  2002-01-26

I ran his test and I don't get the infinite loop.  However,
it does stop at exactly the same place his does.   It hung there
for 10 minutes before I killed the process.  This is the script I ran:

/java/re/jdk/1.4.0/promoted/latest/binaries/solsparc/bin/javadoc -verbose -J-Xmx1000M -d html -sourcepath . -subpackages atg

where the current directory contains the top-level package:

I ran this on jano.sfbay, so I don't think machine resources should
be a problem.  So I am changing the synopsis in the bug from:
   -subpackages loops infinitely if given topmost level package

   Hangs when loading javax/swing/JPanel$AccessibleJPanel.class from rt.jar
and reassigning it to  xxxxx@xxxxx 

 xxxxx@xxxxx  2002-01-26

Am committing to hopper, as it is the only hang I'm aware of.  
Perhaps Jamie can work on this bug.
 xxxxx@xxxxx  2002-01-26