RFR: 6519127 : user.home property not set correctly

Alan Bateman Alan.Bateman at oracle.com
Fri Oct 19 09:51:08 UTC 2012

On 19/10/2012 10:11, Sean Chou wrote:
> Hi all,
>      This is a fix for 651912. The fix removed the code to read
> user.home from registry and get profile directory instead of the
> parent of desktop directory. As suggested in
> http://blogs.msdn.com/b/oldnewthing/archive/2003/11/03/55532.aspx from
> the bug page.
> Bug: bugs.sun.com/bugdatabase/view_bug.do?bug_id=6519127
> Webrev: http://cr.openjdk.java.net/~zhouyx/6519127/webrev.00/
> Please take a look.
Sean - any changes to the value of user.home has the potential to cause 
issues, this is an area whether there has been a lot of ambiguity in 
past as to where the user's home directory is, particularly with 
non-standard configurations. 6519127 and several others have been 
waiting for a brave person to re-examine this issue.

Do you think you could provide a detailed write-up on this topic? I 
think we need to understand how the home directory, as reported by the 
shell API, might differ to the value that the registry reports, also 
USERPROFILE. In addition we've had so many issues over the years with 
different configurations, Windows Terminal Server, local accounts vs. 
accounts in the domain, etc. and I think we need to understand as much 
as possible as what might change. Sorry there is lot more work here than 
you might have expected but this is an area where we have to be very 


More information about the core-libs-dev mailing list