[concurrency-interest] ThreadLocalRandom clinit troubles

Alan Bateman Alan.Bateman at oracle.com
Wed Jul 2 06:44:47 EDT 2014

On 02/07/2014 08:10, Peter Firmstone wrote:
> How does one go about it? I'm not on any lists or part of the project.
This is the best place to start:

At a high-level then the types of improvements that you listed in your 
summary seem to be very good. It's probably best to start out small and 
propose one or two small changes first to get used to contributing and 
also because some of the changes are likely to require detailed review 
and discussion.

> :
>> On RFC 3986 (you mentioned this a number of times) then there were 
>> previous attempts bring URI up to this, unfortunately had to be 
>> backed out due to compatibility issues and other breakage. It's 
>> definitely something that needs to be looked at again.
> Yes turns out RFC 3986 is very useful.  Our implementation is strictly 
> compliant, with the exception of making upper case path comparisons 
> for file URI's on certain platforms, it's based on Harmony's URI, but 
> has been refactored with immutability; final class, final fields, 
> isn't serializable etc.  The parser is called during construction 
> using static methods.  Exceptions thrown by constructors are thrown by 
> static methods prior to Object's default constructor being called, to 
> avoid creating partially constructed objects.
java.net.URI is based on the older RFC 2396 and 2732. I think bringing 
it up to RFC 3986 while maintaining compatible could be a major project 
(previously attempts to do this back in JDK 6 were backed out due to the 
breakage that it caused). It's a good topic to bring up on the OpenJDK 
net-dev list where URI is maintained, it's definitely something that 
needs to happen at some point.


More information about the Concurrency-interest mailing list