Mark Little commented on JBTM-653:
The reason basic Xid handling was in ArjunaCore was due to recovery. With the exception of
the node identifier, which was introduced specifically for XA recovery, all of the other
XA specific needs for recovery were moved elsewhere over the years. I see all of what
remained has been moved, but the docs remained talking about XA. However, by some suitable
modifications to the docs as well as the prospect of the related JIRA, it's possible
to make it a clean break.
refactor Xid handling
Project: JBoss Transaction Manager
Issue Type: Task
Security Level: Public(Everyone can see)
Components: Documentation, JTA, JTS, Transaction Core
Affects Versions: 4.8.0
Reporter: Jonathan Halliday
Assignee: Jonathan Halliday
Fix For: 4.9.0
Xid handling is a horrible mess. Creation is excruciatingly slow because XATxConverter
uses Uid.stringFrom, then promptly turns the string back into a bunch of bytes. XID has no
business being in ArjunaCore, or indeed externally exposed at all - it should be internal
to XidImple. Some static utility methods on XATxConverter and elsewhere would be better
off in XidImple too.
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
For more information on JIRA, see: http://www.atlassian.com/software/jira