[
https://issues.apache.org/jira/browse/DERBY-6796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14342793#comment-14342793
]
Trejkaz commented on DERBY-6796:
--------------------------------
Nope, I can't figure out how to test this in isolation.
If I do the obvious least-effort test (define a new system property, point it
to a subclass of DirStorageFactory), the issue still occurs (in the debugger, I
can clearly see it getting the wrong value), but somehow Derby somehow falls
back to the other factory and a database is still created at the location I
specified.
Is there some way to turn off that sort of fallback?
> Impossible to define new sub-subprotocols
> -----------------------------------------
>
> Key: DERBY-6796
> URL: https://issues.apache.org/jira/browse/DERBY-6796
> Project: Derby
> Issue Type: Bug
> Components: Miscellaneous
> Affects Versions: 10.9.1.0
> Reporter: Trejkaz
> Attachments: wrong-prefix-cut.patch
>
>
> We discovered this in 10.9.1.0 but I looked at the source in trunk and the
> issue is still there.
> To define a new sub-subprotocol (or I guess override an existing one?) by
> setting a system property, you are supposed to set
> "derby.subSubProtocol.myproto" with the name of your storage factory class as
> the value.
> Problem is, the way these are loaded strips the wrong prefix off the front of
> the string. BaseMonitor.java:
> {code}
> if (key.startsWith(Property.MODULE_PREFIX)) {
> tag = key.substring(Property.MODULE_PREFIX.length());
> } else if (key.startsWith(Property.SUB_SUB_PROTOCOL_PREFIX)) {
> tag = key.substring(Property.MODULE_PREFIX.length());
> } else {
> continue nextModule;
> }
> {code}
> So you get "rotocol.myproto" as the protocol you have to use in your
> connection strings.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
