This content has been marked as final. Show 8 replies
Thanks for the feedback, it is good to know I am not alone :-)
I did some further tests.
I enabled logging in sqldeveloper.conf, and found this query I believe to be the culprit:
Prepared:with src_name as (select source n from Dba_java_classes
where lower(name)=lower(:NAME) and rownum = 1)
select text||decode(substr(text,length(text)),chr(10),null,chr(13),null,chr(10)) source
where owner= :OWNER
and type = 'JAVA SOURCE'
and (name in (select n from src_name) or name in (select owner || '/' || n from src_name)
or lower(name) = lower(:NAME))
order by line
Previously :NAME was populated with the java short name, while now it has the java long name.
In fact, if I compile a class where short name and long name are the same, it works.
In DBA_JAVA_CLASSES name is the long name, while in DBA_SOURCE it is the short name (at least this is what I observed).
So I believe that this is truly a regression.
Anyone (from dev) can confirm this?
Thanks for reporting this. Looks like you will have to keep 3.1 available alongside 3.2 for the time being. I logged a bug:
Bug 14535076 - FORUM - REGRESSION - UNABLE TO BROWSE JAVA SOURCE IN THE DB
SQL Developer Team
I logged bug 14535076 without a test case, so it went nowhere -- often the case if the developer has no time to create his/her own case.
Fortunately a similar bug, Bug 14844305 - JAVA SOURCE NOT SHOWN, was logged with a test case, so odds are it will get fixed in the next release. Bug 14535076 has been closed as a duplicate of that more recent bug.