Thanks for catching this, Myrna, and thanks for everyone else's
commentary. I introduced an assertion in DDMWriter which wasn't
protected by a SanityManager.DEBUG check. I have added that check and
the "SANITY >>>" diagnostic went away.
Thanks,
-Rick
Myrna van Lunteren wrote:
On Wed, Jan 20, 2010 at 2:40 PM, Knut Anders Hatlen <[email protected]> wrote:
Myrna van Lunteren <[email protected]> writes:
On Wed, Jan 20, 2010 at 6:46 AM, Rick Hillegas <[email protected]> wrote:
Hi Myrna,
These look like more fallout from a large submission I committed on Friday
(subversion revision 899733 for DERBY-4491). I have fixed the DRDAConnThread
javadoc warning with revision 901219.
I don't see the second diagnostic ("Failed with following errors") when I
build the javadoc. What command are you using to build the docs?
Thanks,
-Rick
Myrna van Lunteren wrote:
I apologize - I'm using a slightly hokey homegrown utility that checks
on expected output of the build by checking each line against a list
of lines to ignore as 'ok'.
So that's where that line came from, not ant javadoc at all.
This was (after doing ant -Ddeprecation=off -Dsane=false all, twice)
the result from
ant -Ddeprecation=off buildjars. The output contains this section:
derbynetjar:
[echo] Beginning derbynet.jar insane build
[echo] creating net.list
[java] SANITY >>> /org/apache/derby/impl/drda/DDMWriter.class
[echo] creating new net.properties file
That line with SANITY is not on my list of lines to be ignored, and
so, it is new.
It's the only occurrence of that line in my build output, and looking
at SanityManager.java I don't understand why this would come up in the
build output...
Any ideas?
classlister.java prints this warning if insane jars contain the
SanityManager class. One of the commits Rick mentioned above introduced
a call to a SanityManager method in insane jars, which caused the
SanityManager class to be included also in the insane jars, hence the
warning. I think Rick backed out the changes that caused it. Or are you
still seeing the warning?
--
Knut Anders
Well, yes.
I synced up to 901399 and am (still) seeing this.
Maybe you're referring to Rick's fixing a sealing violation with
revision 899819...
(http://svn.apache.org/viewvc?view=revision&revision=899819)
Maybe this is then another instance of the same?
Myrna
