Hi Amin,
I updated your patch and attached it to HSEARCH-421, hope you've seen
it and are able to look into it, especially the PerDPQueueProcessor
needs some more thoughts.
Also I was going to ask you to prepare a test as I couldn't really
figure it all out by reading the code, but as it's quite tricky to
test this and because I wasn't sure it would be possible I wrote it
myself to find out.
The test is almost complete but some lines are commented out, as exact
implementation depends on how you're going to implement the changes
needed in PerDPQueueProcessor.
Cheers,
Sanne
2010/4/2 Amin Mohammed-Coleman <aminmc(a)gmail.com>:
Hi
I attached a patch for HSearch 421 yesterday which Sanne is reviewing (there 3 changes
that I believe he wanted to make). If Sanne is happy with the patch I can work on the
documentation side of things tomorrow and hopefully get something on Sunday/Monday.
Thanks
Amin
On 2 Apr 2010, at 15:35, Emmanuel Bernard wrote:
> I've reviewed the list of bugs you've sent as well as all the bugs assigned
to 3.2
>
> I think the right thing to do is:
> - cut a 3.2 CR1 based on what is labeled 3.2 beta2
> - move 3.2 beta3 and co to 3.3
>
> That would mean delay the use of Lucene 3 to HSearch 3.3. I think most issues are in
control for CR1, the big one is Amin's work on exceptions.
>
> Sanne, let's try and chat by IM, I have a couple of questions on beta3's bugs
related to what you've done. Aside from that, let's try and target CR1 for end of
next week (I'm likely busy till Wed and free afterwards). Amin do you think you can
get HSEARCH0-421 done by that time?
>
> Emmanuel
>
> On 1 avr. 2010, at 19:51, Emmanuel Bernard wrote:
>
>> IM me tomorrow.
>>
>> On 1 avr. 2010, at 19:46, Sanne Grinovero wrote:
>>
>>> Hi Emmanuel,
>>> maybe you could spare some time to do some planning on Search? There
>>> are so many needed fixes here :D
>>> Also my current task is in wait for feedback.
>>>
>>> From the forums it appears that many have already upgraded to Lucene
>>> 2.9 wo checking compatibility - just assuming it would work - and are
>>> waiting or having trouble.
>>> And it's quite likely people are going to use Core 3.5 now, having
>>> Search unreleased is annoying.
>>>
>>> below a quick overview, I actually believe that you could cut a new
>>> beta in short time:
>>>
>>> Urgently needed stuff
>>> ============================
>>>
>>> HSEARCH-484 Building documentation fails Unassigned
>>>
>>> HSEARCH-476 Work queue contains the same operation multiple times
and
>>> some add/delete operations have null primary keys Emmanuel Bernard
>>> * you probably fixed this already, to give it another look; maybe I
>>> can add some additional tests in next days
>>>
>>> HSEARCH-421 Exceptions happening in backend are unnoticed
Unassigned
>>> * Amin is working on it, he should be close
>>>
>>> HSEARCH-425 Reimplement support for compressed fields (support
>>> removed in Lucene3) Sanne Grinovero
>>> * I'm waiting for feedback, we need to talk about this - I'm also
>>> open to remove it
>>>
>>> HSEARCH-428 Update the Getting Started section of the documentation
>>> Hardy Ferentschik
>>> HSEARCH-483 update readme.txt Emmanuel Bernard
>>>
>>> Other things which could be moved on
>>> ============================
>>> Refactoring can wait, and others are related to bugs we did postpone
>>> many other times as we can't reproduce, or are actually fixed already
>>> but need to be checked;
>>> stress tests would be nice, but I'm not going to make them soon :
>>>
>>> HSEARCH-346 Setup hudson job testing Search against several
databases
>>> Juraci Paixao Krohling
>>> HSEARCH-318 Extend IndexWriterSetting to take additional constructor
>>> argument Unassigned
>>> HSEARCH-467 publish Maven artifacts containing the testsuite
Hardy
>>> Ferentschik
>>> HSEARCH-468 Refactor build to have a module setup Hardy
Ferentschik
>>> HSEARCH-444 Search fails to start when core uses XML mappings making
>>> use of entity-name attribute Emmanuel Bernard
>>> HSEARCH-393 unit test of the NotSharedReaderPerfTest hangs on with
>>> some database Emmanuel Bernard
>>> HSEARCH-320 Hibernate Search conflicts with Spring's class path
>>> scanning Hardy Ferentschik
>>> HSEARCH-228 Write a reliability test under stress Hardy
Ferentschik
>>> HSEARCH-314 ThreadLocal in ContextHolder causes memory leak when
>>> deployed in a web container Unassigned
>>> HSEARCH-333 Remove or improve log warning in
>>> DocumentBuilderContainedEntity.checkDocumentId Hardy Ferentschik
>>> HSEARCH-371 Refactor DocumentBuilder and introduce abstract base
>>> class Hardy Ferentschik
>>>
>>> Cheers,
>>> Sanne
>>
>> _______________________________________________
>> hibernate-dev mailing list
>> hibernate-dev(a)lists.jboss.org
>>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>