[hibernate-dev] NoORM IRC meeting minutes

Tue Apr 9 09:50:26 EDT 2019

Hi,

Here are the minutes of this week's meeting.

The bot was not responding so it's plain text FTW:

15:19 < gsmet> #startmeeting
15:19 < gsmet> hmmm
15:20 < gsmet> looks like the bot has a meeting too
15:20 < koentsje> lol
15:20 < gsmet> I'll copy paste things, I suppose
15:20 < fax4ever> \o/
15:20 < gsmet> #topic Progress Davide
15:20 < DavideD> Hi
15:21 < DavideD> I worked on the delete and persist of an entity using the
reactive approach
15:21 < DavideD> I think I finally figure out what's needed
15:21 < gsmet> nice
15:22 < sannegrinovero> DavideD, there's a lot of people eager to see that
working ;)
15:22 < DavideD> It's nce because the query is still generated by ORM but
executed with the asyncronous driver
15:23 < DavideD> TO be fair there are still some part that are not
ayncronous but for simple case scenario shouldn't matter
15:23 < DavideD> I'm working on the find right now
15:23 < DavideD> I would like to finish that one because it's a bit more
complicated as it requires to return the entity
15:24 < DavideD> In the other cases I have nothing to return
15:24 < DavideD> I think that's it from me
15:24 < gsmet> #topic Next 2 weeks Davide
15:25 < DavideD> I will work on session.find
15:25 < DavideD> and then if there is time, an updata so that at least
there is a working example for the CRUD operations
15:25 < DavideD> I mean on entity updates
15:26 < gsmet> it's based on ORM 6 or 5.x?
15:26 < DavideD> ORM 6
15:26 < gsmet> cool
15:26 < DavideD> I think that's all
15:26 < gsmet> #topic Progress Fabio
15:27 < fax4ever> In the last two weeks...
15:27 < fax4ever> Firstly, I finished the three issues related to override
and ignore the analyzers,
15:27 < fax4ever> for the comparison values provided in the queries.
15:27 < fax4ever> The feature has been released within the 6.0.0.Alpha4.
15:27 < fax4ever> Secondly, I fixed some mismatch between Lucene and
Elasticsearch behavior.
15:27 < fax4ever> In particular, on how which they handle the
`minimumShouldMatch` parameter.
15:28 < fax4ever> In agreement with Yoann, I forced the Elasticsearch
backend to behave as Lucene one.
15:28 < fax4ever> Thirdly, I fixed a wrong error message, inherited from
Search 5.
15:28 < fax4ever> Fourthly, I added the generic type to the PropertyHadler
mapper Pojo class,
15:28 < fax4ever> so that we would able to avoid some casts.
15:28 < fax4ever> Fiftly, I reviewed several pull requests.
15:28 < fax4ever> Finally, yesterday I started the issue to support the
`indexNullAs` field parameter again.
15:28 < fax4ever> I'm talking about it in the next topic.
15:29 < fax4ever> I think we can move to it, thanks.
15:29 < gsmet> #topic Next 2 weeks Fabio
15:29 < fax4ever> Continuing on 'indexNullAs' parameter issue.
15:29 < fax4ever> I can tell you that the Elasticsearch part has been
already done and
15:29 < fax4ever> you can find it in a preview pull request.
15:29 < fax4ever> Essentially, we have used the 'null_value' Elasticsearch
field mapping option.
15:29 < fax4ever> With this implementation the DSL API method indexNullAs
is type safe.
15:30 < fax4ever> It means that the indexNullAs method requires a value
consistent with the type of the field.
15:30 < fax4ever> Then using this feature in the bridges is type safe.
15:30 < fax4ever> On the other hand, the annotation property requires a
string,
15:30 < fax4ever> so I'm working on the conversion ( parsing? ) part,
15:30 < fax4ever> to reuse the method through the annotation interface.
15:30 < fax4ever> Furthermore, the list of the issues Yoann has left me
it's still long :)
15:31 < fax4ever> As usual, I'll follow the precedence order defined in the
current sprint.
15:31 < fax4ever> You can watch our progress going to the sprint page:
HSEARCH - 2019-06.
15:31 < fax4ever> In a nutshell, they are about:
15:31 < fax4ever> (1) Strengthen some test on dates
15:31 < fax4ever> (2) Use the new Lucene 8 default normalizer
15:31 < fax4ever> (3) Make some changes on our mass indexer progress monitor
15:31 < fax4ever> (4) Allow annotations to be defined on default interface
methods too
15:31 < fax4ever> (5) Restore some compatibility checks on analyzers
15:31 < fax4ever> (6) Make some classes thread safe
15:32 < fax4ever> (7) Support BigDecimal and BigInteger.
15:32 < fax4ever> The last would use some introspections from Hibernate ORM
on field definition,
15:32 < fax4ever> such as @Column.lenght, to convert the BigDecimal into a
long.
15:32 < gsmet> I don't think we need the full list of items of the next
sprint
15:32 < fax4ever> Told that I think that's all from me.
15:32 < gsmet> OK
15:32 < gsmet> #topic Progress Koen
15:32 < fax4ever> Thank you for your attention.
15:33 < koentsje> i have released hibernate tools 5.4.2.Final
15:33 < koentsje> i have also integrated this new release into jboss tools
15:33 < koentsje> i did some testing and verification using java 12 and 13
15:33 < koentsje> on jboss tools hibernate that is
15:33 < koentsje> i have found no issues so far
15:34 < koentsje> (which doesn’t mean they don’t exist)
15:34 < koentsje> and then i worked on quarkus-eclipse, the eclipse plugins
for quarkus
15:34 -!- sfikes [~sfikes at c-69-246-28-50.hsd1.ms.comcast.net] has joined
#hibernate-dev
15:34 < koentsje> i have tried to make this ready for publication
15:35 < koentsje> making it ready from a legal point of view by adding
license and copyright info
15:35 < koentsje> creating a readme, build and installation guides and a
getting started guide
15:36 < koentsje> that’s it for the past sprint
15:36 < gsmet> #topic Next 2 weeks Koen
15:36 < koentsje> i need to setup continuous integration for quarkus-eclipse
15:37 < koentsje> finish the contribution guide which i started
15:37 < koentsje> and finally make the whole repository public
15:38 < gsmet> out of curiosity, what does it do?
15:38 < koentsje> further i want to create a quarkus-eclipse launch
configuration to run and debug inside eclipse
15:38 < koentsje> i still need to develop the quarkus getting started cheat
sheet (and maybe others)
15:39 < koentsje> gsmet, it helps users get started with quarkus
15:39 < gsmet> sure :)
15:39 < gsmet> but it allows to create a Quarkus project?
15:39 < gsmet> with the bom setup and everything?
15:40 < koentsje> i.e. create a project, install an extension, provide a
view of available extensions
15:40 < gsmet> OK, cool
15:40 < koentsje> yes with bom etc
15:40 < koentsje> i desperately need to compile a wish list of
functionality that would help users
15:41 < koentsje> so that’s one of the tasks
15:41 < gsmet> just send an email on quarkus-dev and see how it goes
15:41 < koentsje> finally i also want to continue some of the cleanup and
restructuring work on the hibernate eclipse tooling that is still going on
15:42 < koentsje> gsmet, yes that’s the plan
15:42 < koentsje> that’s it for me
15:42 < gsmet> ok, thanks
15:42 < gsmet> #topic Progress Guillaume
15:42 < gsmet> not much from me
15:42 < gsmet> I worked a bit on HV but mostly worked on Quarkus
15:43 < gsmet> I'm making very slow progress on the Search extension being
sidetracked by a lot of things
15:43 < gsmet> but I'm getting to it
15:43 < gsmet> I have an SSL related issue to fix and some cleanup to do on
the ORM side (I made it work but it could use some cleanup)
15:43 < gsmet> hopefully, I'll be able to push it this week
15:44 < gsmet> #topic Next 2 weeks Guillaume
15:44 < gsmet> I need to spend some time on HV, we have a few PRs waiting
for my attention
15:44 < gsmet> and pushing the Search Quarkus extension
15:45 < gsmet> I also would like to integrate HV and ORM on Quarkus as they
are not integrated yet
15:45 < gsmet> that's pretty much it
15:45 < gsmet> #topic Progress Sanne
15:45 < gsmet> sannegrinovero: anything on your side?
15:46 < gsmet> apparently not :)
15:46 < koentsje> :D
15:46 < gsmet> let's close this meeting then, thanks everyone