Inactivity?

[nyc2]

Hi Querydsl team,

I just wonder about the great inactivity realized when looking at Github commits and the number of open issues.

I have used Querydsl since years and was always excited about quick answers and best support mainly provided by Timo Westkämper. Thank you very much.

It would be a pity if there are no concrete plans on how to proceed with this project. Do you have any information that could be relevant for the community users?

Bye, nyc2

[Ruben Dijkstra]

Hello,

Querydsl is not dead, but a few of the members have been busy for quite a while. (Myself included)

The future is not quite certain, but at least what is certain is that I will take a more prominent role in the project.

Bear with us, Querydsl will survive.

Best regards,
Ruben

[Filipe Sousa]

Now I’m worried about the uncertainty.

> --
> You received this message because you are subscribed to the Google Groups "Querydsl" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to querydsl+u...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Benjamin Ross]

Sorry for spamming a bit, but I seem to be unable to post a new thread to this group and this seems to be the most appropriate thread to ask what's going on.

Thanks.

[Ruben Dijkstra]

Hi Benjamin,

That's fine, the forum is moderated so we had to still accept your message.

Best regards,
Ruben

[Richard Vítek]

Hi Ruben,

would you have any update on the topic? I'd love to see QueryDSL survive and thrive again.

To be more specific, are there actually any plans for at least maintenance releases any time soon?

Thanks and good luck,
Richard

[Ruben Dijkstra]

Hi,

Of course!

On Friday, 5 May 2017 17:32:05 UTC+2, Richard Vítek wrote:

Hi Ruben,

would you have any update on the topic? I'd love to see QueryDSL survive and thrive again.

To be more specific, are there actually any plans for at least maintenance releases any time soon?

Looking at everything now, the other Querydsl team members are having some life issues.

Nothing severe, but just important enough to wait with hobbies.

Currently, Timo is the only one who has a working release setup.

I'll try and see if I can also get it working.

 

Thanks and good luck,
Richard

Thank you, Querydsl is not dead, but thanks for your patience!

Best regards,

Ruben 

[Kara Hatherly]

Hi, I've been meaning to reach out for a few weeks now, stumbled over this thread and thought it was the right place to post. 

We make heavy use of querydsl-sql within Atlassian as part of quite a few of our products - in particular in JIRA where we use it to dynamically construct some pretty gnarly SQL from our search DSL. It's a great project and we're committed to, and invested in its long term success.

I've got two PRs which have been sitting ready to merge for a few months now and on reading this thread feel like I (and Atlassian in general) could be more involved in the project given that we're making such heavy use of it and that it sounds like it could use some extra help.

@Ruben - would you be interested in bringing on an extra maintainer or two? I imagine I or one of my colleagues could get some of the ready-to-merge bugfix/small-improvements PRs merged in and figure out how to get a release out if you want the help. We've also got a handful of useful utilities and extensions we're using internally which would be good to push upstream too.

Longer-term: we're hitting some of the limits of the vistor-based recursion in querydsl-sql (read: StackOverflowErrors) and would be interested in eventually taking on a bigger project of restructuring the implementation quite a bit, but that's probably a conversation worth having in a different thread. 

Kara

[Ruben Dijkstra]

Hello,

On Tuesday, 1 August 2017 09:19:35 UTC+2, Kara Hatherly wrote:

Hi, I've been meaning to reach out for a few weeks now, stumbled over this thread and thought it was the right place to post. 

Yes, very understandable.

Sorry for the downtime, I couldn't get a hold of the other members yet..

I was telling myself, if this continues I'll just merge, release and everything on my own for a while.

We make heavy use of querydsl-sql within Atlassian as part of quite a few of our products - in particular in JIRA where we use it to dynamically construct some pretty gnarly SQL from our search DSL. It's a great project and we're committed to, and invested in its long term success.

I've got two PRs which have been sitting ready to merge for a few months now and on reading this thread feel like I (and Atlassian in general) could be more involved in the project given that we're making such heavy use of it and that it sounds like it could use some extra help.

@Ruben - would you be interested in bringing on an extra maintainer or two? I imagine I or one of my colleagues could get some of the ready-to-merge bugfix/small-improvements PRs merged in and figure out how to get a release out if you want the help. We've also got a handful of useful utilities and extensions we're using internally which would be good to push upstream too.

I would not be opposed, I'll try one more time to ask the others what they think, primarily Timo, who has ownership on the GitHub organization

Longer-term: we're hitting some of the limits of the vistor-based recursion in querydsl-sql (read: StackOverflowErrors) and would be interested in eventually taking on a bigger project of restructuring the implementation quite a bit, but that's probably a conversation worth having in a different thread. 

Yes, when the time comes we can discuss that, if it's worthwhile I'm definitely interested.

Best regards,

Ruben 

[David Berkman]

Still no activity? When is the right time to restart this project, start merging PRs, making some medium term plans, and implementing? I need to make some decisions of which tool to use in the short term and I had my hopes pegged on QueryDSL, but inactivity this long is a bad sign. I'm willing to help reverse that. I guess I'm cloning and building first.

[Ruben Dijkstra]

Hello,

Yes I also find it rather weird, but I'll take matters in my own hands and see when the rest of the team returns.

Best regards,

Ruben

[Kara Hatherly]

Hey again - if there's anything I (or Atlassian more broadly) can do to help please let me know. 

You noted earlier in the thread that you don't have a working release setup - is that just a matter of getting vagrant, puppet, etc. all working? If so would it be worth getting that automated (in travis or otherwise) such that releases are a push-button operation?

Kara

[David Berkman]

I pulled the project and attempted to build, which doesn't quite work due to...

[ERROR] Failed to execute goal on project querydsl-sql: Could not resolve dependencies for project com.querydsl:querydsl-sql:jar:4.1.5.BUILD-SNAPSHOT: The following artifacts could not be resolved: com.oracle:ojdbc6:jar:12.1.0.1-atlassian-hosted, oracle:sdoapi:jar:11.2.0, org.opengeo:geodb:jar:0.7, cubrid:cubrid-jdbc:jar:9.3.1.0005: Could not find artifact com.oracle:ojdbc6:jar:12.1.0.1-atlassian-hosted in nexus (https://dev.forio.com/nexus/content/groups/public) -> [Help 1]

...and that's after sifting through mvnrepo and updating a few of the dependencies, which still don't seem to actually exist (and one of those is com.oracle:ojdbc6:jar:12.1.0.1-atlassian-hosted). I can try turning these missing dependencies into <optional>true</optional>, but I'm not sure what that would mean for the build. Ultimately, all dependencies should either be available in maven central, or we need to provide a maven repo for the extra bits and be willing to keep that up and available.

Any suggestions on the current build and missing dependencies?

David

[Ruben Dijkstra]

We should be hosting them on http://querydsl.com/maven/

ojdbc6: http://querydsl.com/maven/com/oracle/ojdbc6/

Maybe you have a settings.xml that overrides the repositories used?

Best regards,

Ruben

[Gili]

Ruben,

Unless I am misreading the tea leaves:

1. Most committers have stepped away from this project.

2. You are doing your best to support this project but even you don't seem to have enough time to dedicate to it.

3. Atlassian is offering paid employees to work on this project.

As much as I'm not a fan of having a single organization take ownership of this project (multiple committers from different companies is ideal) I suggest it's time to reach out to Atlassian and have them take a more active role in the stewardship of this project. Worse-case scenario you guys can retake control when life settles down.

I've had to abandon some of my own open-source projects over the years and unfortunately life doesn't get less busy over time :( The only thing I have time to focus on nowadays are open-source projects that are directly related to my 9-5 job. It's understandable, but it's time to move on; otherwise, QueryDSL will die as a project.

Gili

[Filipe Sousa]

This weekend, Timo was busy reviewing pull requests. Maybe there is hope 🙏

[Ruben Dijkstra]

Yes, there is hope, because I have heard what Timo's intentions are.

So, the main focus is going to be Querydsl 4, I will try to see if I can get a few last releases out for Querydsl 3, but that will have to be under com.querydsl.

It's only temporary, but it's at least to not leave Querydsl 3 in the dark, so bear with us.

Anyway, I did never and do not intend to drop the project, by any means, there even have been some Atlassian employees who offered their help, which I accepted, so there is enough manpower.

So there are still committers from different companies, and there is still time for the hobby I love.

But I do appreciate your concern, although the tea leaf analogy wasn't strictly necessary.

[gili.tzabari]

Glad to hear it. I'll save my doomsday tea leaves for another day :)

Gili

--
You received this message because you are subscribed to a topic in the Google Groups "Querydsl" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/querydsl/B5UOzCYbNcE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to querydsl+u...@googlegroups.com.

[Adam Latuszek]

Is there any ETA of new release?

[Ruben Dijkstra]

First need to address Hibernate 5.2
But the easiest would be to update Querydsl version to 5 and introduce this breaking change and keep supporting Querydsl 4, and drop Querydsl 3 after the last few releases.

[Lovro Pandzic]

I agree and suggest that you take the Querydsl 5 option. Supporting it in Querydsl 4 will be a waste of time, specially considering how much development effort the project has received in the past year.

[Robert Bain]

Hello Kara

On this, please see https://github.com/querydsl/querydsl/issues/2459. Timo has approved handing over the project and so a new owner and contributors are required.  If interested, please comment directly on the issue.