Replies: 5 comments 6 replies
-
|
Can only answer as a random contributor as I don't have any other access to the project
Seems stable enough but also a work in progress. This wasmjs PR appears to be the most requested feature. It needs to be merged for people interested in trying it out and giving feedback Improve health with more frequent releases of fixes/features (that don't break api). Assume this is the process https://github.com/sqldelight/sqldelight/blob/master/RELEASING.md.
Don't know - contributors can do the development, however PRs need reviewing/merging from maintainers
Would be supported for as long as CashApp depend on it internally
The only manual task is creating the change-log and that can be made easier using GitHub cli to list the merged PRs e.g |
Beta Was this translation helpful? Give feedback.
-
|
As expected, you @griffio seems to be the only one responding to issues and discussions :) Who are the current maintainers of this project? The last commit of the top committers @AlecKazakova and @JakeWharton is from 3 months ago. It just seems really quiet around this project. We as a team/company would like to know if we can rely on this project. For us it's the best SQL lib out there. |
Beta Was this translation helpful? Give feedback.
-
|
What about QoL fixes e.g. a release of sqldelight/sql-psi#623? Is there anything blocking a 2.0.3 release? |
Beta Was this translation helpful? Give feedback.
-
|
That message is merely annoying but otherwise harmless. I would not do a release for that. Alec is out for a bit. I wouldn't count on a release until he gets back. |
Beta Was this translation helpful? Give feedback.
-
|
If Alec is kind of the single point of failure, could this project be managed differently such that there will be a constant release cycle? I'm worried that people might back out of this library because it appears not so well maintained. |
Beta Was this translation helpful? Give feedback.
-
|
Unless otherwise specified as being part of an official product, our corporate open source projects are maintained on a volunteer basis. Their goal is not to amass users. Whether people use the software or not is generally immaterial to us. We use it, and it being open source is a way of showing our work. People only tend to notice this fact (which is true for the overwhelming majority of our projects) when they have problems. If you are solely a JDBC Sqlite or Android Sqlite user of SQL Delight you are probably unaware of the pains of a PostgreSQL user or someone who wants WasmJs support. |
Beta Was this translation helpful? Give feedback.
-
|
Great, this is a clear statement. If you already know that people only tend to notice this fact when they have problems, it might help them and the people working on this project to make this also more clear on the main page of this project. When I read the main page for the first time I was under the impression that PostgresSQL support was in line with SQLlite. It's totally fine that this is not, and we thank you and and the devs making this great project. But a line on the main page like this is a showcase project, no guarantee or
will help to avoid some misunderstanding. For Kotlin backend production use, SQLdelight may not be safest choice right now. If there are people like @griffio who are willing to make Backend/PostgreSQL more stable I'm also willing to chime in. Amazing project ❤️ |
Beta Was this translation helpful? Give feedback.
-
|
It might help to get rid of the @square-clippy issues that are created. If ppl come to this project for the first time and see the amount of issues open they might run away. https://github.com/sqldelight/sqldelight/issues?q=is%3Aissue%20state%3Aopen%20author%3Asquare-clippy |
Beta Was this translation helpful? Give feedback.
-
|
The "unmaintained" problem is even worse for the IntelliJ plugin. To a point that it's rating is now dropping rapidly. The plugin simply does not work on recent InteliJ versions. I consider the plugin a fundamental part of the project: without it SQLDelight's DX is not that good. |
Beta Was this translation helpful? Give feedback.
-
|
I had a baby thats why i've been awol. The maintainer pool is better now than it was a couple years ago with @hfhbd and @griffio helping but I'm still extremely load bearing especially for releases (I hope to fix this next chunk of time I have when I don't need to prep a release). Internal usage has also jumped at Block with server side support and so the time I do have for SQLDelight has mostly been spent helping folks get it running internally (as Jake mentioned that's really the priority) And also echoing what Jake said, most people use this solely on Android which has been pretty stable and unchanging for awhile. Everything still works fine in Cash where I'm primarily exposed to problems with it, this library was way more active when we were noticing problems in our day to day and fixing them. It's probably next most used in multiplatform setups. We used it a bit years ago but I'm not exposed to that part of the library as a consumer anymore. Jake and Jesse are so sometimes they fix things that cause them problems. Nowadays I'm most exposed to the MySQL dialect since thats what we use for Cash server, so I work on bugfixes for that whenever someone internally encounters a bug Some folks at faire spent a bunch of time and contributions getting cockroachdb working Derek wrote the web stuff because he was authoring a web site that he wanted to use sqldelight for Everything follows the same trend, contributors are contributing based off of what they're using. Becoming a contributor isn't easy, this library has many rabbit holes, but when there's long stretches with nothing happening its likely because there aren't any contributors facing the same problems. I'd love to just have someone hired at Block working on this full time so we can change that to be a bit more community driven instead of contributor driven, but I don't know when/if that will happen, and we've done well in a contributor driven model up to know so I chose to invest there. Opening up dialects and drivers to be externally implemented is part of that, moving this into a GH org will eventually mean the dialects and drivers move out of this repo so its more obvious that "sqldelight" (the core library) is stable and rarely releases, but something like the postgresql dialect or wasm driver which are way newer may have 100s of issues and more actively released. |
Beta Was this translation helpful? Give feedback.
-
SQLdelight is arguably the best SQL lib out there. We heavily rely on it for Kotlin backends, and are very happy with it.
However, there hasn’t been a new release for a long while now. The latest commit to
masterwas last year October. The only one responding to issues seems to be @griffio . @square-clippy is polluting the issues.Question: what is the current health of this project? Do Cashapp developers still get time to develop? Can other devs rely on this lib to be supported in the coming years? Do you guys need help?
Beta Was this translation helpful? Give feedback.
All reactions