Is Echo still maintained ?

[adrianlungu]

Hello,

I wanna start out by mentioning that I'm a really big fan of Echo and have enjoyed working with it since v2.

Recently, while implementing another project on Echo, I've run into #1356 which really breaks flows when using Bind, and also saw a couple of other Pull Requests that are marked won't fix although they do provide value for the router imho (i.e. #1527 ).

Out of concern, I'm just asking if Echo is still considered maintained, or if it might be a better idea to fork it and maintain it separately while also keeping track of it in case it ends up being maintained at some later point ?

Thanks!

[iambenkay]

This is also a major concern for me. Progress seems to have halted and issues are piling up... If you do fork it @adrianlungu I would love to contribute 🚀

[lammel]

@adrianlungu Yes, Echo can still be considered as maintained, as long as people are standing up and contributing.
But currently real-life is taking away much of the few maintainers time. It is also not always easy to review and merge a PR as all implications need to be thought through and possible tested.

It is a free and open source project after all, with all ups and downs.A fork is always a possibility, but remember also a fork needs to be maintained. So I think it does make sense to resolve the perceived staleness right for this project.

@iambenkay Your contributions currently seem to be limited to a single "Why is this not merged yet" question yet, of course your contributions are welcome here too. I for one started out with router enhancements some months ago and try to help where I can.

We (current maintainers) are discussing how to proceed and if/how new maintainers can join in.
If you are willing to help, let us know ;-)

[iambenkay]

sure I am willing to help.

been looking at echo for 3 weeks now... it's an amazing library.

[iambenkay]

I just didn't want to make any PRs and have them unattended to considering the other great PRs I saw that have not been merged.

If it's going to be more active then I would make contributions.

[adrianlungu]

@lammel thanks for the update!

All points you mention make sense and I agree, all PRs and updates do need to be carefully checked in order to ensure the stability of the router and backwards compatibility; the question popped into my head considering for example the bind issue which is over a year old, and it also impacted updating from 4.1.6 to any version post that one in a quite significant way.

I am open to helping out with reviews, PRs and anything else that might be needed to keep the router in good shape, as much as my time permits, and I think allowing people to join in as maintainers would be really good to speed things up.

I would definitely be in favor of helping the current repo instead of forking, so, please let me know how you think it would be best moving forward!

[lammel]

Great guys. Thanks for your feedback.

I stumbled upon echo in the same way and work based on the "if I have time" pragma anyway.
Guess the whole OSS community is built upon that.

There are also other quite active recent contributors that might be willing to help like @pafuent or @arun0009.

[aldas]

Another cosmetic thing. maybe that stale bot should mark stale issues some other label as won't fix. won't fix has that negative vibe that your problem is disregarded as irrelevant.

[lammel]

Another cosmetic thing. maybe that stale bot should mark stale issues some other label as won't fix. won't fix has that negative vibe that your problem is disregarded as irrelevant.

We might use a dedicated "autoclose" label, which does show the intention. We'll take that into consideration,

[lammel]

Just for info... the stale bot was just relaed a little and uses the 'stale' label now instead of wontfix.
We also have 30 days now before the issue is closed.

[lammel]

Great to see so many contributions from you @aldas and @iambenkay and @arun0009 . In the meantime @pafuent has joined as a maintainer now and we are back on track. Please keep it coming.

Concerning the bind issue which was one of the reasons discussion mostly continues in #1681 now. @adrianlungu please also keep an eye on it, as we try to come up with a way to make things just work with v4 now and keep further improvements for v5 (which will change the interface and a new PR will be opened later).

Feel free to close this one, otherwise we can let our lovely relaxed stale-bot handle it.

[lammel]

Thanks to the work of numerous contributors, we even have a new release!
So the question is answered ;-)

[jamelt]

How can you not allow someone to bind JSON Arrays for months on end?

I would turn the stale bot off completely, IMHO.

[aldas]

@jamelt answer for your problem is here #1565 (comment) also binding is detailed in documentation https://echo.labstack.com/guide/binding/ which mentions methods you should use for binding body.

[jamelt]

@aldas thanks for the reply, i respect that.

this patch shared by @benoitmasson was all that was really needed to fix this. i literally copied and pasted the entire bind.go file and added these two lines to the file. set the e.Binder to the patched code when bootstrapping the app and everything was taken care of.

i guess I would just expect that something as rudimentary as binding a json array would be supported out of the box. an array is valid json.

the closer a feature is to core functionality, the greater expectation is that it will work and be good at it. as you start to veer off into niche features, that expectation tends to subside.

i am more than willing to submit a PR if it is going to be accepted into master.