Roadmaps for 4.0.0 and 5.0.0 #300
Comments
|
Hi, just curious, is their any timeline when this would be released? |
|
on a similar note, how would we grade the stability of current latest (4.0-alpha)? should it not be considered production-ready? if so, what is work is outstanding to reach stable/production-ready? |
|
Hey folks, sorry for the delayed response (I was on family vacations 😄). The plan for v4.0 is essentially this:
The first 2 items are the ones which would modify the API, apart from that everything is pretty much stable. Although I do use the alpha version in production, I'd suggest you wait until those items are done to avoid having to modify things on your apps (or you use it knowing that things will change). I don't have a timeline (sorry) but my hope is to tackle these things in the upcoming months (after I release a new version of @chimeraphp packages). |
Which features are you looking to use from PHP 7.4? |
|
Just bumping the min version to current stable is sufficient 👍 Besides that, typed properties, obviously |
|
Are you planning to change from the BSD 3 license to the MIT license after version 4? |
|
@ZhangChengLin we can surely change, AFAIK BSD-3 is as permissive as MIT so I didn't bother to change back then. However, I'm not an expert in that area and if that's an issue for you I can definitely modify it. |
I do n’t know much about this either. I just see that many projects are using the MIT license, so I feel that if they are the same, then maybe you can introduce more about the idea of using this license in the readme.5.5k star, Someone can give you the answer you want. I think that the current license has no effect on me. |
|
BSD/MIT are pretty much the same: changing license is much more effort
(need to hunt down all previous contributors and ask for their opinion)
…On Sat, May 30, 2020, 20:52 张成林 ***@***.***> wrote:
@ZhangChengLin <https://github.com/zhangchenglin> we can surely change,
AFAIK BSD-3 is as permissive as MIT so I didn't bother to change back then.
However, I'm not an expert in that area and if that's an issue for you I
can definitely modify it.
I do n’t know much about this either. I just see that many projects are
using the MIT license, so I feel that if they are the same, then maybe you
can introduce more about the idea of using this license in the readme.
Someone can give you the answer you want. I think that the current license
has no effect on me.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#300 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AABFVEGAQ3URYPEPU3DDK33RUFIXZANCNFSM4HYZ36JQ>
.
|
|
RE PHP version stats from composer, only 1 in 5 people are using PHP 7.4. PHP 7.2 is still the most popular version... |
I don't understand this, so it's so troublesome? |
|
Yes, **changing** license is a lot of work
…On Sat, May 30, 2020, 21:58 张成林 ***@***.***> wrote:
BSD/MIT are pretty much the same: changing license is much more effort
(need to hunt down all previous contributors and ask for their opinion)
I don't understand this, so it's so troublesome?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#300 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AABFVEECWHQCUKZEE4Q3JPDRUFQPJANCNFSM4HYZ36JQ>
.
|
It all your properties are private, typing them in the language is only gonna give you a slowdown at runtime for no benefits, compared to using a static analyzer and phpdoc? |
|
It will add runtime invariants, should any initialisation escape our analysis. In general, the overhead is negligible, compared to the advantages: if you are worried about this kind of overhead, you can write a plugin to drop all typed properties post installation. |
|
Besides, adding types explicitly can aid compilers and reflectors 👍 |
That would be a bad idea, if someone is reflecting on the code. ;) |
|
I'm getting the feeling that the PHP7.4 yay/nay discussion might be better served elsewhere.... Regardless, I'd be very interested in having JWK support, to the point of adding it myself but for PHP 7.2 and higher, as that is indeed the constraint for the libs I am currently integrating with... |
|
Per the PHP 7 question I agree with @Potherca this is a bit off-topic here and also, this ship has sailed months ago. The README was updated last week. I am a co-maintainer of the Simple OAuth/OpenID Connect module for Drupal and have an interest in #32, per above (so will also take a look at #339, which is indicated as a blocker.) |
|
I would really appreciate the release of a stable version @lcobucci If you like, I could create a PR where I make version |
|
@chris-doehring my plan is to release v4 before php 8.0 GA - or at least really close to it. I want for v3.x to only serve as migration path for the new API, the only tricky thing is to backport BC-breaking changes. |
|
@lcobucci is there anything I can do to help out? Currently this is blocking PHP 8 support on league/oauth2-server and laravel/passport. Let me know what can be done to speed this up :) |
|
@lcobucci so If I go by #300 (comment) it seems #452, #339 & " Backport changes to v3.x in a BC-compatible manner to create a migration path" are the only three things left for v4? #339 is a bit out of date it seems but do you have any more feedback for those PRs? Not sure what you mean with the backport of changes to v3? |
|
Okay, folks, I took some time to organise stuff - about time, right? (sorry). So I have some dates which I'm aiming for v4.0.0-beta1 and v3.4.0 (assigned on the milestones) and added issues to backport stuff. v4.0.0-beta1 issues: https://github.com/lcobucci/jwt/milestones/4.0.0-beta1 With that said, I'll be closing this issue and invest some time reaching those expected dates. Any help on getting stuff done is really appreciated! |
|
That's great @lcobucci! I'll check those out soon to see if I can help out there 👍 |
I'd be interested in helping to stabilize the v4 code APIs. Do you have a roadmap on what is still needing to be completed?
The text was updated successfully, but these errors were encountered: