Is this library dead? And whats the alternative? #12111
Replies: 4 comments 12 replies
-
|
Related current issues: |
Beta Was this translation helpful? Give feedback.
-
|
Hi @Martinspire, I'm almost ok apart from the spare time on open source. What I can say - it's open source, any contribution is welcome, and anyone can contribute. I'm happy to review and approve, and if someone can dedicate time to the library, more than happy to set the person as main maintainer with all permissions. May things are complex, and I wish everyone can find motivation to solve complexities. |
Beta Was this translation helpful? Give feedback.
-
|
And instead of creating discussions - pick up a bug and try to solve it. v20 is failing because of a very simple test assertion, because a part of generated HTML has been removed. Super simple fix. This is how you and anyone can start: https://github.com/help-me-mom/ng-mocks/blob/master/CONTRIBUTING.md |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
|
I don't think it's unfair to create this discussion considering some PRs have been raised to fix bugs and have failed to get much attention nor merged after many months. Whilst the reasons for this are understandable of course, it's a fair question/discussion point. Hopefully someone can pick this library up as a maintainer to increase the cadence of PR reviews and merges |
Beta Was this translation helpful? Give feedback.
-
|
Nice to hear from you satanTime. I have been trying to get into this library but have failed to make any useful contribution. Perhaps others are influenced however to pick up where some issues are left at. I don't mind making contributions to move this project forward while my time is spent on other projects. Regardless of that, it now makes more sense that this library didn't get many updates recently, if you've left the Angular space. Which is totally fine. Perhaps its better to move foward and see if there might be a need to change ownership, or at least have other folks who can review and accept PRs, since we still require support for the latest features and then it wouldn't need to rely on you to fix things. Also, a different part that might help more folks is to provide more documentation on the code and the way things are, since its very hard to get into the current codebase if you want to do something, but are unfamiliar with its context and reasons why things are the way they are. Do you have any idea what it would take to make this project up to standards for the latest Angular versions? Because that is the most pressing issue right now. I can handle workarounds, its just that many of the main features of v19 and 20 aren't supported yet, with more on the way with 21. Especially when 22 hits with Signal Forms that will be part of 21 as experimental phase. Are you still interested in supporting this project or are you kind of done with it now and looking to hand this over to other people? Because that makes a difference in how we move forward. |
Beta Was this translation helpful? Give feedback.
-
|
Would it be possible to reduce the complexity of the codebase by letting go of the wider range of supported version? I think most would be fine if only actively supported angular versions were supported and maybe even drop non-ivy and non-esbuild support? |
Beta Was this translation helpful? Give feedback.
-
|
Seeing how Angular is transforming, zoneless, standalone, signals, etc. I think it would make sense to focus on that and indeed drop most of the other stuff and only provide minor support for older versions. Regardless, for now it doesn't look like there is much uptick in activity if nobody is willing to dive into the codebase and pick this library up. |
Beta Was this translation helpful? Give feedback.
-
|
As someone who actively uses and loves this library, dropping old versions support is not a problem to me |
Beta Was this translation helpful? Give feedback.
-
|
this is indeed a good idea |
Beta Was this translation helpful? Give feedback.
-
|
I'm locally checking wether dropping the angular versions and fixing the todo's mentioned with Angular is viable. If i've got it up and running i'm willing to open a PR. So far i've touched 261 files.. |
Beta Was this translation helpful? Give feedback.
-
|
Just to give some perspective, I've removed the older versions in tests and implemented todo's and deprecations referencing old angular versions. I've still got 18 failing unit tests that I still need to fix. As I'm only using small parts of ng-mocks in my own application there is a learning curve on the parts that I don't use. |
Beta Was this translation helpful? Give feedback.
-
|
These issues seems to be pile up since now in Angular V21 Vitest is new default test runner. If I'm not mistaken vitest is not officially supported by ng-mocks. I found this conversation #5064. 😐 |
Beta Was this translation helpful? Give feedback.
-
|
I have no issues using the AnalogJS vitest test runner. Not tried the official one but I want to wait for a bit for that to be more stable. Though I don't expect there to be a problem? Though you might want to make a new issue for that. I also bet that angular signal forms are going to cause issues with this library. Which is probably the moment for me to look elsewhere if that happens. |
Beta Was this translation helpful? Give feedback.
-
|
I spent a few hours testing how it works. I quickly came to the same conclusion. It's best to wait until it’s more stable and the performance issues are fixed 😐 I'll recheck the test cases once the implementation is more mature and create new issues if needed 👍 I just started testing Signal Forms. The implementation looked pretty safe — it mostly uses computed, signal, and WritableSignal, and I haven’t had any issues with those when using |
Beta Was this translation helpful? Give feedback.
-
|
At least some of the problems I encountered seem to point to Angular components + jsdom |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
So I've been enjoying this library for a lot of years now and appreciate the work that has been done in it. I applaud @satanTime for his efforts and can totally understand how a library like this can eat your time away, or adding frustration to a busy day already.
I hope to be using it for longer, but I think I'm at a crossroad now where I just have little confidence that the project is going to survive for long. When standalone hit, we already noticed a few problems related to how things were mocked and with newer signals its clear that this project needs a few big updates to support the latest features. Let alone when Signal Forms drop.
However, the amount of updates the project got were low and it seems there isn't enough documentation or perhaps even skill for other folks to pick up where satanTime left off. I think there are some smart folks out there that could help the project but I'm not sure if its really possible to be effective, given that the project uses a lot of fixes and workarounds to make it work with a wide range of codebases (which I totally get). I'm not sure how much time satanTime is still willing to spend on it and whether some other problems behind the scenes have caused his time to shrink to almost zero to work on fixes the community is waiting for. I have no problem providing donations to keep work going, but I need to see a bit more effort before I start donating some more. Though anybody else willing to donate some, is always appreciated. Its just not clear where we stand right now and what the possibilities are.
Another problem I noticed, is that there isn't really an alternative right now to mocking these components. I see that a lot of the Angular team have moved to no longer mocking stuff (which I find really weird, defeating the purpose of unit tests and making it difficult to spot where bugs are coming from). I would applaud anybody that would fork this project (to make it work with the latest features of angular 20 and soon to be 21), or to build your own library. I would love it if the Angular team itself would put effort in a mocking solution for the platform, or even take over this project, but I can also understand why that would not provide any short term solutions either. If any at all.
In this discussion I hope to find some answers on the following:
I would love to keep using NgMocks and I've been using a few workarounds to at least get my components tested easily. I stopped using most of the new signals for viewchildren, contentchildren, etc. Only really using input/output/model/linkedSignal/computed/effect, which seems to work still. But no input.required either. And sometimes making custom mocks for when its just not possible to work around sensibly. I'm still on the fence about the future. I don't really like to nót mock my dependencies, but I can also see how much work that would be. Manually making mocks for each component seems very tedious and I'd rather not create duplicates that can cause all other kinds of failures. But without a maintained testing library, I see no other way to keep using it, since its essential to my pipeline and would block future angular updates entirely.
Beta Was this translation helpful? Give feedback.
All reactions