Support workspaces and singlefile recents

[Sharsie]

Moved over from albertlauncher/python#225

[Sharsie]

Since this never got reviewed (and hence merged), I'm lagging behind the plugin interface version, I currently have 0.33 installed and need to verify the changes work with changes introduced with interface version v5, specifically running the search when query is empty, i.e. using matcher = Matcher(ctx.query) when query has nothing but trigger.

I'll try to get to this during the week, until then I consider this PR not ready

[Sharsie]

Tested against the v5 interface, lgtm, I'm not sure what your expectation for reviewing this is @ManuelSchneid3r though
@Silureth if you've updated past 0.34, feel free to use this branch instead of the old one, if you are still on interface version 4, you can use one here:
https://github.com/Sharsie/albert-plugin-python-vscode-projects/tree/hotfix/vscode_projects_recents_interface_v4
If you are still on interface v3, you can ignore me

[ManuelSchneid3r]

I'm not sure what your expectation for reviewing this is @ManuelSchneid3r

I dont get the point of this question. I try to achieve peer reviewing of the plugins (well having volunteers is a problem though). I mean this is code that runs on the users machines and probably close to nobody reviews the code before checking the enable plugin checkbox (see more here). I am not exactly sure if this answers your question.

[Sharsie]

Sorry I should have been more specific, as you mentioned, getting the plugins reviewed is a strugle.

If somebody opened a pull request against this repository, as a maintainer, I'm more than happy to review it, even though I have very limited python experience, but I can probably do that for the code I wrote.

But what is the expectation when the maintainer (in this case me) opens a pull request and there is noone to review it? I mean, this has been sitting here for more than 6 months, I understand we don't want to merge unverified changes, but at the same time, there is no point in me maintaining the plugin if the changes never get merged.

Am I alone in this? Or does this affect other plugins as well?

If that would be the case, maybe a different approach to 3rd-party plugins would be more appropriate, e.g. docs linking to 3rd-party repositories that can be used by downloading the code into the plugins directory locally. Clearly stating that the code is unverified and used at user's own risk.

[ManuelSchneid3r]

The purpose of this special case is that this is code that eventually runs on my machine (and other devs working on main). I dont want to give anyone the possibility to inject unreviewed code that does so. I hope you agree with that.

Now the reviewing needs to be more progressive. That's why there is the CODEOWNSERS file which contains maintainers and the reviewer team. But I just added them recently so there was nobody assigned to this PR yet.

Since in this case I am the only one that responsible for a review I have to admit that I overlooked it in the past. There are too much of other issues in my queue. That's why I try to setup this contribution pipelines with auto assignments. So hopefully in future there will be faster reviews. For some other plugins this seems to begin to work out.

There is the readme of the org repo containing third party plugins that have not yet gotten enough attention to be merged into the public plugins. And if nobody cares about them they will not land there, because we actually need some pioneers that go ahead and review/maintain them for the non tech audience.

I the end this has to be communicated properly. So I continously work on the contributing page and plan to put a community primer in the plugin settings placeholder.

[ManuelSchneid3r]

for future PRs please do not squash and use conventional commit messages since the release script auto bumps version and builds changelogs from it. thank you for your work and patience

[Sharsie]

Oh yea sorry about that, this PR is a copy from the old repo, before contributing docs were a thing, I'll make sure to familiarize myself with the rules.

The purpose of this special case is that this is code that eventually runs on my machine (and other devs working on main). I dont want to give anyone the possibility to inject unreviewed code that does so. I hope you agree with that.

Absolutely

But I just added them recently so there was nobody assigned to this PR yet.

Oh I was actually referring to the fact that there was a request for the review from the python reviewers group very soon after this PR was opened:

I wouldn't expect you to do the reviews, that would be insane :)

Anyway thanks for paying attention and expending time to handle all this, I appreciate it. I will leave you alone now.

[ManuelSchneid3r]

Oh I was actually referring to the fact that there was a request for the review from the python reviewers group very soon after this PR was opened:

im not sure if the members got pinged. with codeowners particular random members are assigned.

Anyway thanks for paying attention and expending time to handle all this, I appreciate it. I will leave you alone now.

all good. thank you for work on this plugin.