Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Revert "Revert "qutebrowser: 2.5.4 -> 3.0.0"" #251668

Merged
merged 1 commit into from
Aug 26, 2023

Conversation

dotlambda
Copy link
Member

@dotlambda dotlambda commented Aug 26, 2023

Reverts #251660
A downgrade would lead to data loss.

ghost
ghost previously requested changes Aug 26, 2023
Copy link

@ghost ghost left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ghost
Copy link

ghost commented Aug 26, 2023

See also: #251660 (comment)

@ghost
Copy link

ghost commented Aug 26, 2023

A downgrade would lead to data loss.

I don't understand what you mean by this.

@dotlambda
Copy link
Member Author

Downgrading from pyqt6-webengine to pyqtwebengine leads to data loss afaik. We can't afford to ship the downgrade to nixos-unstable. Let's discuss properly taking doing any action.

@dotlambda
Copy link
Member Author

A downgrade would lead to data loss.

I don't understand what you mean by this.

Qt5's webengine can't deal with the SQLite database after migrations were applied

@dotlambda
Copy link
Member Author

Downgrading from pyqt6-webengine to pyqtwebengine leads to data loss afaik. We can't afford to ship the downgrade to nixos-unstable. Let's discuss properly taking doing any action.

Hence I'm merging this now and ask you to wait for comments before making changes, just like @rnhmjoj should have waited before merging #250171.

@dotlambda dotlambda dismissed ghost ’s stale review August 26, 2023 20:58

We have to use Qt6 by default now that some people's databases were migrated over.

@dotlambda dotlambda merged commit 9b81404 into master Aug 26, 2023
4 of 5 checks passed
@dotlambda dotlambda deleted the revert-251660-pr/revert/250171 branch August 26, 2023 20:58
@ghost
Copy link

ghost commented Aug 26, 2023

Downgrading from pyqt6-webengine to pyqtwebengine leads to data loss afaik.

No

If you end up accidentally downgrading from running Qt 6 to Qt 5,
the underlying Chromium will discard your browsing data (such as
cookies). qutebrowser will warn about this while starting

You get a warning, and even if you ignore it it's just the cookies.

Hence I'm merging this now and ask you to wait for comments before making changes, just like @rnhmjoj should have waited before merging #250171.

Rules for thee but not for me....

@dotlambda
Copy link
Member Author

dotlambda commented Aug 26, 2023

Hence I'm merging this now and ask you to wait for comments before making changes, just like @rnhmjoj should have waited before merging #250171.

Rules for thee but not for me....

Not true. I'm only undoing something that was done just a second ago so we can have a discussion before (potentially) making that change again.

If you end up accidentally downgrading from running Qt 6 to Qt 5,
the underlying Chromium will discard your browsing data (such as
cookies). qutebrowser will warn about this while starting

You get a warning, and even if you ignore it it's just the cookies.

Exactly, data loss. If we want the warning to be effective at all we'd at least have to tell people about the existence of qutebrowser-qt6 in it.

@ghost
Copy link

ghost commented Aug 26, 2023

Exactly, data loss.

What next, we remove rm from coreutils because it can be used for rm -rf?

If anything, the silent data loss occured when #250171 silently caused an irreversible upgrade to peoples' cookies. The longer we leave that on master the more people will experience this. By stalling you are causing more of this "cookie loss".

Frankly it smells like stalling as a negotiating tactic while claiming that time is on your side. That's pretty dishonest.

If this is the big deal that you claim it is why didn't #250171 even touch the release notes?

@dotlambda
Copy link
Member Author

dotlambda commented Aug 26, 2023

If anything, the silent data loss occured when #250171 silently caused an irreversible upgrade to peoples' cookies. The longer we leave that on master the more people will experience this.

The sad news is that #250171 has already made it to nixos-unstable. I think it shouldn't have been merged this fast but we can't do anything about that now.
I'm not saying that having data loss for some people is the worst tradeoff but we should discuss it before taking action.
I'd like to hear your opinion on #251671 for example.

Frankly it smells like stalling as a negotiating tactic while claiming that time is on your side. That's pretty dishonest.

It's not. I just want to have a discussion first.

@ghost
Copy link

ghost commented Aug 26, 2023

we can't do anything about that now.

That is a really sleazy attitude.

It's not. I just want to have a discussion first.

No you don't! You've declared that nothing can be done.

You're stalling to deliberately increase the number of people who've been silently upgraded.

@dotlambda
Copy link
Member Author

we can't do anything about that now.

That is a really sleazy attitude.

I don't mean that we can't revert the change. I just mean that some people will inevitably end up with a downgrade if we do.

@dotlambda
Copy link
Member Author

You've declared that nothing can be done.

Nothing can be done about a Qt6 version having made it to nixos-unstable. Stop misinterpreting me on purpose.

@dotlambda
Copy link
Member Author

These accusations don't get us further. I'd rather discuss options. Maybe Matrix is the best place for that?

@samueldr
Copy link
Member

If anything, the silent data loss occured when #250171 silently caused an irreversible upgrade to peoples' cookies. The longer we leave that on master the more people will experience this. By stalling you are causing more of this "cookie loss".

Schema migrations are to be expected when upgrading packages. Assuming this is a package upgrade, this is a schema migration, and not data loss.

Reverting schema migrations / making them forward+backward compatible is left as an issue for upstream to solve, we can't handle every schema out there.

Upstream made this the current version, maintainers upgraded, schemas got migrated. No data loss.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants