Cherry Pick Many Space Wizards Commits - #77
Merged
Merged
Conversation
Yes unironically it only happens on that culture. Also gigantic skill issue on our part here.
Fixes #222, hopefully Yes I did test this. No idea why this is only an issue for some people.
* Supply engine version CVar from launcher * Clarify comment * Clarify comment, hopefully
Right now, running the ARM64 launcher after previously having ran a x64 launcher could cause errors, as the installed engine versions would be x64. We now clear old engine versions on startup if we detect the architecture changed. This seemed like the simplest way to handle this scenario. I also thought of having the architecture in every EngineInstallation and EngineModule row, but that would've been way more complicated.
Avalonia moment. Fixes #227 Probably fixes #92
Now when the launcher gets an error during downloading, it will keep all the blobs that were successfully downloaded. This means restarting the download won't have to redownload that data. There were basically two problems that needed solving to implement this. The first problem is that the launcher wants all content blobs to be referenced, or else it'll cull them on update. This is easy enough to deal with: we just track interrupted downloads with some new DB tables and make the culling logic aware of those. The second problem is keeping state consistent. We can't practically keep the entire database consistent during the download process, so we relied solely on the SQLite transaction to undo everything if something went wrong. Given that we now want to persist the content blobs, that won't do. Now, if an error gets caught during downloads, we still commit the database but manually delete the broken data. Given that this is obviously a dangerous situation to be in, I rewrote much of the logic to avoid writing invalid data. For example, previously we wrote the content manifest directly to the DB as it was getting read in, which also meant immediately writing empty placeholder blobs. Now the blobs just get added as they're downloaded. All things considered this cleans up the code too so it's probably a good thing. At first, I wanted to make all downloads first be stored as an interrupted download, and then swap the data out to a proper content version later. I ended up not doing this however as it'd make the joinery between the delta and zip download system more complicated. It would allow us to do intermittent commits, I guess, but it's something to consider for later. Related to #187, but does not close it (ideally we'd automatically retry for transient network errors).
Yeah that was getting pretty unwieldy.
Keep it in sync with the engine, support the new library names
Avoid having to load the entire .NET Framework runtime on launcher startup. Gigantic waste. This is going to require changes to the publishing workflow... Ugh. Oh well.
Also added --debug flag to bootstrap so that console.bat can just run bootstrap instead.
Windows is broken so we can't use these until DXGI is fixed.
DEATHB4DEFEAT
added
Priority: 1-Urgent
* Make AngleBox shrink automatically Co-authored-by: Pieter-Jan Briers <pieterjan.briers+git@gmail.com> * Add icon files * Use icons * Better licensing info * Cleanup * Cleanup * Cleanup * Cleanup * Cleanup * Cleanup * Remove Favorite -> Unfavorite * Recreate SVGs from scratch (public domain), add instructions and change some UI text * Set star icon purely through XAML. * Undo ServerEntryView.xaml.cs changes entirely --------- Co-authored-by: Pieter-Jan Briers <pieterjan.briers+git@gmail.com>
Fixes #241
Also fixed a random documention commentS to use "their" instead of "his"
Fixes #250
This reverts commit 99a067b.
Instead of having to copy all the replay files into the content DB, the loader can now load the files directly out of the zip. Has some requirements, but none of those are relevant for 300 MB SS14 replays.
Setup for future work to allow us to make file access not completely buffered to hell
SDL 3.4.0
Still present in the (newer) development tab.
This reverts commit 7cfecb3. This version of SDL3 erroneously raises SDL_QUIT when closing secondary windows. This is breaking OpenDream, so revert it. It's not critical anyways as it only adds the fancy loading progress bars to bleeding-edge servers.
Nixpkgs maintainers annoyingly readded back dotnet 8 to the package, following #163 I assume this is the cause.
DEATHB4DEFEAT
marked this pull request as ready for review
March 11, 2026 02:07
Member
Author
|
I did it again with the squash augh |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
11 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
A yearish or something commits, I don't remember, hooray!
Need to double check mac support now that I have one, noting these commits I didn't add for later:
space-wizards/SS14.Launcher@76235d4
space-wizards/SS14.Launcher@e5a01fb
space-wizards/SS14.Launcher@69baa4b