-
Notifications
You must be signed in to change notification settings - Fork 496
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
Bump CurrentProtocol #567
Bump CurrentProtocol #567
Conversation
Codecov Report
@@ Coverage Diff @@
## steamclient-beta #567 +/- ##
=================================================
Coverage 22.12% 22.12%
=================================================
Files 86 86
Lines 8666 8666
Branches 714 714
=================================================
Hits 1917 1917
Misses 6624 6624
Partials 125 125
Continue to review full report at Codecov.
|
Do we know what the side-effects are? This usually indicates some kind of breaking change. |
I didn't find out any broken functionality yet, but then again I tested only logging in part for now. Would be cool if we had some unit tests for that. Do you have any idea how we could find out breaking changes? I mean, apart from me putting this into my toys and finding out what broke the hard way 😅. I'll do that the moment I'm done with basic group chat implementation. |
We can only really rely on it being an observable change. For example, AFAIK we’ve still never seen a CMsgProtobufWrapped. |
I have a theory that the protocol change might be related to new chat implementation, if the bump happened in the last beta client. I didn't check what value current stable uses. I base my theory on the fact that when you set current UI mode to chat of I mean, this is the only thing that I noticed myself. It's entirely possible that the change has nothing to do with it, but just saying what I happened to find. I tested current UI mode with both protocol versions and everything seems to be the same (you can enable beta chat on old protcol too). It doesn't look like any existing functionality changed based on that, but we won't know for sure without more people testing that - "it works for me" is not good enough. I think it'd be a good idea to put it into alpha before beta chat makes it into stable, so any potential problems could be fixed in time, with eventual revert if needed. It's better to find out about it sooner than later, especially now when we can still test things and not when beta goes stable, potentially forcing us to newer one (Valve probably keeps very low amount of backwards compatibility, if any, considering that all Steam clients have forced auto-updates). |
I think the bump happened just prior to the new beta. @xPaw do you recall? FWIW, Valve keep a fairly long trail of backwards compatibility since IIRC game servers don’t auto-update the Steam components, or at least didn’t used to. |
@yaakov-h No idea when it happened, I only saw it in the friends ui. |
Just checked the binaries, and current stable uses the bumped version too. It seems that maybe we already do already know what's in the new protocol: https://github.com/JustArchi/SteamKit/blob/75b11fbc6acbaa1b4dc033c01971979a2a17b02d/Resources/SteamLanguage/steammsg.steamd#L73-L74 😕 If you can re-target this PR against master, and everything works, I'll merge it. @psychonic do you know what |
I don't know the significance of this, but I assume it's related to the |
I'd also like to revert the change to `client_package_version` - rather not
touch that field.
…On Sun, Jun 17, 2018 at 5:26 PM, Nicholas Hastings ***@***.*** > wrote:
I don't know the significance of this, but I assume it's related to the
client_instance_id field in the logon request/response messages.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#567 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ACUdfVmPP2thVI-jb9cN4CR57RfI5gk8ks5t9td6gaJpZM4UoBvk>
.
|
I updated those by hand if it matters, latest values from beta SteamClient.
Let me know if there is anything else I should commit together with this PR, this worked for me. It doesn't look like a prerequisite for anything right now (managed to get new callbacks I wanted also on old protocol), but always best to track latest one.