Search results

An exception in the middle of Sun/Oracle's source code. While I introduced it to try to fix the issue, it's possible that the code I added for naive SSL caused it, so I've made it opt-in. Can you try again to see if you get the same stack trace?

10658.

Debugging this myself, I determined that the settings weren't actually saving for https.*, so that part is now fixed. I installed a small web proxy (polipo) and it seems like HTTP traffic coming out of KoLmafia is logged, but HTTPS traffic doesn't get logged in it, which matches what you've...

Someone's reported this previously, but I haven't been able to reproduce it. The latest build should give us some more information on why the stack trace happens?

10656.

Autocast if no effect present, prompt the user twice if there is one?

Thanks, I think I'm starting to understand that most people would rather tell others to work around the problem of "remove malignant status effects is not working the way holatuwol intends it to" than help me fix it. If that's the case here, that's fine, but the bug essentially stays open until...

When you say "ignored" do you mean that it skips the proxy and directly connects to KoL without it, or that nothing happens on the proxy that you can see? 10649 basically disables the native SSL security stuff that's in Java since some random posts throughout the Internet suggest it as a first...

10648.

Let me know if 10647 handles everything for y'all.

Found out that specialCommand doesn't get called when you hit the 'auto' button in the Louvre, but rather the 'Go to Goal' button. Should be fixed in 10647.

Not the same, but HP restoration causes effect removal, so making it so that HP restoration does not fire should result in a fix.

Yeah, we've got 315 (according to grep -R "new StringBuffer"). Most can switch, but Java never updated appendReplacement() to take in StringBuilder also, so there will need to be some massaging whenever Matchers are in play. Or the most frequently accessed code that uses StringBuffers (the...

Edit 2: 10639

Decided to go with your original suggestion to just remove it. 10624.

10623.

10616

The idea of an unnamed mood is when you specify multiple moods at once (the idea of "extends" doesn't come to play with this unnamed/anonymous mood); unless I had some miraculous foresight, I do not believe you can have the unnamed/anonymous mood have just one parent without cheating using...

I don't think you can delete moods via CLI yet.

Sorry, I've added an alias to recompile before it sends the commit to the server. :(