Search results

  1. holatuwol

    Feature - Implemented Add boolean function goals_exist() or equivalent to ASH

    Aha, I see the formatting problem now. We now have a situation where you have to parse (yourself) information that KoLmafia has at the time this map is being generated. Namely, the type of goal it is -- whether it's a regular item goal, a wildcard item goal, or if it's something that's not an...
  2. holatuwol

    Feature - Implemented Add boolean function goals_exist() or equivalent to ASH

    10755. Instead of goal_string() and has_goals(), get_goals() now returns a string[int] representing all the goals that KoLmafia is set to try to satisfy. If there's no goals, then the count() on that will be zero.
  3. holatuwol

    Feature - Implemented Add boolean function goals_exist() or equivalent to ASH

    Before you start implementing stuff ... How about we just give you a string[int] with everything split in advance to avoid the whole "split_string" that you're likely to do right afterwards?
  4. holatuwol

    Feature - Implemented Add boolean function goals_exist() or equivalent to ASH

    I can't think of a likeable name since the internal name KoLmafia uses is "hasGoals()". That being said, I could remove it entirely, and you could just check if the string from goal_string is empty.
  5. holatuwol

    Bug - Cannot Reproduce Custom Combat Script autorepeat

    Every relay request clears continuation state. To pin it down, it's probably some wonky interaction between that reset of continuation state, what happens when you push the CCS button, and what browser you did things in.
  6. holatuwol

    Feature - Implemented Add boolean function goals_exist() or equivalent to ASH

    10753. I'm not sure where your other feature request was that resulted in limbo due to formatting, but Winterbay had an open feature request which made sense to me and didn't appear to have any formatting problems?
  7. holatuwol

    Anti-mafia whining on G-D

    Yeah, once we start pushing a few things to the relay browser (my first target will be login, because that's the first thing you interact with), we can reach out and see if the relay browser itself is usable. I don't think it'd be worthwhile until at least that part works.
  8. holatuwol

    Anti-mafia whining on G-D

    Yep. We'd probably have to put a bunch of tooling in place to achieve it because it needs to be really easy to build things in HTML with some scripting in the middle, which is the interesting part. I'm leaning towards something that translates some HTML-like input file into an ASH relay...
  9. holatuwol

    Anti-mafia whining on G-D

    At one point, I wanted to torch the main UI and make everything equivalent to custom relay override paths accessible through the browser. That'd make the notion of a 'mobile' KoLmafia theoretically possible for people who have their computer on but only want to run CLI commands and chat while...
  10. holatuwol

    ASH Map error

    Ah, I was saying that your inner loop with the 'j' counter is slow because of how you shift the elements left. In simpler terms, if I threw the list of all 5600 items in the game at your shuffler, on average, the nested loop implementation would do something like 8 million shift operations (15...
  11. holatuwol

    Bug - Fixed Proxy Settings in r10587

    I added additional logging to see if that code is even called. If I were to guess, it's a bug with early access.
  12. holatuwol

    ASH Map error

    Because it's vaguely relevant, the shuffling algorithm you're trying to implement is only O(n) when the remove operation is O(1). The way you've implemented it, remove is O(n) and therefore the algorithm winds up being O(n²). Personally, I'd use a simpler fake shuffle. void shuffle( int[int]...
  13. holatuwol

    ASH Map error

    That's because you're thinking of maps as arrays, so a coding mistake that would normally result in an index out of bounds exception doesn't happen. More explicitly: input[j]=input[j+1]; What happens when 'j' is 6? You look at the value in input[7] which is not defined (and therefore zero)...
  14. holatuwol

    Bug - Fixed Proxy Settings in r10587

    The log is helpful, at least. It says that Java 1.7.0 update 4 early access (which never worked), unlike Java 1.7.0 update 3 (which worked), still tries to use certificate validation no matter what we mark there. Rather than implement a naive one, we'll probably have to create a custom...
  15. holatuwol

    Bug - Not A Bug Not logging in

    Yeah, I'm trying to dig deeper into the core Java code to see where in that code it's throwing the exception, so I put in a stack trace.
  16. holatuwol

    Bug - Not A Bug Not logging in

    Ah, Veracity asked if you could enable debug logging. KoLmafia created a DEBUG_[date].txt After checking to make sure there's no passwords in there (and removing any such passwords if they are still there -- we should have removed it, but it never hurts to be extra careful), can you upload it?
  17. holatuwol

    Bug - Not A Bug Not logging in

    Can you upload the stack trace?
  18. holatuwol

    Bug - Fixed Proxy Settings in r10587

    1. Setup a proxy server locally. 2. Install Java 1.6.0 update 23 to a VM using host-only networking. 3. Run KoLmafia, adjust your proxy settings, and attempt to login. If it doesn't work: 4. Follow the instructions here: http://java.com/en/download/help/proxy_setup.xml 5. Delete your existing...
  19. holatuwol

    Bug - Fixed Proxy Settings in r10587

    Not too good, I think I need a 32-bit Windows XP to test, since the computer that I installed the proxy server on won't let VirtualBox use 64-bit Windows. That being said, I didn't think of trying to test with 32-bit Ubuntu with the older version of the JDK to see if it's Java 1.6.0 update 23...
Back
Top