sxammy

Legend!

G'day guys, I'm using a trained Markov Chain to create random user names for an upcoming tutorial island script, however, when trying to load the saved markov_chain.dat file I'm getting the following exception: I have verified that the file is found, of note the bytes available on the input stream are zero and the type of the InputStream is java.util.zip.ZipFile.ZipFileInflaterInputStream, is OSBot trying to do some magic for me or is it slapping my wrists for not following a convention? Begin stack trace for Exception: [java.io.IOException: Stream closed] java.util.zip.InflaterInputStream.ensureOpen(InflaterInputStream.java:67) java.util.zip.InflaterInputStream.read(InflaterInputStream.java:142) java.io.ObjectInputStream$PeekInputStream.read(ObjectInputStream.java:2781) java.io.ObjectInputStream$PeekInputStream.readFully(ObjectInputStream.java:2797) java.io.ObjectInputStream$BlockDataInputStream.readShort(ObjectInputStream.java:3274) java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:934) java.io.ObjectInputStream.<init>(ObjectInputStream.java:396) org.sxam.osrs.bots.GarboIslando.utils.NameGenerator.MarkovChainGenerator.loadFromFile(MarkovChainGenerator.java:42) org.sxam.osrs.bots.GarboIslando.utils.NameGenerator.NameGenerator.generateName(NameGenerator.java:21) org.sxam.osrs.bots.GarboIslando.tasks.HandleGielinorGuideTask.sxamExecute(HandleGielinorGuideTask.java:77) org.sxam.osrs.bot.structure.SXAMStateTask.executeCurrentTask(SXAMStateTask.java:116) org.sxam.osrs.bot.structure.SXAMHarness.performStateWork(SXAMHarness.java:271) org.sxam.osrs.bot.structure.SXAMHarness.TICK(SXAMHarness.java:255) org.sxam.osrs.bot.structure.SXAMInterfaces$SXAMBotBaseInterface.sxamOnLoop(SXAMInterfaces.java:108) org.sxam.osrs.bot.structure.osbot.looping.OSBotLoopRunner.onLoop(OSBotLoopRunner.java:69) org.osbot.rs07.event.ScriptExecutor$InternalExecutor.run(ip:52) java.lang.Thread.run(Thread.java:748) The code: Reading from getScriptResourceAsStream: public InputStream getResourceStream(String fileName) { return getPlatformMethodProvider().getScriptResourceAsStream(fileName); } Loading from InputStream: NameGenerator.generator = MarkovChainGenerator.loadFromFile( SXAMExecutionEnvironment.Core.getResourceStream("markov_chain.dat")); The Uber jar structure: Any ideas guys? Am I barking up the wrong tree here? Cheers, Mitch.

Could you point us in the direction of that one? Keen to see what's been done before and released officially.

Cheers mate! I'll give that one a read

Cheers guys, Do we know why these types of scripts aren't publically available? Is there a rule of thumb or a nice set of guidelines we should be following?

G'day everyone, I'm an aussie enterprise level developer (albeit for C#/ASP.NET) - however my roots were a lot greener than the gray skies I live in today, I started programming around 10 years ago with a exceptionally awfully written looping willow tree chopper&dropper in that little pond close to lumby and then the yew trees next to Edgeville with banking. Since then I've picked up a 4 year long piece of paper and learnt about the wicked ways of design patterns, now; I'm looking to kick start a new type of botting run time. While the typical standard for a bot is designed for a single purpose such as chopping logs and dropping them or something more advanced like murdering off thousands of lava dragons in the wilderness with PK detection and all the related goodies; I thought we could reach further since this single purpose leaves us with a predicament - we're doing the same activities for a long period of time, and for larger bot farms all smashing the same resource or activity, the ever growing machine learning bot detection software at Jagex is doing what it does and learning, which is resulting in bot detection that is finding us quicker. So I've been thinking (and programming) - let's enter a new world of letting our bots be smarter, what if after a cutla hours of chopping trees we move over to some agility? What if after we've done our agility we walk over to the GE and sell those logs we chopped to buy better equipment? What if we buy quest supplies and knock off a cutla quests? Essentially my question is; what if we start making our activity patterns look more human like? I'd like to open this discussion and get some feedback from you all; my potential customers - is this something we've thought of before and failed horrendously at? I've got some working code for a different platform before they decided to fuck off, though that being said I never really asked the community what they thought, so I thought I'd move my projecto ver here to OSBot and see how far I can go this time before the preverbal rug is pulled from under me again. I've got a big vision for my project - too big to realistically achieve in a healthy time frame, I'd like to invite my fellow developers into a single platform for deploying their botting purposes across all botting clients, and since I'm here taking your feedback on how my ideas will sit with the consumer base, let me give you something in return: https://pastebin.com/aW2fv17j. While you can't copy paste this bad boy and see results immediately, some tweaking will get you a Publish/Subscribe pattern ready to go - how you publish the events into the pattern is left to you - though feel free to shoot me a DM and I'd be happy to help you learn. Since I'm giving let me give something else - I've learnt most of what I know today from this book: "Design Patterns: Elements of Reusable Object-Oriented Software" - it's a bit outdated now (there's some code language examples in there I've never even heard of before, but the major ideas still apply even in an Enterprise setting). Cheers for reading, I'm looking forward to hearing from you all, sxammy.