blueboy_

It's very difficult to comment on all changes that occur through merging code. I haven't purposely made any of these changes. I only change code relevant to playerbot and any other changes must be attributed to merges with the mangos zero core. I've just finsihed updating the code and did note some strange things. The changes applied by two commits were not present. I had to re-apply these to resolve a compile issue on windows. Interestingly, the .gitattribrutes. https://github.com/blueboy/portalzero/commit/1caa01b6cd981321bd5f8699febcacf31fbde5d1 comes and goes. Some commits show it and others don't. The time stamp on commit history is also screwed up. Work that I posted in August this year is listed as being 'one year old'. Are you experiencing any problems with the changes? I will try to revert unwanted changes, but it may take awhile. Cheers

Sorry, If I gave out my Skype Id, I'd never get any work done I will try to answer any questions through the forum as soon as I can. There is more to playerbot than just the files contained in the 'playerbot' folder. If you have just moved the code from this folder, I'm not surprised that it doesn't compile. Please Note: We only support offical core code and not hybrid code used to generate repacks. Example:Merge zeronew-ai code into the latest mangos zero core run the following 'bash' script from within the 'GitBash' shell on windows. If you haven't got 'GitBash' you can download it from http://msysgit.github.com/ #!/bin/bash -x git clone git://github.com/mangos-zero/server.git mangos cd mangos git pull git://github.com/blueboy/portalzero.git zeronew-ai This should merge the code without any conflicts. Then using VS2008 (win\\VC90.sln) or VS2010 (win\\VC100.sln) compile the code. This again should compile without issue. If not, you must give me details on the errors your getting, I can't read minds.. Hope this helps

Hi Swockale, O.K I've had a chance to test the code with windows and there was a problem. It just shows how popular playerbot is ; at least on windows. Nine months ago some code was removed from the core ( You're the first person to draw this to my attention) and for some unknown reason it was not translated to our code. There was a core merge with lots of conflicts at that time and I may have overrode the removal of this code. Whatever the cause, I do apologise and it's fixed now. I've re-applied the missing commits and updated both repos. I can confirm that they now merge without conflict with the current mangos zero core and compile without issue. 11> 6 file(s) copied. ========== Build: 11 succeeded, 0 failed, 0 up-to-date, 0 skipped ========== If you're still having problems, let me know.

Sorry I don't Skype, but I'm sure we can sort this out. I assume you are using mangos zero (Classic) core. Which version of Visual Studio are you using? Was the compile successful? I assume you used git to do this Did you get any merge conflicts? Which version of 'Playerbot for MaNGOS Zero' are you using? https://github.com/blueboy/portalzero https://github.com/blueboy/portalzero/tree/zeronew-ai In the meantime I'll check that our code compiles with the latest core.

I will help if I can (no charge ), but I need to know where to start... Are you able to compile the core on windows? Have you taken a look at the playerbot wiki linked in first post? Let me know

I'll update playerbot for R2 once I've finished the 'onenew-ai' port. Shouldn't be too long Hope this helps

I couldn't agree with you more The backports are fun to play and they give the developers an opportunity to refine the code and coding skills (we're all learning), for the benefit of all versions. Thanks for your interest in playerbot and your contributions to the core

Sorry, I've only just replied to your post on Github. https://github.com/blueboy/portalzero/issues/1 I have been busy updating the portalzero master code to be compatible with the current mangos zero core. Long overdue I know :rolleyes: , but very few people seemed to be interested in the port since I created it. There were tons of merge conflicts to fix, including the GetTrainerSpellState issue ( that you addressed yourself) , but I'm please to say that they're all done now. I'll be pushing the code shortly and then I'll do the same with the portalone zeronew-ai branch. Hope this helps

All the changes that kennumen refers to are only on the portal/new-ai branch. Eventually we will be merging this code into the portal/master branch, but not yet. It's not that confusing. Only playerbot for MaNGOS currently has a repository for release (tested) code. https://github.com/playerbot/mangos Beta test code (General) The portalone/master (playerbot for MaNGOS one - tbc) is equivalent to portal/master (playerbot for MaNGOS - wotlk) branch. Alpha test code (Development) The portalone/onenew-ai (playerbot for MaNGOS one - tbc) is equivalent to portal/new-ai (playerbot for MaNGOS - wotlk) branch. Please note that the branches maybe equivalent, but the code is not identical, so the files are not transferable. That's why the branches are kept separate. If you wish to use the latest features, I would recommend you use the respective 'Alpha test code'. Although the code might be a little buggy Hope this helps

Sorry it was my fault. I was too focused on the portalone code and I 'jumped tracks' in the discussion. I've now edited the post http://getmangos.eu/bb/post/1609/#p1609. (post 29) and thanks for the clarification on 'combat orders'. I've been working with the revised code and it works great @Update I'm currently testing code (on portalone/onenew-ai code, sorry ) that will help keep ranged bots at an appropriate distance from the target. If the ranged bot does attract the mob, only then does it resort to melee combat. Obviously if the bot is out of ammo, it will automatically switch to melee combat and not use it's bow. Once I'm happy that it work properly, I'll let you guys test it out. Hope this helps

It would be best for you to speak with kennumen about this. I'm sure it would be possible, and I believe it maybe handled by the 'talent specs' feature, that he developed for playerbot. If you can't get his attention here, you might try the playerbot forum. Hope this helps

Yeah, Sorry I've just found it and sent you a reply. Sorry I thought you were notified by default, when someone sends you a PM, obviously not. When we originally put playerbot version info into revision_sql.h file I could see the pitfalls. I agree that it might be a better idea to move this to a dedicated playerbot file. Much like we did when moved the playerbot mod info from mangosd.conf.dist to playerbot.conf.dist, to prevent conflicts. We'll have to have a think about this Cheers

Hi Nooblord, Sorry for the delay in replying to your question, but I have been busy working on a playerbot port for MaNGOS one. Unfortunately this merge issue with "src/shared/revision_sql.h" will occur everytime the core or playerbot updates it's sql files. It's simple to fix and I will illustrate this with the current merge (remove text in red and reorganize text in green) revision_sql.h with conflicts #ifndef __REVISION_SQL_H__ #define __REVISION_SQL_H__ <<<<<<< HEAD #define REVISION_DB_CHARACTERS "required_11785_02_characters_instance" #define REVISION_DB_MANGOS "required_12091_01_mangos_spell_template" #define REVISION_DB_REALMD "required_10008_01_realmd_realmd_db_version" ======= #define REVISION_DB_CHARACTERS "required_11785_02_characters_instance" #define REVISION_DB_PLAYERBOTAI "required_5_playerbotai_auto_follow" #define REVISION_DB_MANGOS "required_12012_01_mangos_spell_template" #define REVISION_DB_REALMD "required_10008_01_realmd_realmd_db_version" >>>>>>> new-ai #endif // __REVISION_SQL_H__ The order of 'REVISION_DB' version is always the same. characters playerbot mangos realmd Simply decide what has changed. You will see here that only the 'REVISION_DB_MANGOS' revision has changed. Use your editor to then update the file to look like this and save the changes revision_sql.h fixed #ifndef __REVISION_SQL_H__ #define __REVISION_SQL_H__ #define REVISION_DB_CHARACTERS "required_11785_02_characters_instance" #define REVISION_DB_PLAYERBOTAI "required_5_playerbotai_auto_follow" #define REVISION_DB_MANGOS "required_12091_01_mangos_spell_template" #define REVISION_DB_REALMD #endif // __REVISION_SQL_H__ Hope this helps

O.K. so there may well be problems in the portal code. How close is the huntress to the victim as combat starts? In close combat, ranged attackers will change to melee attack for self protection. I'm sure you have tried attacking a target from a distance, also without success. This prompted us to improve the code in the new-ai branch. Thanks for the feedback @EDIT Remove reference to portalone code, when I should have been talking about portal new-ai code. Sorry my bad

I see what you were doing now. Your trying to get the hunter to assist the player. That's fine /s .bot co <botname> assist <player> but a player does not have any fix orders. If the player joins in combat, the hunter will tend to assist as a melee dps. Also if the hunter is your only bot, it maybe the center of attention during the attack (highest threat). In which case it will resort to melee combat. Try including a tank bot and a healer bot to your party. You may well be correct, the AI code on the 'new-ai' is quite fluid at present and the ranged attack maybe messed up. Just a few ideas, let me know if it helps

Thanks for the feedback. We are aware of many of the issues you mention, but it's good (not really ) to see that other people getting twitching bots. This occurs particularly if the bots are very close to a terrain feature (i.e tree or fence) or on an incline. This effect I believe is related to the movement system in the core. The closer the bots are to the master, the smaller the effect. This issue is unfortunately aggravated by the new 'follow auto' feature in the 'new-ai' code. If you turn this feature off, the twitching stops. Currently we have no solution for this, but would welcome any help in resolving it. To set combat orders for 'assist' (Note: make sure you use '/s' say rather than '/w' whisper) For a ranged assist, choose a bot that will hold back from the combat (i.e healer). /s .bot co <bot name> assist <healer name> For a dps assist, choose a bot the tank. /s .bot co <bot name> assist <tank name> Hope this helps

Please remember if you have cloned the portal repo to your harddrive, you will already have the remotes setup (i.e no need to use fetch). I'm lazy, I don't bother updating existing code, I just do a clean 'clone' . >git clone git://github.com/blueboy/portal.git >git branch -a * master remotes/origin/HEAD -> origin/master remotes/origin/autobot remotes/origin/craftbot remotes/origin/master remotes/origin/new-ai remotes/origin/sharedbots Then to merge >git merge origin/new-ai Then to update the code, if you wish >git fetch origin To explain what happening for you. I will use the master branch as an example, using the 'fetch' approach explained in my last post. 1. The merge.. >git merge portal/new-ai 2. Fix any conflicts and save file changes. (If you get any conflicts you will need to fix and then commit them before the merge takes). >git commit -a (edit commit header, when the editor pops up and save changes). 3. Checks >git status # On branch master # Your branch is ahead of 'origin/master' by 188 commits. # nothing to commit (working directory clean) and or compare the commit history in your current branch, with that on the 'new-ai' branch. >git log If you try to merge again, you with no changes > git merge portal/new-ai Already up-to-date. I need more info, to be able to help @Edit Nice one Cheers

Hi, I know what you have done. You have copied the browser github link address for the new-ai branch. Git can't use >git fetch https://github.com/blueboy/portal/tree/new-ai fatal: https://github.com/blueboy/portal/tree/new-ai/info/refs not found: did you run git update-server-info on the server? To use fetch to merge the 'new-ai' code into your local branch, do this 1. Setup a remote labelled 'portal' >git remote add portal git://gitihub.com/blueboy/portal.git 2. Download the current status of all remote branches from the portal github ( Also re-use to update) >git fetch portal remote: Counting objects: 77037, done. remote: Compressing objects: 100% (11225/11225), done. remote: Total 75411 (delta 64891), reused 74553 (delta 64105) Receiving objects: 100% (75411/75411), 36.97 MiB | 2.01 MiB/s, done. Resolving deltas: 100% (64891/64891), completed with 1295 local objects. From git://github.com/blueboy/portal * [new branch] autobot -> portal/autobot * [new branch] craftbot -> portal/craftbot * [new branch] master -> portal/master * [new branch] new-ai -> portal/new-ai * [new branch] sharedbots -> portal/sharedbots 3. To merge 'new-ai' into the current branch >git merge portal/new-ai Hope this helps

Glad you enjoy playerbot. The code on the portal master is more stable, but lacks many of the features currently being tested on the new-ai branch. We endeavour to fix issues quickly, so yes we would encourage you to test the new-ai code so we can then merge this code into the master Cheers

@Svarr I have made some changes to the code that I hope resolves the issue your getting. The revised code is on our new-ai branch https://github.com/blueboy/portal/tree/new-ai and you'll find more information on this patch here https://github.com/blueboy/portal/issues/10 Please let us know if helps in anyway. @UnkleNuke Always a pleasure to talk with you

Hi, Thanks for the feedback. We are aware of this issue and we're currently looking for a solution. The bug has been posted here, https://github.com/blueboy/portal/issues/10 A lot of changes have been made to the combat AI recently and it might take awhile to locate the cause. When I have more information I will post to the issue section on github. Cheers

If your interested you can get more information on Playerbot for MangosR2 from our forum (see first post). Taster of what the code will do: https://github.com/blueboy/portalR2/commit/140679e3505e5ce0b15479d8e5269d23d7c45687 The wiki will show you how to merge the code with the MangosR2 core and you can try it out for yourself.

The bots are pretty good, but a bot that kicks everyones ass, has no master and that touches on autonomy. I attempted to achieve a self fighting bot for FFA (free for all) arenas, without success. I did get the bots to turn hostile to the master and indeed they attacked. Problem was that I couldn't stop them attacking once they had started :lol: This would certainly be a challenging project for someone who's interested. It shouldn't be too long before working arenas are a reality on MaNGOS (I know R2 have had some success here). When it does, I have the code to get bots into arenas, fight as a team and then exit properly at the end.

What is Playerbot? It lets you add additional characters from your account (normally upto 9), as bots that you control. The bots assist in all aspects of normal game play, combat, trading, questing and much much more…. Sorry they can't make coffee yet More Information: Playerbot Wiki Source: Playerbot Development Code Report bugs: Playerbot Issues (If possible) Additional Links: Playerbot Forum This is a community project and all contributions are welcome. I would like to acknowledge the hard work of all those who have previously contributed and made Playerbot what it is today

What is Playerbot? It lets you add additional characters from your account (normally upto 9), as bots that you control. The bots assist in all aspects of normal game play, combat, trading, questing and much much more…. Sorry they can't make coffee yet More Information: Playerbot Wiki Source: Playerbot Release Code Playerbot Test Code Report bugs: Playerbot Issues (If possible) Additional Links: Playerbot Forum This is a community project and all contributions are welcome. I would like to acknowledge the hard work of all those who have previously contributed and made Playerbot what it is today