blueboy

Hi, I see what you are doing now. O.K. What's your repo called. I just searched github to see if I could glean some more info, and could not find a reference to klunk. Is it a public or private repo? It sounds like you did exactly what I did with my repo (you forked off Playerbot), and your going to suffer all the headaches I did. If your using this a means to appreciate 'git' & 'github' better, then thats fine. I'll guide you through any pitfalls. However if you want a repo that you can quickly use, you should have forked off blueboy. The tell tale signs that this is so, are the issues with GOSSIP menu. From MaNGOS[8898] they moved alot of code around on the main MaNGOS branch, and it was necessary to move the playerbot code with it. They did not make the move in one commit, so this caused further issues. O.K so your following my post #394, yes. I will assume you are using linux, or know how to convert bash scripts to something that is compatible with your system. I created the two scripts to fully automate the patch creation process. When run they will create the patch without any intervention. If you are having issues with the gethash.sh script, you can of course revert to manually obtaining the <commit HASH> as I did in the begining. Run the 'pbot.sh' script in your working directory. This will call 'gethash.sh' when required. Script Process 1. We need to determine what is the 'HEAD' or latest version of MaNGOS branch at the time of the current 'Playerbot' code. It is understood that we are looking at and not Or you will get the wrong code. I believe this is where you are going wrong. I suggest you change the line in 'pbot.sh', # get hash for latest mangos release [XXXX], associated with playerbot release git pull git://github.com/playerbot/mangos.git master to # get hash for latest mangos release [XXXX], associated with playerbot release git pull git://github.com/blueboy/mangos.git master and # add remote playerbot branch to compare with local mangos branch git remote add playerbot git://github.com/playerbot/mangos.git to # add remote playerbot branch to compare with local mangos branch git remote add playerbot git://github.com/blueboy/mangos.git Apply the created patch to your clean MaNGOS[9162] branch on your harddrive patch --dry-run -p1 < playerbot.patch (I like to use --dry-run to test the patch first) To commit back to your repo, git status // will show all files modified git add <modified files> // will add modified files to commit git commit -a -m "Commit description" // 'git commit' without options for merges git push // upload commit If it still doesn't work, let me know

Hi klunk, Sorry I do not understand what you are trying to do. Are you trying to create your own online repo ? or Are you just trying to merge MaNGOS with Playerbot on your local hardrive? I will help you if I can, but I need to know clearly what you are trying to achieve. Please get back to me @erazare Good to hear the code on blueboy is behaving itself. I'm trying out a few ideas that I hope will improve the AI. Keep on trucking

Hi warrior, I can't say I have experienced this on my server, however I will need details. Can you describe fully what happens when your player dies. For instance, Are you transported to a graveyard? Is there an angel in the graveyard to force resurrection? What happens when you find your corpse? Are there any unusual messages popping up on your client or server? The non-resurrection fix would only effect the bots and not the player Please get back to me.

Hi WOLFMAN, I do not wish to get angry with users, but I am forever telling people that the version of MaNGOS I say the code is compatible with, is the ONLY version that it is guaranteed to work with. Occasionally, the code will work with a future release of MaNGOS, but this is not guaranteed. In my post #539, I stated that the latest code is compatible with MaNGOS[9162],and not MaNGOS[9173]. If you look at the MaNGOS changelog, you will see that there has been a change to CharacterHandler.cpp in MaNGOS[9164]. I will address this issue when I update blueboy next. EDIT: I just pulled MaNGOS [9173], and applied the 'playerbot.patch' create from the MaNGOS[9162] merge. I had no conflicts? Have you got any other core mods in the code, that may be conflicting with the merge? Look inside CharacterHandler.cpp, the system will mark where the conflict occurs, when a merge fails. Cheers

Hi Guys, For those of you who are interested, I have at last updated the Blueboy fork. It is compatible with MaNGOS[9162], and compiles without issue Cheers

Hi WOLFMAN, Yeah i've just found a link to 'Attunement to the Core' which is a requirement for access to Molten Core http://www.wowguru.com/guides/molten-core.php Cheers

Hi WOLFMAN, I see what you mean, I've just taken a small group of deathknights to the gates of MoltenCore. It only allowed the player through the portal. ???? Yeh, your guess is as good as mine. Maybe it is an issue that need addressing with playerbot. I don't think the system acknowledges a small group as being a RAID group, although it appears to do the conversion. but thanks for drawing it to our attention. Cheers

Hi WOLFMAN. I wouldn't think that a 5 man group would be classed as a RAID group. If you said 30 or 40, that would be more realistic. My understanding was that RAID groups are an amalgamation two or three GUILD groups! It sounds like Molten Core is quite a hot spot :rolleyes: and you will need an army, rather than a small group. The 'player' defaults to the party leader. It would make no sense for a bot to become leader 'dog wags the tail, the tail does not wag the dog' I did .tele moltencore and got the message about being in a raid group Reading your last post I'm a little confused, :confused: 1. Start in Ironforge. 2. .tele group MoltenCore 3. In front of the gates of MoltenCore Did the .tele work or not? I thought your probelm was teleporting to Molten Core. The message you get might be a warning that your group is not prepared for what is on the other side of the gate. Is the gate entrance a portal? I know portals act like teleports. Cheers

Hi WOLFMAN, I don't get the chance to play WOW these days because I am too busy solving issues. I don't even know where 'moltencore' is? However I know where your going wrong. I just tried taking a group of bots to molten core, with one of my GM accounts. If you read the Message it says Are you in a raid group? Hope this helps

Hi, How are you at scripting under windows, I use linux, sorry the scripts will need converting. Here goes; You will first need to clone the MaNGOS repo to your hardrive. Then, change to the appropriate commit version git clone git://github.com/mangos/mangos.git git log // find the <commit HASH> for [9147] git checkout -b 9147 <Commit HASH> git log // check that [9147] is the new HEAD Auctionhousebot integrates without issue. It's Playerbot and Vehicle code that clashes. Both update Player.h at the same point. My process will require you to create three patches. 1. auctionhousebot.patch git://github.com/Naicisum/mangos.git 2. Vehicle.patch git://github.com/Tasssadar/Valhalla-Project.git 3. playerbot.patch git://github.com/blueboy/mangos.git // my copy of the Main Playerbot branch 1. To create auctionhousebot.patch use the following script; #!/bin/bash -x git clone git://github.com/Naicisum/mangos.git ahbot cd ahbot git checkout origin/ahbot git checkout -b ahbot git diff master ahbot > auctionhousebot.patch the patch is stored in the ahbot directory. Copy this to your MaNGOS source root directory 2 & 3 Use the bash scripts I gave in post #394, of this thread. These were made to create the playerbot.patch, but can be adapted to create the vehicle.patch too. 3. Playerbot. Check that the script gets its code from blueboy rather than playerbot. The scripts should then run unattended, copy the patch from the playerbot directory to the mangos directory. 2. Vehicle. Change all 'playerbot' references to 'vehicle'. Make sure the script looks for code on the Tasssadar repo. Again run script, and copy the created patch from the vehicle directory to the mangos directory. Right now the fun part. Editing the patches so they work: 1. playerbot.patch must be edited as follows, @ line 306 , within ChatHandler.cpp, void WorldSession::HandleMessagechatOpcode( WorldPacket & recv_data ) // a single space is missing between 'lang,' and player. GetPlayer()->Whisper(msg, lang,player->GetGUID()); must be changed to GetPlayer()->Whisper(msg, lang, player->GetGUID()); @ line 1086, within PlayerbotAI.cpp, void PlayerbotAI::HandleBotOutgoingPacket(const WorldPacket& packet) // remove the '2' from MOVEMENTFLAG_FLYING2. m_bot->m_movementInfo.AddMovementFlag(MOVEMENTFLAG_FLYING2); must be changed to m_bot->m_movementInfo.AddMovementFlag(MOVEMENTFLAG_FLYING); repeat this @ line 1098 m_bot->m_movementInfo.RemoveMovementFlag(MOVEMENTFLAG_FLYING2); 2. Vehicle.patch must be edited as follows, change the block of code @@ -2305,6 +2313,11 @@ class MANGOS_DLL_SPEC Player : public Unit bool isActiveObject() const { return true; } bool canSeeSpellClickOn(Creature const* creature) const; + uint32 GetActionButtonSpell(uint8 button) const + { + ActionButtonList::const_iterator ab = m_actionButtons.find(button); + return ab != m_actionButtons.end() && ab->second.uState != ACTIONBUTTON_DELETED && ab->second.GetType() == ACTION_BUTTON_SPELL ? ab->second.GetAction() : 0; + } protected: uint32 m_contestedPvPTimer; to @@ -2305,6 +2313,11 @@ class MANGOS_DLL_SPEC Player : public Unit bool isActiveObject() const { return true; } bool canSeeSpellClickOn(Creature const* creature) const; + uint32 GetActionButtonSpell(uint8 button) const + { + ActionButtonList::const_iterator ab = m_actionButtons.find(button); + return ab != m_actionButtons.end() && ab->second.uState != ACTIONBUTTON_DELETED && ab->second.GetType() == ACTION_BUTTON_SPELL ? ab->second.GetAction() : 0; + } // Playerbot mod: // A Player can either have a playerbotMgr (to manage its bots), or have playerbotAI (if it is a bot), or The patches must be applied in a specific order. I suggest, N.B. --dry-run allows you to test the patch before actually applying. 1. patch --dry-run -p1 < playerbot.patch 2. patch --dry-run -p1 < auctionhouse.patch 3. patch --dry-run -p1 < vehicle.patch Hope this helps.

Hi DaemonCantor I have just finshed compiling Playerbot from blueboy with (Naicisum's) AHBot patch & (Tasssadar's)Vehicle patch without issue. MaNGOS[9147] SD2[1543] The fix is quite tricky, so I must ask you whether your prepared to follow a quite lengthy process, with alot of keyboard work? Oh, Are you using Linux or Windows? Let me know.

Cheers DaemonCantor

Hi DaemonCantor, O.K I've just upgraded my clients to 3.3.0 and tested the following; MaNGOS [9146] SD2[1543] PSMDB[281] This presently only supports 3.2.2a. However it will load with a few errors. I applied the relevant sqls from the MaNGOS source to update the World & Character databases. It shouldn't be long before Silvermoon are supporting 3.3.0a. I got the same single issue as erazare, when compiling. They have changed the use of MOVEMENTFLAG_FLYING2 to MOVEMENTFLAG_FLYING. This requires changing in PlayerbotAI.cpp. It will then compile without fault. I will update blueboy presently. Initially I only compiled the MaNGOS core with Playerbot. I also use AHBOT and I will check it out this evening. I haven't used Vehicles, but if you can let know where I can get the mod from, I'll give it a try, and get back to you. Regards blueboy

Hi infamousd, It still sounds like your problem was with the 'botguy' (Pappy Looter). Moving Playerbot support from MaNGOS [8898] to [9106], broke the botguy code (due to GOSSIP menu changes). I figured out a fix that seemed to work with linux, but not windows. Meanwhile Stillhard and yad02 found a fix that worked with Windows, but not linux. After a great deal of debate it was thought best to remove the botguy code from the Playerbot core. If you wish to use 'botguy' code then as I said in my last post, there is information in this thread to put the code back. I find commandline summoning far more reliable, and it does not require a GM to summon the 'bot recruiter'. I create macros to summon the bots, so its a one click process. If you have found a common solution to the 'botguy', I am sure that members would be interested to see your fix. Cheers.

Hi erazare, I just searched the code for 'taxi_path' and got a bunch of references to the character database in CharacterHandler.cpp & Player.cpp files. Hope this helps

Hi lecails, If you compare the MaNGOS and Trinitycore you will find that they are similar in many respects, but very different in others. In simple terms, 'they both follow different paths, but arrive at the same destination! One main difference is that MaNGOS utlilizes quest scripts, developed by ScriptDev2.com. Trinity uses its own custom scripts, integrated into the core.

Hi erazare, I assume the 'next flightmaster' you mentioned, is in the same Continent. Sorry I know that is a stupid question! :rolleyes: As you progress from one flightmaster to the next, the character database should be updated to reflect new flight paths found. It sounds like this is not occuring. Maybe the flight path info in your modified DB is incorrect! Just an idea

Hi erazare, I guess you must have tried all DB's compatible with 3.3.0. What aspect of the flightmaster doesn't work? Do they refuse to respond or do they actually crash the server, like our friend 'botguy' did. I would have thought that the mechanics of the flightmasters would be in the MaNGOS code. For instance, when you select a new flightmaster, the system has to recalculate flight paths for the player. The DB would just personalise the flightmasters, gossip menu stuff etc... When I get around to looking at evo-X, I'll be in a better position to help. Cheers

Hi, Yes I normally like to check that the code compiles first, but I was strapped for time and I was hoping that someone could check it out for me. Particularly since I was requested to provide support for evo-X. Now that MaNGOS has switch to 3.3.0, I'll have client machines available to test it out. So watch this space. I believe the issue is with the core rather than playerbot. There seems to be some version of scriptdev2 included with the source. They might be custom scripts, just as Trinitycore includes 'trinityscripts'. Plus a number of configurational issues. I should have checked that the core code compiled, before integrating playerbot :rolleyes:

Hi erazare, Bump, I just finished saying that it would not be long before MaNGOS went to 3.3.0. . I just tried compiling the code used on blueboy/evo-X, and it is buggy as hell. It will take some time to sort out, so be patient. Which is the main evo-X branch? beleko or salja I want to use the branch that everyone else merges from. Do you know the address? Is it http://github.com/beleko/evo-X-Core I did a search on github and these were the only repos found. MaNGOS 3.3.0 support is my prority! Cheers

Hi guys, Just to help, I have integrated the playerbot code into eve-X-core. I am presently supporting the MaNGOS branch which still runs with client 3.2.2a. I do not know whether the code works or not. This is where you come in. If you have already upgraded your client to 3.3.0, then perhaps you can test it out and give me some feedback. The github repo can be located at; http://github.com/blueboy/evo-X-Core There appears to be two evo-X forks, one run by Salja and the other by beleko. I have forked off beleko, shoot me if I got it wrong :rolleyes:

Hi Guys @mec69, SVN had its day, although I still use it to obtain code from certain sites. Git is the future in 'Revison Control' software (HG similar). Once you get the hang of it, its great. True. 'git' was first developed by 'Linus Torvald' the father of linux, but I believe there is access support for other platforms http://github.com/guides/using-git-and-github-for-the-windows-for-newbies @mec69, @erazare & @Hanpuch I've just had contact with Sajla, and yes I believe he is busy trying to resolve the 'Eye of Acherus' quest. I have briefly looked at their repo, and there is a great deal of activity at present. This commonly happens when switching client versions, with all flavours of mangos. There is always a flood of issues that need resolving, and the DB support can be poor. I am sure that it won't be long before MaNGOS does the same. I'd wait a whille before making the jump to 3.3.0, thus allowing for core, DB & mod support to catch up. @Hanpuch There is a saying "If it works, don't mess with it". I am not defending poor support in any way, but maybe the support activity seems poor because Trinity essentially works. To my knowledge the 'eye of acherus' quest doesn't work on any other flavour apart from Trinity. That said, I am very happy with MaNGOS, because of the excellent support it provides I haven't come across the 'mine car' quest but its great that it works on evo-X. The more choices available the better in my opinion... Cheers all

Playerbot was created as a core modicfication to MaNGOS, by lots of people. All my spare time is devoted to maintaning & improving this code on blueboy. Most of the playerbot code is contained in standalone files, separate from the MaNGOS core. I do not know how evo-X differs from MaNGOS, but I am sure it would be possible translate. (Warning: I am presently comparing Trinitycore with MaNGOS. Not to support playerbot, but to fathom out why certain things work on Trinitycore and not on MaNGOS. They are quite different! So be prepared for lots of issues with evo-X. Have you tried applying 'playerbot' to evo-X yet? If you manage to integrate the code into evo-X, maybe could 'fork' off the evo-X branch, and provide ongoing support with your own repo, like I did with blueboy;) Cheers blueboy

Hi mec69, I just missed your last reply, and I don't think this forum notifies users when posts have been edited. So, take a look at my last post Cheers

Hi, The playerbot repo is integrated into a version of MaNGOS. If you look at the commit 'history' of [vladimir: link removed] you will see just below my last commits, a commit whoose description includes [9106]. This was the HEAD of the MaNGOS branch when I last merged. Most people pull and compile the code as is. You don't really need to create a separate patch. However, I do. So I will point you in the right direction. You will need to 'pull' a fresh copy of the MaNGOS branch, and change to commit [9106]. Do you use a Windows or Linux server? If you use linux then I provided the 'bash' scripts to create a playerbot.patch in post #394, in early December on this thread. If you use Windows then I am sure you could adapt these scripts. Please note that these scripts were written when I used to pull from the 'playerbot' repo, change the address to point at 'blueboy' :rolleyes: It is important to know which version of MaNGOS the current playerbot code is compatible with. I frequently get questions asking why the code will not work with any version of MaNGOS. MaNGOS is changing all the time and playerbot code has to change with it. Hope this helps [vladimir: make fork of mangos sources not problem ofc base at GPL rights granted to you, BUT copy sources with removing authors credits not acceptable. You also not make any referecnes about source project for your fork. Maybe this is less problematic but also not show you from non-nice side. I also can't welcome posting links at forum to repos with merged to mangos sources scripting project by legal reason ] Sorry, if I have caused any offence. It was not my intention to deny credit to the original authors of the referenced code. When I merged the playerbot code with code from the main MaNGOS branch, 'Git' automatically assigned me as the author of the commit. I will check if it is possible to override this feature of git, before I commit again. [vladimir: ok, if this not planned result of merge. If you find way fix this have free drop all my notes in post and restore links. About scripting merge to repo. It's your repo and in independent from mangos team position not our problem mostly. If you use cherry-pick then use -x options