balrok

thank you,

i will run a testcompile first and then upload it.. outdoor-0.12 gets fixed in a few minutes too

edit: i'm not sure if this is my problem or mangos-0.12 but for outdoor-0.12 i get

"make[3]: *** No rule to make target `2008_12_22_01_mangos_creature_equip_template.sql', needed by `all-am'. Stop." so i couldn't run a testcompile..

i have currently not much time to investigate, so maybe some can test it for me

edit: thank you fullburned.. and freghar already fixed it

thank you for pointing it out

again a filebeam error - for the update i fixed the spelleffects-merge-error (very easy)

then for getmap i had to change in outdoorpvpsi.cpp

Map * map = MapManager::Instance().GetMap(plr->GetMapId(), plr);

to

Map * map = plr->GetMap();

in function: HandleDropFlag

i hope everything else will play well together with the getmap-patch

if you get strange new errors please report

-----

outdoor-0.12 is also updated (again the spelleffect-mergething).. i haven't checked for compile-problems, but since mangos-0.12 is just a subset of master, and the getmap-patch is not in mangos-0.12, i think there should be no problems

ah i thought you want it to master, cause you used this formal title-design

i don't know about minigame-gameobjects (i even haven't heared, such exist^^)

i have done a quite similar thing for gameobjects, where i wanted that some gameobjects have extra-information

http://github.com/balrok/mangos/commit/2a61e79ce005732dcc13cb9654f43314e4bb2f3c

instead of adding this function to battlegroundmanger, you can add it to objmgr

and then, from inside gameobject::use you'll also have access to the objmgr and can ask if the gameobject has this special attribute

edit maybe this table is supposed to do such things - but i don't know anything about it

http://wiki.udbforums.org/index.php/Gameobject_scripts

maybe a message can done with Source->GetSession()->SendNotification("Your Level is too low");

if you want to improve you patch further: move the mysql-query somewhere at the startup-loads (at the place where all gameobjects are loaded).. cause else this query will be called at every gameobject use, which is maybe to much for a larger server..

so just do at the beginning a query, give a variable in gameobject-class the teleportid (or whatever is needed - haven't looked close)

and then you just check at gameobject-use if teleportid is present or not..

but i believe this patch couldn't be applied to mangos.. i think such custom-features are not very welcome (also ahbot, jail couldn't get to mangos)

ps: also it would be great if you make 4spaces instead of tabs

thank you, while moving things into the header i accidently changed return from false to true

sorry - i don't know how it happened.. i'm pretty sure, that i compiled after everything..

a fixing commit is now uploaded

and the patch by thyros is wrong

i had to revert this change

first problem was, that i mixed up OutdoorPVP with OutdoorPVPObjective

and 2nd bug, was that the playerclass in the header wasn't declared fully.. so i just moved it back..

@mythos: lol ^^

i send some patches to outdoor (not -0.12) i just cleaned up the codestyle

but if something behaves different from those functions we know what it was (:

made some virtual-function in outdoor to normal functions

i think it's more clear, if functions, which don't get overwritten

don't have the virtual-tag

if some subclass needs to redefine them, we can readd the virtual

later

also fixed compile warnings

moved some virtual function-bodies from .cpp to .h

also changed some logoutput (added "OutdoorPvPObjective" in the beginning)

more work on logoutput

used vims autointendent for a better codestyle

about the virtual-function-part i'm not very sure.. but i think, if i've would done a mistake there, the compiler would complain about it

also i splitted up the setgoartkit into another branch (artkit)

if someone knows something about this artkit, please tell me - i guess it will changes the display of the gameobject..

@ spy from blizz:

you're investigations were right ^^

@netsky the patch you posted, is already inside the source

commit 8ac221b6 or d95a7328 .. in this commit i even managed to add yad02 as author ) )

edit: or at least should be, if you say i forgot to push, i'll look at it

ahh maybe you're checking out the wrong sources, the new source is in my repository, something like: github.com/balrok/mangos

@why not implemented:

this code was currently not reviewed.. also the bug , that stealthed players caused a clientside lag (which was fixed in e5911a8b1ae6afb37cd0562e976bd2618e1b7d2e )

showed that this patch needs to be looked through..

also this patch introduces much redundant code (addobject/addcreature/updateworldstates) which can maybe somehow merged together with those equivalent functions

also codestyle needs to be improved too and some bad hacks must be removed..

if someone wants to help with this you're welcome

as i said already i won't have time for it in the nearest future

mh whats the problem?

my attempt to fix is, to rewrite those del-function, so that they look more similar to those one in battleground.cpp

but i would've done generally the same as you..

hopefully i fixed the problem with that..

no the link is right - also this works for me:

git clone  git://github.com/mangos/mangos.git
cd mangos
git pull git://github.com/balrok/mangos.git outdoor

when i checkout foreign branches i do:

git remote add balrok git://github.com/mangos/mangos.git
git fetch balrok
git branch -a
git merge balrok/outdoor

ok - from another server (0.12) came no complaints too

so i added it to outdoor and outdoor-0.12

also yad02's patch is included in outdoor and outdoor-0.12

thank you both for your help

thank you for all your work so far and good luck on your future paths

i will update this patch (but until alterac-valley is in master i won't actively do something here) - also a friend, who's developing on wintergrasp may get writeaccess then too (i don't know yet, if it's better to create a new branch for wintergrasp or not.. - currently i think it's ok to have it inside this patch too)

if someone is interested in contributing to this branch, you can pm me then and i will add you to my repository as collaborator

in future this patch can be found at:

http://github.com/balrok/mangos/tree/outdoor

http://github.com/balrok/mangos/tree/outdoor-0.12

--

@andriuspel:

you're right this is realy offtopic - but you just need to go in a library or bookstore - there you'll find what you're searching..

also i guess somewhere in this forum is actually a thread about this

so please search this thread or open a new thread somewhere else

@thyros: thx it's corrected

@cybermonk: maybe you forgot the outdoorpvp.sql (in the folder sql/outdoor/)

@splinterc:

yes, i think this suggestion will improve code readability - but your idea is currently wrong (as it seems to me) cause you don't do any checks about which team can use which trigger

i mean something like this:

if(plr->GetTeam() == ALLIANCE and trigger != SI_AREATRIGGER_A)
// aliance player on alliance-trigger shouldn't be handled
return false;

(and same for horde)

if you want, you can create a patch for it then i can include it.. else i will just keep it in mind and will do this next time when i review this patch

edit: another thing:

you don't need to check

if(plr->GetTeam())

i think this would never fail

maybe you add this in the top part of the function those should be all cases when this function can return false

if !player->hasaura(...)
return false
if trigger==bla and plr->getteam=xyz
return false
if trigger==blub and plr->getteam=zyx
return false

and then for checking the win_condition this can be used for alliance and horde

if team==ally
{ m_gathered_a++;
if m_gathered_a>123)
{
...
..
}
}

for outdor(master) i also had mergeconflicts, so i updated that too.. also i removed some wrong things from this branch (charlie changed with the last merge accidently very much - and i removed most of those changes except:

 contrib/extractor/debug/zlib.lib            |  Bin 198300 -> 0 bytes
contrib/extractor/release/zlib.lib          |  Bin 212284 -> 0 bytes

(if somebody knows how to fix it, please tell me)

also i cleaned up this patch a bit (just removing tabs for spaces, some intendentaion)

then i tried to fix the bug sarjuuk mentioned, it's about setting the tower-owners in terrokar after 6h back to neutral - i hope this patch works, cause i currently have no client to test

also i fixed the bug, that the banner-gameobject of the nortwest-tower had a wrong rotation

(this is done through the line:

-    AddCapturePoint(TFCapturePoints[type].entry,TFCapturePoints[type].map,TFCapturePoints[type].x,TFCapturePoints[type].y,TFCapturePoints[type].z,TFCapturePoints[type].o,TFCapturePoints[type].rot0,TFCapturePoints[type].rot1,TFCapturePoints[type].rot2,TFCapturePoints[type].rot3);
+    AddCapturePoint(TFCapturePoints[type].entry,TFCapturePoints[type].map,TFCapturePoints[type].x,TFCapturePoints[type].y,TFCapturePoints[type].z,TFCapturePoints[type].o,TFCapturePoints[type].rot0,TFCapturePoints[type].rot1, sin(TFCapturePoints[type].o / 2), cos(TFCapturePoints[type].o / 2));

the whole patch can be found here - it would be nice if someone can test it

http://uploaded.to/file/axewey

you can search in battlegroundmgr for "Find a random arena, that can be create", edit the lines below

and then join directly through the battlemaster to this map

also you could remove

 1521     // don't let enter battlegrounds without assigned battleground id (for example through areatrigger)...
1522     // don't let gm level > 1 either
1523     if(!InBattleGround() && mEntry->IsBattleGroundOrArena())
1524         return false;

from player.cpp and then i guess it works - but maybe it will crash then somehow.. i don't know

good luck with your patch

merged with mangos-0.12

also i have two small questions, maybe someone knows this

the server where i play, now also has the outdoor-patch running, but there cam lags with stealth-remove from players..

arrai suggested this fix but we are not 100% sure, if it's ok to just remove those lines or if we would cut off some functionality..

but maybe those lines got only accidential inside and we can realy remove them

diff --git a/src/game/SpellAuras.cpp b/src/game/SpellAuras.cpp
index b7229a9..5935ab0 100644
--- a/src/game/SpellAuras.cpp
+++ b/src/game/SpellAuras.cpp
@@ -3359,7 +3359,8 @@ void Aura::HandleModStealth(bool apply, bool Real)
                {
                    if(OutdoorPvP * pvp = ((Player*)pTarget)->GetOutdoorPvP())
                        pvp->HandlePlayerActivityChanged((Player*)pTarget);
-                    pTarget->SendUpdateToPlayer((Player*)pTarget);
+                    // by openpvp-patch introduced -> produced lags
+                    // pTarget->SendUpdateToPlayer((Player*)pTarget);
                }
            }

@@ -3432,7 +3433,8 @@ void Aura::HandleInvisibility(bool apply, bool Real)
                    if(OutdoorPvP * pvp = ((Player*)m_target)->GetOutdoorPvP())
                        pvp->HandlePlayerActivityChanged((Player*)m_target);

-                    m_target->SendUpdateToPlayer((Player*)m_target);
+                    // by openpvp-patch introduced -> produced lags
+                    // m_target->SendUpdateToPlayer((Player*)m_target);
                }
            }
        }

next thing:

can someone update the eventai.sql in outdoor_0.12?

i know to less about eventai to do this, and i think someone else here can do it in a few minutes too

@bison: sorry, i can't help with it.. i think psdb has different flags for some creatures/objects than udb.. maybe you try exporting all needed objects/creatures

from udb and apply it to the psdb (if it then will work, you can post your sql here and we can include it in the repo)

i updated outdoor_0.12.. i changed a wrong mysql statement and then merged with newest mangos-0.12, while merging there were some conflicts - and if i made something wrong the worldstates won't work correctly after this.. but i think i made everything right - and it compiles at least^^

edit: i haven't fixed the eventai.sql - maybe someone here can suggest a working one - i realy never looked at eventai yet (:

@ragnababy:

i think you need to run git commit -a or git checkout -f to tell git, what to do with your uncommitted stuff

sorry, with bad hacks i meant the original script from the first topic, sorry for the misunderstanding

ps: i don't know if this is intended by you.. but you disabled for other users to send you pms

what about a simple

while true; do nice -n 20 ./path/to/bin; sleep 5; date; done

it doesn't look like your script has more functionality your script even has bad hacks and unneeded code

@sturm you don't need to use special file extensions

linux just looks at the first byte of a binary to find out how to handle it, and txt-files with a #!/bin/bla at the beginning, will be executed as /bin/bla my_txt_file

to use file-extensions is just your choice

btw (your question looks a bit like you asked for that)

linux itself also has many shellscripts without the .sh extension for example your daemon-startscripts, many git-tools which you use are also shellscripts, the makefiles are shellscripts too

but 0 in the template means, that it will get dbc-values

hello, cause i havent found a gdb-restartscript here i will post mine, which i got through looking at: http://www.trinitycore.org/forum/f5/16line-autorestarter-gdb-tracer-t3679.html

my resulting script, which is already some days in use and seems to work fine, is:

http://mangos-files.eu/patches/startmangos

and the restartscript which is also in use:

http://mangos-files.eu/patches/mangrestart

and i post the startmangosscript here too, cause i don't know if it will be always on this server

#!/bin/bash
LOGDIR="/home/dbs/server/logs/"
LOG_LOG0="log_mangos"
LOG_ERR0="err_mangos"
LOG_GDB0="gdb_mangos"


LOG_LOG=$LOGDIR$LOG_LOG0
LOG_ERR=$LOGDIR$LOG_ERR0
LOG_GDB=$LOGDIR$LOG_GDB0

echo "run" > .core.gdb
echo "bt" >> .core.gdb
echo "bt full" >> .core.gdb
echo "thread apply all backtrace full" >> .core.gdb

while true
do
   date=`date +%Y%m%d_%H.%M.%S`
   #gdb ./server/bin/mangos-worldd -x .core.gdb --batch > $LOG_LOG".log" 2> $LOG_ERR".log"
   gdb ./server/bin/mangos-worldd -x .core.gdb --batch &>$LOG_LOG".log"
   grep -B 40 -A 1800 "Program received signal SIG" $LOG_LOG".log" > $LOG_GDB$date".log"
   mv $LOG_LOG".log" $LOG_LOG$date".log"
   #mv $LOG_ERR".log" $LOG_ERR$date".log"

   rm $LOGDIR"last_"$LOG_LOG0".log"
   ln -s $LOG_LOG$date".log" $LOGDIR"last_"$LOG_LOG0".log"
   #rm $LOGDIR"last_"$LOG_ERR0".log"
   #ln -s $LOG_ERR$date".log" $LOGDIR"last_"$LOG_ERR0".log"
   rm $LOGDIR"last_"$LOG_GDB0".log"
   ln -s $LOG_GDB$date".log" $LOGDIR"last_"$LOG_GDB0".log"
   echo `date`" restarting"
   sleep 11 # let the operating system clean up some stuff, and the tester cool down a bit 
done

cause i found no way, so that mangos still can write to stdout, and cause i don't need this feature, you have to do

tail -f /path/to/mangoslog

after starting, but then you won't have the console..

if mangos doesn't write to stdout it's also realy much faster (10seconds without stdout and over 60s with stdout)

also i don't take any warranty if this script doesn't work at your computer or if this will destroy something - please read the code first

i only post it here so that other people can get inspiration how this can be done

downforce: if you get compile errors with this patch post it in the outdoor-thread in the under-review section

your suggestion is wrong

and i compiled this patch yesterday, and everything worked fine..

mh it doesn't seem to work if i follow the readme, it still writes cmake files into the mangosroot:

balrok@webserver:~/mangos$ mkdir build
balrok@webserver:~/mangos$ cd build/
balrok@webserver:~/mangos/build$ cmake ../
.
.
.
-- Build files have been written to: /home/balrok/mangos

balrok@webserver:~/mangos/build$ cd ../
balrok@webserver:~/mangos$ make
[  0%] Built target genrev
[  0%] Built target revision.h
[  4%] Built target g3dlite
[ 11%] Built target mangossockets
[ 16%] Built target zlib
[ 26%] Built target ZThread
[ 26%] Building CXX object src/framework/CMakeFiles/mangosframework.dir/Policies/ObjectLifeTime.cpp.o
In file included from /home/balrok/mangos/src/framework/Policies/ObjectLifeTime.h:23,
                from /home/balrok/mangos/src/framework/Policies/ObjectLifeTime.cpp:20:
/home/balrok/mangos/src/framework/Platform/Define.h:24:29: error: ace/Basic_Types.h: No such file or directory
/home/balrok/mangos/src/framework/Platform/Define.h:25:28: error: ace/ACE_export.h: No such file or directory
In file included from /home/balrok/mangos/src/framework/Policies/ObjectLifeTime.h:23,
                from /home/balrok/mangos/src/framework/Policies/ObjectLifeTime.cpp:20:
/home/balrok/mangos/src/framework/Platform/Define.h:109: error: ‘ACE_INT64’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:110: error: ‘ACE_INT32’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:111: error: ‘ACE_INT16’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:112: error: ‘ACE_INT8’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:113: error: ‘ACE_UINT64’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:114: error: ‘ACE_UINT32’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:115: error: ‘ACE_UINT16’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:116: error: ‘ACE_UINT8’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:119: error: ‘uint16’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:120: error: ‘uint32’ does not name a type
/home/balrok/mangos/src/framework/Platform/Define.h:123: error: ‘uint64’ does not name a type
make[2]: *** [src/framework/CMakeFiles/mangosframework.dir/Policies/ObjectLifeTime.cpp.o] Error 1
make[1]: *** [src/framework/CMakeFiles/mangosframework.dir/all] Error 2
make: *** [all] Error 2
balrok@webserver:~/mangos$ 

the same error also comes if i do "cmake ." and then "make" in mangosroot

oh right, it's now removed