sart13

And sart, you should try pushing MR2 commits to master and keep the project alive

begin again the old and silly attacks "you do not help mangos"? Yes, I can not do it - or we must pull up mangos for about six months to a year of development, or R2 to roll back the same amount. such as the ongoing development of IOC - how its push into mangos, if there is not as much as 2 levels of the subsystems necessary for it to work? I do not mention about the regular removal of my posts here.

sart13, clearly for me you not for help at forum but just for trolling. Grow up own ego by bashing project?

I am not "bashing" the project, he quietly die without my intervention. prior to your departure from the project still had a chance, now I am afraid it is not relevant. Unfortunately, the endless death of the mangos project means death (perhaps a little lazy) spent me the effort... so rejoice in any case there is nothing.

Vladimir has retired?

If you do not know, it was formally over 4 months ago. and in fact - more than six months. more in Mangos are no active developers in general ...

But after testing I found a problem. I took a level 2 paladin to a trainer and was able to learn all spells upto 'Blessing of Kings', that requires level 22.

https://github.com/blueboy/portal/commit/ae1d5be88f6b3bd6a8bffbbee6d244d81c37c17c

I have pushed a revison to portal to test out. I checked my server and the paladin can now only learn 'Devotion of Aura' as expected.

Hope this helps

checked without bot - same problem. not need big code rewrite after changes like this - problem in another place.

With this help i got one of the errors solved, but now i forund another one:

9>..\\..\\src\\game\\playerbot\\PlayerbotAI.cpp(7555) : error C2660: 'Player::GetTrainerSpellState': Function doesn't accept 1 arguments

The error-message is freely translated, cause i'm german.

I hope you can help on this one.

I really love your work with the playerbot, and i really want it in my game

https://github.com/mangosR2/mangos/commit/9237a89346b09ae4a04830e47c675e423660a56a

As far as I can tell, has always existed. mention there are at least starting with 2.0.1

in this case your must manually insert record for each user at beginning. use REPLACE instead UPDATE.

I make fair draft of "heartbert resist" method.

https://github.com/mangosR2/mangos/commit/1fdadbea569a4b0b5fa6d68dea8e673350621acb

any views and suggestions are welcome. now method in live test.

If I was to create a server from mangosR2 code, under what circumstances would this crash occur?

1. working threads in server >2

2. online >200 (approx)

3. active bot usage.

this crush impossible reproduce without mtmaps environment (parallel map processing). may be appeared after teleport master to other map, probability 0,000(period).

now no mmaps in mangos, but mmaps almost full ready to implement in master.

I do not see any reason to develop a cyberium version - it is still based on the TC code (~50%), this dead-end direction. reply to angeleyes was due to the fact that he published here my implementation bugs (there are many, although for the most part already polished implementation.)

I have long not criticizing anybody - no sence. sorry if you do not understand ...

constructive lessons your may give in R2 branch. you can offer the best option?

I have no desire, and I can not convince anyone of the correctness and necessity made me the code. like - use, do not like - forget it.

@angeleyes, my version of DF (fully different with cyberium's) not discussied here...

in most cases blizz use auras for this. example - Saurfang's blood beasts.

solve of problem your may find in R2 (commit mr1355). this not solve of source problem (not been studied in detail), but removal of consequences.

This aura does not appear to be in places that are not listed in Aura requirements, checking on it will go into the spell::CheckCast. In these places the order of removal and imposing aura is determined by stacking. If you do not - want to implement check (but not hack like this). Sorry, but the code of m-one I did not look very long ...

is nonsense. problems of aura stacking are solved in another place and very different methods ...

1. PersistentAura has your own update method

2. handle aura by published way wrong in basic. need common solution.

3. all this - wrong way. first effect of 45042 must summon creature-trigger, on his must be casted 45043.

perhaps we still speak different languages:) I can not understand how my desire to do what I'm wondering, can interfere with "The MaNGOS Foundation believes it is a fundamental right for everyone to learn without restriction or reprisals". I also hard to understand why you think that I was separated from the community - rather, it is very few remaining active members of the mangos team (team != community) were separated, and are far behind me (or rather from us). I am always ready to learn something, unfortunately something new this year came from members of the mangos team a few times. not be called a new and clever those stupid things out - for whom we have quarreled with Schmoozerd.

so you go in probably the last existing actively updated brunch mangos, share ideas, "to give others the opportunity to learn. That is why we are supposed to be here".

"here" - not exclusive github/mangos. github/mangosR2 - also "here".

I do not really understand what you want me to say. that the code does not match R2 mangos coding standarts? this is not true, in fact, commit mangos last three months to meet these standards, much less than many commits R2. "do the above mangos" - quite funny, if you look carefully, on the home page was listed R2 - "R2 - not a separate project! only custom fork of mangos project".

difference is not at all, how a developer decides to handle the rejection. difference is that the R2 team is doing something at all, and, moreover, seeks (and finds) the necessary code from third-party developers on their own, rather than waiting until it will bring on a silver platter. and in any case, I will never smear snot on the table and that "the patch is not presented for adoption and not finished." if the code is required - I can finish it himself.

http://pastebin.com/H7X8H4cJ

the problem is not in haks or workarounds, and certainly not in the "mangos standards", but in a simple lack of developers, or their unfounded snobbery. ideas and code are ready to actively offer the project when it develops and is ahead of other projects. when he was lagging behind other similar - very few people need it.

It is just annoying to have simple and reasonable patches waiting for a long time.

I did not annoying. It has long been accustomed to.

may be more simple update your GCC?

nope - zps solution is correct

~SELECT_FLAG_IN_MELEE is something different than SELECT_FLAG_NOT_IN_MELEE

in this case need another definition.

PS probably GoogleTranslate gives a very bad translation:) When I wrote about "at least 1 commit per week," not what I mean:)