VladimirMangos

Not point to trinity, lol. Originally patch (maybe only similar) has been writed and suggested at this forum and then i rewrite it for working state and post at this forum also. Main reason that it not has been added in repo in referenced time because it will add some slowdown to code work for most wide cases when this not need (not need have diff GMs at realms). When you talk at some feature you must understand that in most cases feature add not only possibilities but also problems. Slowdown in code work and more resources use most common cases. Not all features bad sides can be avoid just disable its by some option, because sometime related code can't be guarded and excluded from use in this way. So need always compare good/bad sides. Maybe related feature more wide expected that i am think and can be added to repo. I still not sure about.

it possible use some flags and updates field directly set that different in 1.12.1 and 3.x

This is known ACE bug in ACE configs. Showup random at startup. Someting in wrong memory free at config structures destroy deep in ACE code. If someone fix it wiil be fine but this is _not_easy_ task. So need really good dev. skills for find bug in related ACE code.

mangos different from normal open source projects. Many points already has been provided lot times in years. Oh... ok. But can be repeated. What we have in normal open sources projects: some final release that have at some point complete set features. So most users happy with release until next release ready. They mostly ignore development branches. This is possible because features selected by project team. What we have in mangos. It's _always_ in beta state without some complete set features. Features list set _not_ mangos team. And users look at this known features set also. In result any developement branch better from user point any previouse release because have more _expected_ features. So creating any special _public_ development branch will just meaning all will use it and ignore release btanch. So what reason in diff branches? Mangos team _have_ development branches in your meaning for really deep changes req. for next version support. 400 branch is good example. For less deep changes as i point above have branch just useless lost team time.

It can be and _valid_ case: when player (not group leader) have persistant binding to instance and group not have yet.

why you use 2 realmd DB? for 2 realms you need single relamd and related realmd DB. So users at both realms will have different characters but same accounts

Hmm, Lynx3D sometime ago implement on-hit item proc from special attacks in another way

In [10886] fixed possible source of crash. Please, report if this help.

Just clarify: i not meaning array of stacks of movegens. I not see where this can be need... I think more simple fixed size movgen pointers array can work fine. Don't known cases when similar type (same priority) movement can be need stacked instead replaced. In array we have have specialized slots for default movegent (or idle), scripted activated for long time (for escort/long walking case), fighting chase (including ranged if will created), confused, fear slots, short movement (knock, etc) Each slot expect have single movgen that prio (like current way work) above all < slots or NULL if this type movgens not active.

confused movement is random, we not need exactly selected points (or near search), we can just skip any problematic.

getNearPoint will not return LOS-fail point.

https://github.com/cipherCOM/mangos/tree/cmake-0.12 cipherCOM mostly complete cmake use for mangos build. But small still existed problem/limitations not let yet switch to it use currently. Release mode look like complete. Problems related to debug from IDE mostly.

For release, or you add for debug version also? [added]Oh, sorry, i see you already answer to this question in post ^^

Russian law possible more liberal that in Europe, for example it explicitly grand some rights like right modify legal owned program copy "for better compatibility" or same as in europe do reverse engineering in educational purposes, and declarete any license aggriments working only in part that not affect this rights. Ofc, i not law specialist and this my personal; conclusion from reading some time ago. So as i understand you can accept license but ignore conflicting with law part rules.

Is this need for any known FreeBSD version ?

Originally expected that any assign must use VALUE verison, any check use non-VALUE version, this need also for example becase this VALUE can used in some calculations and we must be sure that result will still invalid.

In [s0501]. Thank you! In fact in cases commits without hard conflicts you can just provided short porblem description and commit hash. I use nice script that do prepare backported mangos commit very easy (resolve some problems with git cherry-pick use in application to mangos sources) writen by freghar.

This don't must be case Someone break expected height check in matser also ( INVALID_HEIGHT must be by original design used _only_ for checks. in past exist special greter by abs() value mark real invalid height... Look at still exist vmap mirror: #define VMAP_INVALID_HEIGHT -100000.0f // for check #define VMAP_INVALID_HEIGHT_VALUE -200000.0f // real assigned value in unknown height case

Yeah, I know, if I will gather info about it i will post judgements cooldown patch separately. Right now please review the main patch, as in title - allow processing cooldown for auras applied on all units. It's like closed door for further research. I not see how adding this help in research... also it anyway skipped in proc handlers.

Exist another case when this need do also (maximize skill at level update by config option setting) In [10878] with this addition. Thank you

zergtmn can be another good case for dev pos. He suggest lot small and not so small high quality patches that can be directly go to repo. And many also missed repo (base at irc suggestion, his own repo commits) posisble just because slow process in its adding without commit right. He present and active some time at #mangos and look like can be trusted for team membership.

in [10874] Thank you

I not understand only why files in archive with windows line ends

Not very strong point: description often not discribe all details in work.

You miss another part: at adding to official repo support this changes compatibility just move from 3rd party creators to team and resolving all problems also because team can't ignore this problems after adding anymore. So ofc for 3rd party creators and users all move to nice way all except team members that now need fix all bugs and etc. Ofc, we know that community will provided good support in this but anyway this will let less time for our current projects also important from _our_ point for mangos. This not meaning that any patch must be ideal, in fact i more often partly rewrite patch before adding to report that add it as-is, but need some expection that added patch will not eat all next time for weeks in fixing (weeks just example ). But in most cases patch not added because it not reviewed because lack of time for someone review it or lack need info for sure that sugested way correct in general. This most sad and often case.