DasBlub

tested and patch still required. thanks for fixing it, this item is funny i updated the patch for the newest version (just a small cleanup): http://github.com/DasBlub/mangos/commit/e6a66fe1ae11b5dfa6611d88ef99cb44bfb6d583

not sure whether this patch is correct or not. i queried the data on UDB392: UNIT_FLAG_PVP: SELECT COUNT(*) FROM creature_template WHERE unit_flags & 4096; -> 1542 UNIT_FLAG_PVP_ATTACKABLE: SELECT COUNT(*) FROM creature_template WHERE unit_flags & 8; -> 0

patch not required anymore, such a function now exists: void AddMenuItem(uint8 Icon, int32 itemText, uint32 dtSender, uint32 dtAction, int32 boxText, uint32 BoxMoney, bool Coded = false);

this thread is now useless, the table is now in the characters-db. the table still has the 'realmid' field which isn't used anymore in the code. it can now be dropped. patch to drop it: http://github.com/DasBlub/mangos/commit/200ecd2df162883e66c9e0184b21262e86587b65

the realmd does now support all wow versions, so this patch is outdated (who had thought that with a patch from 2008...? )

this patch should not be required anymore as GOs seem to be handled different now, in GameObject::Use there's a call to Script->GOHello. i couldn't find "10852" in SD2 though, so maybe a script for it is still required (i didn't test the quest). EDIT: tested, script still required. posted @ udbtracker*: http://udbtrac.kicks-ass.net/index.php?showrev=any&quest=10852 * till 10 minutes ago i didn't knew about it either... [EDIT2] Quest 10852 has now been fixed (not yet commited): http://www.scriptdev2.com/project.php?issueid=1084

What bug does the patch fix? What features does the patch add? I updated the project files for the vmap_assembler: * Add x64 version to VC80 & VC90 * Switch from SSE to SSE2 * Wider use of variables instead of hardcoded paths I need someone to test the VC80 files and tell me whether they work or not! I tested the VC90 and VC100 files and they work fine, i modified the VC80 file by manually editing the config files, thus I don't know if it'll work. If you test it, please test if it's set to SSE2 (you can find this in the project settings under "Configuration Properties" -> "C/C++" -> "Code Generation" and there the field "Enable Enhanced Instruction Set". Thanks for testing! For which revision was the patch created? [10471] but should also work for [10474] as this one was the prev. commit in my repo Who has been writing this patch? Please include either forum user names or email addresses. DasBlub Commit http://github.com/DasBlub/mangos/commit/8baa2b7f947d123067b68c4e676065732edd4773 (for those who clone my repo to test it: it's on my new x64 branch)

i prefer a 'real' bugtracker like lighthouse as it has workflows (assigning tickets to someone, 'in work', 'waiting for answer', etc.) and is better to search it. furthermore lighthouse has less spam because you need to register on it and can't use your forum-account from here. i haven't used the github issues yet, but it looks really nice and would be useful as those who are working on mangos either way already have a github account (well, many of them) so they can use that account. but in the end, the devs have to decide, they need to work with it all the time

skynyrd, you mean CTRL+SPACE (to trigger IntelliSense)? and i prefer the this-> too, just because it's more 'clear' and prevents any naming-problems between local variables and class variables (even if i'm using m_ i might still miss one). i also use std:: instead of 'using namespace std'. i guess this is a question of style which everyone has to decide on his own.

created a patch out of it: http://gist.github.com/572267 i also remarked, that files from visual studio don't get excluded, so the .opensdf (VC100) file was also shown in my git-diff. maybe move those VS-specific .gitignore from /win/.gitignore to /.gitignore so that it also counts for all the projects in /contrib/ and /dep/ ?

i assume that it won't work, because the client relies on that map-number

you might want to submit this to mysql directly? http://forge.mysql.com/wiki/Contributing_Code

just download the Platform SDK for Windows 7: http://www.microsoft.com/downloads/details.aspx?FamilyID=c17ba869-9671-4330-a63e-1fd44e0e2505&displaylang=en then you have to set the compiler for x64 in VS (use VS2010: http://www.microsoft.com/express/Downloads/ & http://msdn.microsoft.com/en-us/vstudio/bb984878.aspx) to the compiler in the PSDK. for the db you have to use the 'mysqldump' command to dump the database and then you get a .sql file which you can run again on your new db to re-load it.

What bug does the patch fix? What features does the patch add? It adds a bit more output to Char.log Sample Output As a note: it shows no IP for the first two records because i did those commands directly from the console. For which revision was the patch created? [10326] Who has been writing this patch? Please include either forum user names or email addresses. DasBlub Patch http://gist.github.com/512275

if it would have been purposely, they'd have taken 12345 but yes, looks nice

i think it'll be better to first finish it on master and then properly backport it to 0.12 when has been commited (and i say that as a 0.12 user )

LOLCODE is the way to go!

for the last two spells (1954, 2267) you wrote == instead of !=, is this a typo?

when i implemented the no_char_delete patch, i first used a DATETIME field from MySQL. i later changed it to BIGINT(11) due to the fact, that PGSQL and MySQL don't have the same datatypes and functions for it. it was impossible to implement the checks i needed to do, because i couldn't use any DB-internal functions. and HDDs are very cheap nowadays, you can get 2TB as RAID-edition (= 24/7 HDDs) for something around 300$ and normal hdd's are even cheaper. -- DasBlub

are we now a group of sociology students? how about NOT talking about the human-factor crap in this and just talk about the technical part? Diabolix, you need two or three tables because of the 3rd normal form for databases role (id, name) - role_right (role_id, right_id) - right (id, name) a role can be "user", "admin", etc. a right can be "command .server info", "has logout delay", etc. -- DasBlub

if you're on linux, this is very easy: dd if=/dev/random of=~/mangos count=500000 chmod ugo+x ~/mangos to start it, just use this: ~/mangos how about reading the wiki / faq?

that's not what i said. i said it's useless to check again if group is, what you just set. it's the same if i'd do this: if(int x = 42) { if(x == 42) { ... } }

yes, they'll get support by ACDC i heard rumors that Alestorm, Finntroll, Eluveitie, Ensiferum and Korpiklaani would be on the stage too (at least they'll perform in my media player at that moment ) and yes, congrats to all devs who help keeping this project rolling

this patch looks a bit hacky: + if(Group* group = GetGroup()) + { + if (group == this->GetGroup()) why do you check again if it's that value which you have just set? GetGroup() is the same as this->GetGroup() also, those if's are strange, instead of something like if(a) { if(b) { ... if(z) { ...} } } you can just do if(a && b ... && z) { ... } -- DasBlub

maybe with a grid searcher? i'm not sure, sorry