lekkereten

Ok, I'll commit that.

Added on r16919 Please tell me if it is what you suggested

Here: TranslatedROSTRviewer.zip mirror: http://www.mediafire.com/?dx7hr0077d6vjhe

[status]: Read information about 1 subnetworks. This could be useful only when subnetwork > 1 [info]: Initializing random number generator. Only display on one console. +1 to Brian suggestion.

I guess no one is going to implement support for PostgreSQL so yes +1 for removal.

Mr Euphy had this awesome idea on this topic regarding to a request for modification in monster script command. The request was to deprecate the event label parameter function as size/ai replacement and add two parameters (size and ai). What do you think about implementing it?

+1

When the crash happens is when the conditional is true or false?

In date.c only 4 out of 9 functions are being used. In the following functions: date_get_year, date_get_month, date_get_hour, date_get_min, date_get_sec, the only one which it is used is date_get_day to later be called by is_day_of_sun, is_day_of_moon, is_day_of_star which will be called on other files. My suggestion is that the functions that are not in use to be removed. Otherwise a macro could be made. Something like that: #define date_get_sub(tm_element) { \ time_t t; \ struct tm* lt; \ t = time(NULL); \ lt = localtime(&t); \ return lt->tm_element; \ } int date_get_year(void) { date_get_sub(tm_year + 1900); } {...} Also, it reminded me about Show* functions in showmsg.c. And with Euphy's script-shortener spirit as inspiration, this could be rewritten with a macro: #define ShowSub(msg_const) { \ int ret; \ va_list ap; \ va_start(ap, string); \ ret = _vShowMessage(msg_const, string, ap); \ va_end(ap); \ return ret; \ } int ShowMessage(const char *string, ...) { ShowSub(MSG_NONE); } int ShowStatus(const char *string, ...) { ShowSub(MSG_STATUS); } int ShowSQL(const char *string, ...) { ShowSub(MSG_SQL); } {...} Now, I don't know if the usage of macros is a good programming practice, it's up to you to decide if this is a good thing that I suggested. Bonus: And do you guys think that it would be worth using the restrict keyword, since pointers are everywhere in the code, and I just thought that it could help optimize it. Example: Thanks.

This or this should do the work.

Do you have a screenshot of what that is?

Go to src/map/map.c Find: if (battle_config.debuff_on_logout&1) { Add after: status_change_end(&sd->bl, SC_FUSION, INVALID_TIMER); This will remove the status when someone logs out

What do you mean? To disable the skill for being used or remove the status or something?

I was looking at the monster command and some lines¹ gave me the idea to merge killmonster and killmonster2 into one function, and use killmonster2 as duplicate of killmonster, just like monstersmall/monsterbig and monster. Patch: killmonster(2)_merge.patch [¹]

OH f*ck me I totally forgot that. And sorry for the mistakes, I did the parameter modification in a hurry. Hmm, I thought the sizes were right, I was using this as my reference: enum size { SZ_SMALL = 0, SZ_MEDIUM, //1 SZ_BIG, //2 }; @edit Should work. monster_patch.patch And you should suggest this as the official behavior, it's a great idea

Yeah, you should use: packet_ver: 28 in packet_db.txt and change the packetver like dastgir wrote.

Sure. Can someone try to say if its working? I can only do that tomorrow. New patch with unsigned type vars and syntax proprosed by OP and Euphy. monster_size_ai_syntax_nwe.patch

Please test. It's just a copy of Ind's INF2_NOLP. Go to src/map/skill.c: Find: if( !(skill_get_inf2(group->skill_id)&(INF2_SONG_DANCE|INF2_TRAP)) && map_getcell(bl->m, bl->x, bl->y, CELL_CHKLANDPROTECTOR) ) Change to: if( !(skill_get_inf2(group->skill_id)&(INF2_SONG_DANCE|INF2_TRAP|INF2_NOLP)) && map_getcell(bl->m, bl->x, bl->y, CELL_CHKLANDPROTECTOR) ) Go to src/map/skill.h: Find: INF2_NO_ENEMY = 0x1000, Add after: INF2_NOLP = 0x2000, Go to db/skill_db.txt: Find: 516,0,8,4,-1,0x2,3,10,1,no,0,0,0,weapon,0, GS_DESPERADO,Desperado Change to: 516,0,8,4,-1,0x2,3,10,1,no,0,0x2000,0,weapon,0, GS_DESPERADO,Desperado Remember to recompile your emulator. I can't test, ...your emulator is kinda old.

This ain't a well-made code, but it is working so far. Here's the explanation of the behavior. Go to src/map/skill.c Find: req.amount[i] = skill_db[j].amount[i]; Add after: if (skill == ASC_EDP) { j = pc_search_inventory(sd, 678); if (j < 0) { j = pc_search_inventory(sd, 608); if (j > 0 && sd->status.inventory[j].amount >= 1) { req.itemid[i] = 608; req.amount[i] = 1; } } } Change 608 for the item you desire to be used as an alternative to EDP in j = pc_search_inventory(sd, 608); and req.itemid = 608; Change 1 in sd->status.inventory[j].amount >= 1 to how many 608 it will be required. Do it in req.amount = 1; as well. Recompile and enjoy.

I've updated the patch to the latest revision. Try with it. Here: extend_vending.patch

You can try this. @all I've updated the diff to r16857. I did not test. Here: extending_vending_16857.patch

I'm not much of a client guy, but it seems to be an incompatible version between client/server. Post what version of hexed you're using, and the packet_db.txt configuration as well packet ver from common/mmo.h

Could you please test this patch that I've done? The syntax should work like: *monster "<map name>",<x>,<y>,"<name to show>",<mob id>,<amount>{,<size>, <ai>, "<event label>"}; <size>: 0=small, 1=medium, 2=big <ai>: 0=off, 1=on There should be no trouble compiling, since I did it myself and got no errors nor warnings. However I can't test at the moment, which is why I asked you to do so. Patch: monster_size_ai.patch

You could make a *little* optimization on your code. unsigned int base_penalty = 0; if (battle_config.zeny_penalty > 0 && ||>> sd->status.zeny > 0 && <<|| !map[sd->bl.m].flag.nozenypenalty) <-------------------- { base_penalty = (unsigned int)((double)sd->status.zeny * (double)battle_config.zeny_penalty / 10000.); pc_payzeny(sd, base_penalty); }

Tried both common leveling and @baselevelup?