Founded Code for Fix Callvotebug (JKA), compile to .so?

Eragon · **Posted:** 05 Apr 2009 16:27

Hi aluigi!

I Found the Code for fixing the Callvote bug problem:
http://bugzilla-attachments.icculus.org/attachment.cgi?id=1723

But i need that in a .so file... can you help me?

Best Regards
Eragon

aluigi · **Posted:** 05 Apr 2009 20:03

it's a "diff" file referred to the game source code so you can't build a shared library from it.

Eragon · **Posted:** 05 Apr 2009 23:25

So there is no way to become this bug fixed...?

aluigi · **Posted:** 06 Apr 2009 10:56

everything usually can be fixed but in this case I don't see a quick (simple to write), easy (to implement) and universal (working on any linux executable) solution.

the simplest idea which could come in my mind can be for example adding (maybe through dll injection) a fake Cbuf_ExecuteText where implementing the same patch I wrote for Windows, then calling this function instead of the real one from SV_GameSystemCalls and then calling the real Cbuf_ExecuteText one when the fake one terminates.

Eragon · **Posted:** 06 Apr 2009 19:32

So is it possible that you can do anything? Because thats one level to high for me at the moment >.<.
A fix for that would bring a lot.....

Ty!

aluigi · **Posted:** 06 Apr 2009 22:40

no, I'm not interested and I'm not sure of the results.
then for JA the JA+ mod seems to solve the problem so I don't see problems

Eragon · **Posted:** 08 Apr 2009 00:00

Well, its for a BASE server which should stay base... its a popular server, but this found new bug is getting on my nerves.. and i need vote on my server

evan1715 · **Posted:** 26 Apr 2009 14:53

i dont understand the question...
u want a base mod that fixes bugs? o_O

what mod is the server running?

quandary · **Joined:** 01 May 2009 00:43 **Posts:** 4

Quote:

Index: trunk/code/game/g_cmds.c
===================================================================
--- trunk/code/game/g_cmds.c (revision 1288)
+++ trunk/code/game/g_cmds.c (working copy)
@@ -1239,7 +1239,8 @@
trap_Argv( 1, arg1, sizeof( arg1 ) );
trap_Argv( 2, arg2, sizeof( arg2 ) );

- if( strchr( arg1, ';' ) || strchr( arg2, ';' ) ) {
+ if( strchr( arg1, ';' ) || strchr( arg2, ';' ) ||
+ strchr( arg1, '\n' ) || strchr( arg2, '\n' ) ) {
trap_SendServerCommand( ent-g_entities, "print \"Invalid vote string.\n\"" );
return;
}
@@ -1404,7 +1405,8 @@
trap_Argv( i, &arg2[strlen(arg2)], sizeof( arg2 ) - strlen(arg2) );
}

- if( strchr( arg1, ';' ) || strchr( arg2, ';' ) ) {
+ if( strchr( arg1, ';' ) || strchr( arg2, ';' ) ||
+ strchr( arg1, '\n' ) || strchr( arg2, '\n' ) ) {
trap_SendServerCommand( ent-g_entities, "print \"Invalid vote string.\n\"" );
return;
}

They forgot \r *lol*

aluigi · **Posted:** 01 May 2009 02:01

I don't know of what mod is this fix (not ioquake3) and so if have been applied other fixes somewhere else but the strange thing is that the proof-of-concept for testing the bug uses just the \r char so how is possible that just this char has been forgotten?

(edit: in case someone is interested in the ioquake3 fix: http://svn.icculus.org/quake3/trunk/cod ... f_format=h)

Luigi Auriemma

Founded Code for Fix Callvotebug (JKA), compile to .so?