Reset Password
Existing players used to logging in with their character name and moo password must signup for a website account.
- Komira 46s
- QueenZombean 3m Singing the song that doesn't end, yes it goes...
- Emily 1m I lost myself, in the dark charade.
- Vanashis 39m
- zxq 43s
- AdamBlue9000 27s Rolling 526d6 damage against both of us.
- Fogchild1 9s
- Rillem 1m Why's the weed always gone?
a Mench 3h Doing a bit of everything.
- NightHollow 4h
j Fengshui 1h
- Diamond 18s After Winter, must come Spring.
And 25 more hiding and/or disguised
Connect to Sindome @ moo.sindome.org:5555 or just Play Now

Driving Client Crash (TinTin++)

I have been having an interesting problem over the past several months where occasionally while driving and coming to a halt, shortly afterwards the TinTin++ client will crash. After some puzzling about I've determined this is due to a #Zap function being called by the client which is usually triggered by the user deliberately (this is usually like a kill function to stop running sessions one by one by name). This is the code for that function in TinTin (hopefully my code tags don't break everything):

DO_COMMAND(do_zap)

{

struct session *sesptr;

char arg1[BUFFER_SIZE];

push_call("do_zap(%p,%p)",ses,arg);

sub_arg_in_braces(ses, arg, arg1, GET_ALL, SUB_VAR

Okay so they did break, because of course they did, and ate my entire post. Just a moment!
This is the relevant code for TinTin++ from their Github, for the #zap function to kill active sessions.

I've combed through the raw client logs of 20 or so instances of this #zap being triggered while driving, and there is nothing obvious to me that is being received that could trigger some niche command or client crash (no strange characters or pseudo-code for example):

But this is still a very reliable, apparently deliberate trigger of a session kill by what is known to be a pretty stable and unflappable piece of software, so I am very puzzled what is going on. I also lack the expertise to file intelligent bug reports with the TT++ devs, so I am curious if anyone smarter than me sees something that might be falsely triggered in the code here.

Enable #debug and reproduce.
Thanks I will keep debug for commands on while driving and see what turns up.
Could be something with our color codes or something that are being sent as a part of driving?