justin

i keep on getting crashes if i send clearall message to a jit.cellblock which "refers" to a coll for the data... i realise this is probably not a good idea (from a programming pov), but i accidentally came across it a couple of times until i discovered the cause of the crash.

i'm not getting any crash log either, just beach ball spinning, then i have to force quit max. can anyone else reproduce this?

system specs: osx 10.4.11 / max 4.63 / jitter 1.5

i keep on getting crashes if i send clearall message to a jit.cellblock which "refers" to a coll for the data... i realise this is probably not a good idea (from a programming pov), but i accidentally came across it a couple of times until i discovered the cause of the crash. 

jit-cellblock-and-coll-crash-reproducible

just to add that i get the same crash with max 504...

Yeah, this is really not a good idea. Would it be OK to just error 

out the cellblock "clearall" message to the Max window? If you are 

using a coll attachment, you should really be clearing the coll, not 

On Jul 29, 2008, at 10:35 AM, justin wrote:

> i keep on getting crashes if i send clearall message to a 

> jit.cellblock which "refers" to a coll for the data... i realise 

> this is probably not a good idea (from a programming pov), but i 

> accidentally came across it a couple of times until i discovered the 

> i'm not getting any crash log either, just beach ball spinning, then 

> i have to force quit max. can anyone else reproduce this?

> system specs: osx 10.4.11 / max 4.63 / jitter 1.5

> #P comment 368 120 100 196617 2. click to crash max;

> #P message 320 120 47 196617 clear all;

> #P newobj 63 123 70 196617 coll collcrash;

> #P newex 63 95 120 196617 sprintf %ld col == crash;

> #P message 240 120 78 196617 refer collcrash;

> #P user jit.cellblock 240 149 449 348 3 9 4 10 45 17 0 1 1 0 1 0 1 1 

> 1 0 0 0 255 255 255 0 0 0 0 0 0 191 191 191 0 0 0 215 215 240 1 1 1 

> #P comment 75 27 100 196617 1. bang this;

Yeah, this is really not a good idea.  Would it be OK to just error  
out the cellblock "clearall" message to the Max window?  If you are  
using a coll attachment, you should really be clearing the coll, not  
the cellblock...

>
> i keep on getting crashes if i send clearall message to a  
> jit.cellblock which "refers" to a coll for the data... i realise  
> this is probably not a good idea (from a programming pov), but i  
> accidentally came across it a couple of times until i discovered the  
> cause of the crash.
>
> i'm not getting any crash log either, just beach ball spinning, then  
> i have to force quit max. can anyone else reproduce this?
>
> system specs: osx 10.4.11 / max 4.63 / jitter 1.5
>
> see patch below:
>
> #P window setfont "Sans Serif" 9.;
> #P window linecount 1;
> #P comment 368 120 100 196617 2. click to crash max;
> #P message 320 120 47 196617 clear all;
> #N coll collcrash;
> #P newobj 63 123 70 196617 coll collcrash;
> #P button 33 26 38 0;
> #P newex 63 95 120 196617 sprintf %ld col == crash;
> #P message 240 120 78 196617 refer collcrash;
> #P user jit.cellblock 240 149 449 348 3 9 4 10 45 17 0 1 1 0 1 0 1 1  
> 1 0 0 0 255 255 255 0 0 0 0 0 0 191 191 191 0 0 0 215 215 240 1 1 1  
> 0 4 0 0 1;
> #P newex 33 68 40 196617 uzi 10;
> #P comment 75 27 100 196617 1. bang this;
> #P fasten 7 0 2 0 325 142 245 142;
> #P connect 3 0 2 0;
> #P fasten 1 1 3 0 53 119 245 119;
> #P connect 4 0 6 0;
> #P connect 1 2 4 0;
> #P connect 5 0 1 0;
> #P window clipboard copycount 9;
>


Quote: ddg@cycling74.com wrote on Tue, 29 July 2008 17:57

----------------------------------------------------

> Yeah, this is really not a good idea. Would it be OK to just error 

> out the cellblock "clearall" message to the Max window? If you are 

> using a coll attachment, you should really be clearing the coll, not 

it would better to get an error message than a crash, for sure!

i thought exactly the same about sending the clear all message to jit.cellblock, i just thought that even if it wasnt good programming it still shouldn't crash and hang.

Quote: ddg@cycling74.com wrote on Tue, 29 July 2008 17:57
----------------------------------------------------
> Yeah, this is really not a good idea.  Would it be OK to just error  
> out the cellblock "clearall" message to the Max window?  If you are  
> using a coll attachment, you should really be clearing the coll, not  
> the cellblock...
----------------------------------------------------

I get the same results with Leopard OSX 10.5.4(and using Max5.0.4).

...and here's a crash-log I get after force-quitting(in case Justin wants to send it to support).

On Jul 29, 2008, at 11:09 AM, justin wrote:

> i thought exactly the same about sending the clear all message to 

> jit.cellblock, i just thought that even if it wasnt good programming 

On Jul 29, 2008, at 11:09 AM, justin wrote:
> i thought exactly the same about sending the clear all message to  
> jit.cellblock, i just thought that even if it wasnt good programming  
> it still shouldn't crash and hang.
>
> ;)


jit.cellblock and coll crash - reproducible?