Guillaume Evrard

i got a really important patch in my application that i can't open anymore in max (with 5.0.4).

the OS X report is here : http://guiev.free.fr/report/report.txt

bug-in-5-0-4

I had a crash using the tab object when giving it entries which had argument placeholders. For example, one tab had "#1_Sampler" and another tab(in the same tab object) had "#1_Sequencer". Unfortunately, in a separate unrelated incident, my computer died so I can't test out your patch. You could try opening it in a text editor and deleting any tab objects or anything else with argument-placeholders and see if it works for you(then mail the results you find to support; make sure you save two copies for editing, etc.). Either way, you probably want to copy the crash report into a text file and submit it as well so that support can see that. (or mail them both directly to support@cycling74.com).

Best of luck, sorry it's happening to us both.

I had a crash using the tab object when giving it entries which had argument placeholders. For example, one tab had "#1_Sampler" and another tab(in the same tab object) had "#1_Sequencer".  Unfortunately, in a separate unrelated incident, my computer died so I can't test out your patch. You could try opening it in a text editor and deleting any tab objects or anything else with argument-placeholders and see if it works for you(then mail the results you find to support; make sure you save two copies for editing, etc.). Either way, you probably want to copy the crash report into a text file and submit it as well so that support can see that. (or mail them both directly to support@cycling74.com).

Best of luck, sorry it's happening to us both.


P.S. I emailed a zipped folder of both my crashing patch and the resulting crash log directly to them. You might want to do the same.

it seems that removing arguments of the only bpatcher (wich were #1 and #2) "solved" the problem. By the way i need these arguments, what can i do ?

First, you should probably try and locate which objects used within the bpatcher were using these arguments, then test to see if a new patcher with those objects were causing the app to crash. If you can isolate the problem to a single object using these arguments then(aside from mailing your findings to support) you could try to find a workaround(example, my workaround was to use umenu instead of the tab object).... actually, i just tested by creating a new patcher and instantiating a blank bpatcher, giving it the arguments "#1" and "#2", saving the patch, closing Max5, then reopening the patch and I got the crash... so I guess your workaround might need to involve not using bpatcher for now(for example use a subpatcher with the same arguments or some other encapsulation-scheme and force the user, for now, to double-click on the subpatcher to open it up). I'm not sure, but I've gone ahead and emailed support with the crash log and test patch(just a bpatcher with those arguments) to attach it as a related incident to the crash I reported. Hopefully there will be a fix soon.

(I have to run some errands for an hour or two and then I'll definitely take a further look at your patch... see if we can find a temporary solution while they work on the fix).

Best of luck and patience to both of us.


when you call a bpatcher with 1 argument set to #1, it doesn't crash.

but when you have a bpatcher with 2 arguments set to #1 and #2, max crashes.

files are attached : open argbugparent-1arg.maxpat for no crash, argbugparent-2args for a nice bug report :)

Ok, so I took a look a second look at your patch(in text) and was thinking, what if you instantiated the patch 2PIM-AD in an object box with arguments #1 and #2(for example, this would be in a new object box: [2PIM-AD #1 #2]), then put the object box in presentation mode and put a comment telling people to double-click on it to see the insides? It's a quick, cheap, and not-so-user-friendly fix, but until they actually fix the real problem, it might just have to do?

Let me see if I can get on my wife's computer, install a trial-version of Max5 and upload an example of what I'm talking about(since my own comp's hard-drive recently died and I'm waiting on apple-care)....

I've attached an example to this post. Open tester.maxpat to see the full encapsulation.

yes i could do so, but the bpatcher is a part of a complex interface, and i need to see its graphical interface. i just won't release any version of my app until the bug is fixed.

Ya, I understand and would recommend waiting for the fix. The only other way would be to include the 2PIM-AD patch within the larger patch, but then you would lose the modular ease-of-editing inherent in the encapsulation.

i came back to 5.0.3, waiting for 5.0.5... :-)

> Ya, I understand and would recommend waiting for the fix. The only other

> way would be to include the 2PIM-AD patch within the larger patch, but then

> you would lose the modular ease-of-editing inherent in the encapsulation.

> Glad to offer what little help I could.

>
> Ya, I understand and would recommend waiting for the fix. The only other
> way would be to include the 2PIM-AD patch within the larger patch, but then
> you would lose the modular ease-of-editing inherent in the encapsulation.
>
> Glad to offer what little help I could.
> --
> www.karaokaze.net
>

Bug in 5.0.4 ?