Cycling 74 » Topic: [text]-bug?

[text]-bug?

FRid — Wed, 23 Mar 2011 21:24:53 +0000

I think i may have found a bug or at least some very weird behaviour using the [text]-object. Decided to give this a new thread because it might be overlooked otherwise. Anyway if you want to know the history this discussion led to the discovery:

http://cycling74.com/forums/topic.php?id=26375

I’m using the filein to load 7000+ ints into [text]. This is all working fine except when i wish to dump these ints for further processing then values get lost. I’ve used another [text] to capture the output.

Any help is ofcourse greatly appreciated!

FRid

– Pasted Max Patch, click to expand. –

Re: [text]-bug?

11olsen — Sun, 27 Mar 2011 10:48:18 +0000

anyone of the max team?
is there a limit when passing lists?

Re: [text]-bug?

FRid — Sun, 27 Mar 2011 12:38:36 +0000

As for as i know (list-)limits are gone since max5 but i don’t know for sure. And indeed i was also hoping someone from the max-team spotted this (or anyone else who knows about it).

FRid

Re: [text]-bug?

jvkr — Sun, 27 Mar 2011 13:56:12 +0000

This patch can be brought back to essentials like this. Note by the way that the last number in the text object is 10, which represents the first two digits of the value 1041.

– Pasted Max Patch, click to expand. –

To work around this limitation anyway, you could use a counter and induce a cr (carriage return) every so many values.

Re: [text]-bug?

FRid — Mon, 28 Mar 2011 08:08:10 +0000

Thanks for clearing that up. Bummer, i was hoping for a bug :). I wonder how this works with jit.spill/fill though, they too pass their messages as lists.

Thanks for your suggestion but this will not really help in the setup as it is now since it needs to pass the list at once. I guess it’s back to the mxj-drawingboard. :(

FRid

(just curious; why are there (still) limits to this. Somehow connected to the same limitations existent in ints/floats?)

Re: [text]-bug?

11olsen — Mon, 28 Mar 2011 20:51:55 +0000

hey frid, the cr solution doesn’t help with fileout. maybe you could really collect single ints in an internal array. then it could take another command like “write” or so to confirm everything has been send and than write the file. too bad that rather simple things so often end up in workarounds with max.

Re: [text]-bug?

FRid — Mon, 28 Mar 2011 22:08:42 +0000

“maybe you could really collect single ints in an internal array. then it could take another command like “write” or so to confirm everything has been send and than write the file.”

This is exactly the idea, yes. However as you’ve mentioned it’s always easier said than done. For example i’ve been testing the principle (create array, store incoming ints, bang to print in the maxwindow) and it does compile however when i test it i get errors in the maxwindow?

Anyway thanks for your suggestion and gradually we’ll get there…

FRid

//Here’s the java-code.

import com.cycling74.max.*;

public class arrai extends MaxObject
{

int [] bitt;
int count = 0;

private static final String[] INLET_ASSIST = new String[]{
“inlet 1 help”
};
private static final String[] OUTLET_ASSIST = new String[]{
“outlet 1 help”
};

public arrai(Atom[] args)
{
declareInlets(new int[]{DataTypes.ALL});
declareOutlets(new int[]{DataTypes.ALL});

setInletAssist(INLET_ASSIST);
setOutletAssist(OUTLET_ASSIST);

}

public void bang()
{
int b;

for (int i=0;i b = bitt[i];
post(“b”);
}
}

public void inlet(int i)
{
count+=1;
bitt[count] = i;
}
}