[kepler-users] Size Limitation of ArrayTokens?

[Thomas M. Parris]

If it helps, we regularly read, process and write long streams of 720x360
element double precision matrices using the PN directory without difficulty.
The workflow is rather involved, so many such arrays are in memory at any
given time.  So that's 259,200 elements and 2,073,600 bytes for each array.

That said, I realize arrays are not the same as matrices.  

-- Tom

-----Original Message-----
From: kepler-users-bounces at kepler-project.org
[mailto:kepler-users-bounces at kepler-project.org] On Behalf Of Christopher
Brooks
Sent: Monday, May 21, 2012 12:32 PM
To: Stefan Proell
Cc: kepler-users at kepler-project.org
Subject: Re: [kepler-users] Size Limitation of ArrayTokens?

HI Stefan,
A quick search of the web shows no apparent limits.
A more complete test case would help.

The way I would approach this is by splitting up the problem into first just
reading the data and being sure that worked and then adding more like
encoding.

_Christopher

On 5/21/12 9:20 AM, Stefan Proell wrote:
> Dear Mailing list,
> I wrote my own Actor which I need in order to encode binary files via
> Base64 and pass the resulting encoded String to a REST Service. I feed 
> the binary file with a BinaryFileReader-Actor to my encoding Actor and 
> cast the Token to an Array, as I found no other solution to receive 
> the bytes from the file. I then use a standard (and also deprecated) 
> method for encoding the file first to Base64 and then send it through 
> some URL-safe encoding. The method looks like this:
>
>
> @Override
>     public void fire() throws IllegalActionException {
>
>         super.fire();
>
>         // Read file from InputPort  and convert to a ByteArray
>         ArrayToken inputToken = (ArrayToken) inputPort.get(0);
>         byte[] inputBytes =
> ArrayToken.arrayTokenToUnsignedByteArray(inputToken);
>
>         // Encoding
>         String encode = new  String(Base64.encodeBase64(inputBytes));
>         String encodedString = java.net.URLEncoder.encode(encode);
>
>         output.send(0, new StringToken(encodedString));
>
>     }
>
> My problem is that the encoded file is truncated by the actor and 
> hence not usable for further processing. There seems to be a limit of 
> 20 000 array elements (bytes), which is not sufficient for my purpose.
> Does anyone have an idea why the file is chopped off after 20k letters?
>
> Thanks in advance,
> Stefan
>
> _______________________________________________
> Kepler-users mailing list
> Kepler-users at kepler-project.org
> http://lists.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users

-- 
Christopher Brooks, PMP                       University of California
CHESS Executive Director                      US Mail: 337 Cory Hall
Programmer/Analyst CHESS/Ptolemy/Trust        Berkeley, CA 94720-1774
ph: 510.643.9841                                (Office: 545Q Cory)
home: (F-Tu) 707.665.0131 cell: 707.332.0670

_______________________________________________
Kepler-users mailing list
Kepler-users at kepler-project.org
http://lists.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users