Hi John, Some news on the debugging yet? I managed to deploy blend4j in my local maven installation, so now I can test the client.setChunkedEncodingSize(CHUNKED_ENCODING_SIZE); line on my setting. Unfortunately with ChunkedEncoding enabled, it seems that the file is not (recognized by || sent to) galaxy. this is the errormessage: galaxy.web.framework ERROR 2013-09-10 16:06:15,688 Uncaught exception in exposed API method: Traceback (most recent call last): File "/opt/galaxy_testing/lib/galaxy/web/framework/__init__.py", line 189, in decorator rval = func( self, trans, *args, **kwargs) TypeError: create() takes exactly 3 arguments (2 given) 10.79.24.114 - - [10/Sep/2013:16:06:15 +0200] "POST /galaxy_testing/api/tools?key=1234567890abcdef1234567890abcdef1 HTTP/1.1" 500 - "-" "Java/1.7.0_25" Do you know this problem, or could there also be something wrong with my config? Thanks for the fast reply! On 9 September 2013 19:51, John Chilton <chilton@msi.umn.edu> wrote:
Glad to hear blend4j is working well, annoyed to here it is not streaming files out by default.
Based on your ideas I started some work here, but it is not right. I spent the last half an hour looking around Galaxy trying to figure out why history creation via the API was broken, not realizing it was because I called setChunkedEncodingSize on the client.
https://github.com/jmchilton/blend4j/commit/22608de210f805fc192c86a836ea55c7...
I have created an issue in github for this:
https://github.com/jmchilton/blend4j/issues/5
Feel free to try to hack on this some more, it might actually take some non-trivial refactoring to get client.setChunkedEncodingSize(CHUNKED_ENCODING_SIZE); to be set conditionally somehow. Right now all requests are sharing the same underlying client object.
I will try to work on this more as I have time.
-John
On Mon, Sep 9, 2013 at 10:58 AM, Eric Kuyt <eric.kuijt@wur.nl> wrote:
Hi all,
I'm using blend4j for some time now and I'm quite happy about it, but now I run into some OutOfMemory issues. When I try to upload large files to galaxy (I know http might be not the best option) some byte array is getting to large. A possible solution would be file chunking in jersey, is this possible in blend4j and how could I implement this feature?
Thanks,
Eric
-- Central Veterinary Institute of Wageningen UR (CVI) Department of Infection Biology PO box 65, 8200 AB Lelystad, NL Visiting address: ASG, Edelhertweg 15, 8219 PH Lelystad Tel: +31-(0)320-293391 Fax: +31-(0)320-238153 E-mail: eric.kuijt@wur.nl Web: http://www.cvi.wur.nl