I have added it to manage an pool of stream, it works fine.
public class AEDFileWriter {
private Map<String,FileOutputStream> fosPool= new HashMap();
public void writeLine(@Body String data, @Property("OutputFileName") String
OutputFileName, @Header("CamelSplitComplete") boolean done) throws
IOException {
// Get the outputStream
FileOutputStream fos = fosPool.get(OutputFileName);
// Create it if not exist
if (fos == null) {
fos = new FileOutputStream(new File(OutputFileName));
fosPool.put(OutputFileName, fos);
}
// Add data
fos.write(data.getBytes());
// Eventually close the pool
if (done) {
fos.close();
fos = null;
fosPool.remove(OutputFileName);
}
}
}
My POC is finish, camel can process ( on my laptop ) 1200 line of CSV per
second and this number is growing via threading.
I will put on gitHub the source of this project I think it could be help
full for lot of camel beginner who need to deal with giga bytes of file !
Thanks
JF
2013/3/8 Jean Francois LE BESCONT <jflebesc...@gmail.com>
> It works fine with the simple example that is present but in a seda queue
> executed with concurrentConsumers the bean throw an exception due to
> concurrency ...
>
>
> 2013/3/8 Jean Francois LE BESCONT <jflebesc...@gmail.com>
>
>> OK !
>>
>> I have create a jira : https://issues.apache.org/jira/browse/CAMEL-6147 ( my
>> first :)
>>
>>
>> By the way the solution is a route like this :
>>
>> from("file://C:/Temp/camel/input_test/?noop=true")
>> .setProperty("OutputFileName" ,
>> simple("C:/Temp/camel/output_test/${headers.CamelFileName}"))
>> .split()
>> .tokenize("\n")
>> .streaming()
>> .bean(MyFileWriter.class, "writeLine")
>> .end()
>> .end();
>>
>> With a MyFileWriter like this :
>>
>> public class AEDFileWriter {
>>
>> private FileOutputStream fos;
>>
>> public void writeLine(String data, @Property("OutputFileName") String
>> OutputFileName, @Header("CamelSplitComplete") boolean done) throws
>> IOException {
>>
>> if (fos == null) {
>> fos = new FileOutputStream(new File(OutputFileName));
>> }
>> fos.write(data.getBytes());
>> if (done) {
>> fos.close();
>> fos = null;
>> }
>>
>> }
>> }
>>
>>
>> I am closed to the end :)
>>
>>
>> 2013/3/8 Claus Ibsen <claus.ib...@gmail.com>
>>
>>> On Fri, Mar 8, 2013 at 2:39 PM, jeff <jflebesc...@gmail.com> wrote:
>>> > the more performante way looks to do :
>>> >
>>> > from("file://C:/Temp/camel/input_test/?noop=true")
>>> > .split()
>>> > .tokenize("\n")
>>> > // Business lock with possible reject / enrich
>>> etc ...
>>> > .streaming()
>>> >
>>> > .to("stream:file?fileName=C:/Temp/camel/output_test/out.csv")
>>> > .end()
>>> > .end();
>>> >
>>> > But the stream is not closed ... Please help, I have to finish my
>>> proof of
>>> > concept ! :)
>>> >
>>>
>>> We should improve the stream:file to be able to auto close when it
>>> detects the splitter is done, as we have a completed property on the
>>> exchange to tell us.
>>>
>>> Fell free to log a JIRA ticket, then we can add an option on stream:file
>>>
>>> "stream:file?fileName=C:/Temp/camel/output_test/out.csv&closeOnDone=true"
>>>
>>> Just have to figure out a good name for the option.
>>>
>>>
>>> > Thanks
>>> >
>>> > JF
>>> >
>>> >
>>> >
>>> >
>>> >
>>> > --
>>> > View this message in context:
>>> http://camel.465427.n5.nabble.com/Write-enriched-data-after-a-split-in-a-file-tp5728761p5728829.html
>>> > Sent from the Camel - Users mailing list archive at Nabble.com.
>>>
>>>
>>>
>>> --
>>> Claus Ibsen
>>> -----------------
>>> Red Hat, Inc.
>>> FuseSource is now part of Red Hat
>>> Email: cib...@redhat.com
>>> Web: http://fusesource.com
>>> Twitter: davsclaus
>>> Blog: http://davsclaus.com
>>> Author of Camel in Action: http://www.manning.com/ibsen
>>>
>>
>>
>
