I was trying to use that context to distinct data from the input file, and to use it to give names to the splited files

Hi,
You have  to create several output, don't think you can do it dynamically 'cause I don't know if it s possible to retrieve column name in talend.

Yes of course but he also wants to name his files using the column's name.

Ok sorry didn't get the point !

shong wrote:

Hi
Not need to use context variable, just iterate each row and set a dynamic output file path. The job design looks like:
tFileInputDelimited--row1(main)--tFlowToIterate--iterate--tFixedFlowInput--main--tFileOutputDelimed

on tFixedFlowInput, generate each row of input data, define two columns the schema, name and result.
key:value
name:(String)globalMap.get("row1.name")
result:(String)globalMap.get("row1.result")

on tFileOutputDelimited, set a dynamic output file path, for example:
"D:/file/"+(String)globalMap.get("row1.name")

Check the box 'append' on tFileOutputDelimited.

Best regards
Shong

I have a problem just like this one but this solution does not work for me as I am working with huge files 1Gb+ , iterating with each row reduces the performance.
I am looking for a more dynamic way for this kind of process. Do you have any idea about how can I do it without iterating?