I am trying to split a very large csv-file for further processing. I currently read the input, count the features, divide by a set number of max lines, round to integer and use this attribute as a filename to write new files with a feature writer.
Somehow, the attribute is calculated correctly but using it as a filename results in a maximum number of 100 files being written. The feature writer then starts overwriting the first file again.
Am I missing something?
There are several possible solutions, but the easiest is perhaps to simply use the GroupCounter from the FME Hub: https://hub.safe.com/publishers/lars-de-vries/transformers/groupcounter
You can then use the resulting attribute _group for the fanout.
There are several possible solutions, but the easiest is perhaps to simply use the GroupCounter from the FME Hub: https://hub.safe.com/publishers/lars-de-vries/transformers/groupcounter
You can then use the resulting attribute _group for the fanout.
The attribute is calculated correctly. Only the fanout doesn't work within the featurewriter.
The attribute is calculated correctly. Only the fanout doesn't work within the featurewriter.
That is surprising and unexpected. Can you show us some screenshots of how the fanout is configured? And have you double-checked that the fanout value is indeed correct for values > 99?