Heya,
I am doing some testing where I read in a fixed width file where the last "row" is too short. I expected the last row to get sent to the log port (when dataPolicy is "Controlled") or cause an error (when dataPolicy is "Strict"). Instead, the graph executes, records the good rows, and acts like the last bit of data never existed - no record of data going to the logging port with "Controlled" mode, no halt of the graph in "Strict" mode. Am I doing something wrong? Or is this correct behaviour?
I am using clover 2.8.0.
My data is:
with a FMT of:
My result is 2 rows to port 0 (and nothing to port 1 with a dataPolicy of "Controlled"):
This should be a rare case for me, but it seemed odd..
Thanks,
Anna
I am doing some testing where I read in a fixed width file where the last "row" is too short. I expected the last row to get sent to the log port (when dataPolicy is "Controlled") or cause an error (when dataPolicy is "Strict"). Instead, the graph executes, records the good rows, and acts like the last bit of data never existed - no record of data going to the logging port with "Controlled" mode, no halt of the graph in "Strict" mode. Am I doing something wrong? Or is this correct behaviour?
I am using clover 2.8.0.
My data is:
001JOE SMITH 12/30/2009 1.56002MARY JONES 11/13/2009 22.74003CARRIE COOKE478.30
with a FMT of:
<?xml version="1.0" encoding="UTF-8"?>
<Record name="RECORD_true_fixed_sample_missing_data_txt_" type="fixed">
<Field name="FIELD_true_fixed_sample_missing_data_txt_CLOVER_ROW_NUM" type="numeric" nullable="false" size="10" auto_filling="source_row_count" />
<Field name="FIELD_true_fixed_sample_missing_data_txt_pcl" type="string" nullable="true" size="3" />
<Field name="FIELD_true_fixed_sample_missing_data_txt_name" type="string" nullable="true" size="20" />
<Field name="FIELD_true_fixed_sample_missing_data_txt_date" type="string" nullable="true" size="10" />
<Field name="FIELD_true_fixed_sample_missing_data_txt_amount" type="string" nullable="true" size="15" />
</Record>
My result is 2 rows to port 0 (and nothing to port 1 with a dataPolicy of "Controlled"):
[2010-01-29 15:09:58,348] INFO - INPUT_0 FINISHED_OK
[2010-01-29 15:09:58,348] INFO - %cpu:.. Out:0 2 0 0 0
[2010-01-29 15:09:58,349] INFO - Out:1 0 0 0 0
This should be a rare case for me, but it seemed odd..
Thanks,
Anna
-
Hello Anna,
it's really not proper behavior :-(. I've reported it to our bug tracking system (http://bug.cloveretl.com/view.php?id=3672).
Please sign in to leave a comment.
Comments 1