Create a Job (not a transformation), and use the Unzip file
with regex to unzip all the files in a temp directory. On the Advanced tab
click Add extracted file to result
, to save the list of the extracted file internally.
Then let the job execute a transformation whose first step is a Get row from results
, to get the list of file names in a field, and give the flow to the Get data from XML
. On that step specify you want to get the file name from the previous step.
Then go back to the Job and put a Delete file name from result
. Make sure the arrow is green, so the extracted XML will be deleted only if it was read successfully.
Like that, you tmp directory will contain the file that where not read. Pretty easy to maintain on the long run. Especially if on the Unzip file
, you check the option that automagically appends the date on the extracted xml.