How to define mapping from multiple files to Oracle Tables in 9i

Question

Around 100-200 Flat files are created every 30 minutes and each filename is different - Filename has datetime Stamp as part of the file name apart from the product code as first 12 characters.
Can anyone guide me in How to define mappings to these files using OWB ?
What I can do is consolidate all files into one known single file name and map the files to Oracle tables which I don't want to do because I need to reject errorneous files.
Can anyone provide me some tips on this ?
Thanks in Advance.
Sohan.

Answer 1

As you know, in OWB you need to define the flat file source in a 'static' way (name, location, etc. have to be defined previously), so you cannot deal directly with dinamically generated file names. One solution would be to consolidate them into a single file (which you can define statically in OWB), but prefix every record with the filename. In this way it is easy to understand from which file the rejected records came from. If you are using unix, it is very easy to write a script to do this. Something like this will do:
awk '{printf "%s,%s\n",FILENAME,$1}' yourfilename >> onefile
where yourfile is the name of the file you are currently processing, while onefile is the name of the consolidated file. You can run this for all files in your directory by substituting yourfilename with * .
You can then disregard the file name field in OWB, while processing the rejected records based on the file name prefix by using unix utilities like grep and similar.
Regards:
Igor