I had this prod issue where we FTP a file that came in from different system had unaccepted characters and the program reading it has failed. I had taken a mainframe approach before as below
- From the error output, identify the field where we have the issue data
- Using sort, separated all issue records into different file, by using position of issue character
- Corrected the separated file manually (or using sort) and combined with remaining file to create the corrected final file
This time, I tried a different approach. In our shop, we have the software “Notepad++” which is useful to do any search, replace, copy (at column level) from a text file. This time I used below approach,
- Open the Find pop-up using Ctrl-F or View –> Find
- Put [^\x00-\x7F]+ in search box (This will help you to track or replace all non-ascii character in text file.)
- Select search mode as ‘Regular expression’
- Select Find all in current document.
This would highlight all the issue records (and issue characters) in a new window at the bottom. In my case, all the issue characters were same and were at same position. So I replaced them and FTP the file again and restarted the job.
I have found this approach in this post.
I have started a new finance blog, please visit My Finance Blog
You may also like to look at: