Don,
- The field separator is pipe (|). My apologies for not putting that in the initial post.
- Yes, The reject file is supposed to contain the original contents of every line that was changed by one or more rules in the rules file while producing the output file.
- My apologies for typos, It will have 'Contains' or 'Equals' type condition no other conditions.
-Yes, "contauns" was a typo.
-As I mentioned earlier, it is safe to assume only two conditions 'Contains' or Equals'. The example input given in my earlier post has typos as you pointed out.
-The distinction between two conditions:
1. 'Equals' is used for exact string match. (example, if field1 value equals XX, then replace it with rc).
2. 'Contains' is used for pattern match (example, if field1 value contains $, replace it with <blank>, example, field1 value of rc$21 will become rc21 as '$' gets replaced with '' as field one contained '$'.
Appreciate all your help with any unix script solution to this.

---------- Post updated at 11:15 PM ---------- Previous update was at 11:02 PM ----------

Also I updated the files with field/line delimiters:
Input:

Code:

field1|field2|
rc11|rc12| 
rc$21|rc#21|
XX31|yy32|
rc41|r!42|

Rules:

Code:

field|search|condition|replace|
field1|XX|Equals|rc|
field1|$|contains||   
field2|#|contains||
field2|!|contains|c|
field2|yy|Equals|rc|

Output:

Code:

field1|field2|
rc11|rc12|
rc21|rc22|
rc31|rc32|
rc41|rc42|

Rejects:

Code:

field1|field2|
rc$21|rc#22|
XX31|yy32|
rc41|r!42|

Thank you!

Moderator's Comments:

Please use code tags next time for your code and data.