Hi RudiC,

Let me explain why and what I am looking exactly.

For input file RECORD, earlier code provided by you provides below output -

Input File -

Code:

$ cat RECORD
SOURCE|LOGIN|SERVER|ABC_123_YYYYMMDD_V01P1234YYMMDDHHMISS.txt|ABC_123_V01P.YYYYMMDD.HHMISS.txt|CATEGORY
SOURCE|LOGIN|SERVER|P_XYZ_1234_ABCD.YYYYMMDD.HHMISS.csv|Report.ABCD_csv.YYYYMMDD.HHMISS.csv|CATEGORY
$

Output -

Code:

SOURCE|LOGIN|SERVER|ABC_123_YYYYMMDD_V01P1234YYMMDDHHMISS.txt|ABC_123_V01P.YYYYMMDD.HHMISS.txt|CATEGORY|9|8|YYYYMMDD|........|%Y%m%d|26|6|YYMMDD|......|%y%m%d|32|6|HHMISS|......|%H%M%S|14|8|YYYYMMDD|........|%Y%m%d|23|6|HHMISS|......|%H%M%S
SOURCE|LOGIN|SERVER|P_XYZ_1234_ABCD.YYYYMMDD.HHMISS.csv|Report.ABCD_csv.YYYYMMDD.HHMISS.csv|CATEGORY|17|8|YYYYMMDD|........|%Y%m%d|26|6|HHMISS|......|%H%M%S|17|8|YYYYMMDD|........|%Y%m%d|26|6|HHMISS|......|%H%M%S

Here, As you can see in input file for 1st record column 4 filename is ABC_123_YYYYMMDD_V01P1234YYMMDDHHMISS.txt which contains 2 date values (as specified in date_format.csv) and 1 time value (as specified in timestamp_format.csv).
For 2nd record column 4 filename is P_XYZ_1234_ABCD.YYYYMMDD.HHMISS.csv which contains 1 date value (as specified in date_format.csv) and 1 time value (as specified in timestamp_format.csv).

Similarly, For column 5 we have different number of date/time values. Which results in different number of columns in output depending on number of date/time values in filename. Hence, It is difficult to identify exact position of date/time variables for column 4 and 5.

For this reason, I was thinking of adding 4 more columns in output which will help me to identify how much date/time variables present in column 4 and 5.

Code:

SOURCE|LOGIN|SERVER|ABC_123_YYYYMMDD_V01P1234YYMMDDHHMISS.txt|ABC_123_V01P.YYYYMMDD.HHMISS.txt|CATEGORY|2|1|1|1
SOURCE|LOGIN|SERVER|P_XYZ_1234_ABCD.YYYYMMDD.HHMISS.csv|Report.ABCD_csv.YYYYMMDD.HHMISS.csv|CATEGORY|1|1|1|1

Here, For 1st output record, 2|1|1|1
-- 2 represents number of date variables in columns 4 as specified in date_format.csv
-- 1 represents number of time variables in columns 4 as specified in timestamp_format.csv
-- 1 represents number of date variables in columns 5 as specified in date_format.csv
-- 1 represents number of time variables in columns 5 as specified in timestamp_format.csv
For 2nd output record, 1|1|1|1
-- 1 represents number of date variables in columns 4 as specified in date_format.csv
-- 1 represents number of time variables in columns 4 as specified in timestamp_format.csv
-- 1 represents number of date variables in columns 5 as specified in date_format.csv
-- 1 represents number of time variables in columns 5 as specified in timestamp_format.csv

Hope this clarifies why I am looking for this. It will be even better if we can use our earlier code and produce output as below -

Code:

SOURCE|LOGIN|SERVER|ABC_123_YYYYMMDD_V01P1234YYMMDDHHMISS.txt|ABC_123_V01P.YYYYMMDD.HHMISS.txt|CATEGORY|2|1|1|1|9|8|YYYYMMDD|........|%Y%m%d|26|6|YYMMDD|......|%y%m%d|32|6|HHMISS|......|%H%M%S|14|8|YYYYMMDD|........|%Y%m%d|23|6|HHMISS|......|%H%M%S
SOURCE|LOGIN|SERVER|P_XYZ_1234_ABCD.YYYYMMDD.HHMISS.csv|Report.ABCD_csv.YYYYMMDD.HHMISS.csv|CATEGORY|1|1|1|1|17|8|YYYYMMDD|........|%Y%m%d|26|6|HHMISS|......|%H%M%S|17|8|YYYYMMDD|........|%Y%m%d|26|6|HHMISS|......|%H%M%S

Here, column 7,8,9 and 10 will serve the same purpose as mentioned above to represent number of date/time variables in column 4 and 5 respectively.