RM/COBOL READ: Text file of different record

I have below scenario. 

Input File <<input_file.txt>>

0005abcde                                                                ====> Record1

0020abcdefghijklmnopqrst                                   =====> Record2

Each record has 2 part:

Part#1          4 digit records length (ex: 0005)

Part#2        Actual Data (ex: abcde)

How the FD section should be defined in such case ?

Swapnil,

Assuming that this is a standard text file, with end-of-line/CR between text records, then nothing really special is needed.

01  my-record.
    02  my-count         pic 9999.
    02  my-data.
        03  pic x occurs 1 to 500 depending on my-count.

I this is not a standard text file, then it gets a bit more complicated...

Thanks Tom.. I have LF at the end of the record... I tried but getting 3906 while opening the file.

 FD INPUTFILE.
          01 IN-REC.
              05 IN-REC-LEN                  PIC 9(4).
              05 IN-REC-DATA.
                 10 FILLER                   PIC X
                    OCCURS 1 TO 500 DEPENDING ON IN-REC-LEN.

In the ENVIRONMENT DIVISION/INPUT-OUTPUT SECTION/FILE-CONTROL:

Make sure that the ORGANIZATION clause states:
    ORGANIZATION LINE SEQUENTIAL

Depending on your compiler's configuration, if you omit the word LINE the organization is probably defaulting to BINARY SEQUENTIAL, which causes the mismatch indicated by error 39,06.

Thanks Tom. I tried. It worked fine.

Hi Tom,

I have a follow-up question regarding this. Suppose I want to write multiple records to the file without trailing white space. For example, I have the following two records:

0005abcde                         => Record 1  
0020abcdefghijklmnopqrst         => Record 2

Each record consists of two parts:
Part 1: A 4-digit record length (e.g., 0005)
Part 2: The actual data (e.g., abcde)

How should the FD (File Description) section and the SELECT clause be defined in this case to handle such records?