Read and jump first line and anothers lines in file - cobol

Question

How I can read a .dat file with struct like that: ( A = ALPHANUMERIC && N = NUMERIC )
0AAAAAAAANNNN (233 BLANK SPACES ) 999999 ( SEQUENTIAL NUMBER ONE BY ONE )
1NNNNNNNNNNNNAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
2AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
1NNNNNNNNNNNNAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
2AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
1NNNNNNNNNNNNAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
2AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
1NNNNNNNNNNNNAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
2AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA (194 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
9 (245 BLANK SPACES) 999999 (SEQUENTIAL NUMBER ONE BY ONE)
So, I know, how I can make a program to read this in C/C++ or in C#, but, I try to make in Cobol, just for study....
But, I don't know what the command I need to use to open the file with this style ( I just know the:
ORGANIZATION IS LINE SEQUENTIAL.
I think, exist a another command to open with custon instructions... i don't know...
So, btw, how I can open the file and read the informations ??
( i just need to read the line 1 on time, and, I need to read the line 2 and 3 always paried ... 4 and 5 && 6 and 7 && 8 and 9 ... )
and, I whant to show that information with DISPLAY ( just for study )
Thanks :)

Answer 1

Something like this below your FD:
01 INPUT-RECORD.
05 IR-RECORD-TYPE PIC X.
88 INPUT-RECORD-IS-HEADER VALUE '0'.
88 INPUT-RECORD-IS-DATA1 VALUE '1'.
88 INPUT-RECORD-IS-DATA2 VALUE '2'.
88 INPUT-RECORD-IS-TRAILER VALUE '9'.
05 FILLER PIC X(whatever).
You may need a "trailing" byte for a record-delimiter, I don't know, and you'll have to sort out the lengths, as they seem to vary.
These in Working-Storage:
01 INPUT-RECORD-HEADER.
05 IRH-RECORD-TYPE PIC X.
05 IRH-ITEM1 PIC X(8).
05 IRH-ITEM2 PIC 9(4).
05 FILLER PIC X(233).
05 IRH-SEQUENCE PIC X(6)
01 INPUT-RECORD-DATA1.
05 IRD1-RECORD-TYPE PIC X.
05 IRD1-ITEM1 PIC 9(14).
05 IRD1-ITEM1 PIC X(19).
05 FILLER PIC X(194).
05 IRD1-SEQUENCE PIC X(6)
01 INPUT-RECORD-DATA2.
05 IRD2-RECORD-TYPE PIC X.
05 IRD2-ITEM1 PIC X(33).
05 FILLER PIC X(194).
05 IRD2-SEQUENCE PIC X(6)
01 INPUT-RECORD-TRAILER.
05 IRT-RECORD-TYPE PIC X.
05 FILLER PIC X(245).
05 IRT-SEQUENCE PIC X(6).
You have to read each record, one at a time. Identify it. Put it in the correct W-S definition. When you read a "2" you can process the "1" you have stored along with the "2".
My datanames aren't very good, as I don't know what your data is. Also I have not "formatted" the definitions, which will make them more readable when you do it.

Answer 2

For OpenCOBOL, here is a sample standard in/standard out filter program:
>>SOURCE FORMAT IS FIXED
*> ***************************************************************
*><* ===========
*><* filter
*><* ===========
*><* :Author: Brian Tiffin
*><* :Date: 20090207
*><* :Purpose: Standard IO filters
*><* :Tectonics: cobc -x filter.cob
*> ***************************************************************
identification division.
program-id. filter.
environment division.
configuration section.
input-output section.
file-control.
select standard-input assign to keyboard.
select standard-output assign to display.
data division.
file section.
fd standard-input.
01 stdin-record pic x(32768).
fd standard-output.
01 stdout-record pic x(32768).
working-storage section.
01 file-status pic x value space.
88 end-of-file value high-value
when set to false is low-value.
*> ***************************************************************
procedure division.
main section.
00-main.
perform 01-open
perform 01-read
perform
until end-of-file
perform 01-transform
perform 01-write
perform 01-read
end-perform
.
00-leave.
perform 01-close
.
goback.
*> end main
support section.
01-open.
open input standard-input
open output standard-output
.
01-read.
read standard-input
at end set end-of-file to true
end-read
.
*> All changes here
01-transform.
move stdin-record to stdout-record
.
*>
01-write.
write stdout-record end-write
.
01-close.
close standard-input
close standard-output
.
end program filter.
*><*
*><* Last Update: dd-Mmm-yyyy
and here is a demonstration of using LINAGE that just happens to read in a text file.
*****************************************************************
* Example of LINAGE File Descriptor
* Author: Brian Tiffin
* Date: 10-July-2008
* Tectonics: $ cocb -x linage-demo.cob
* $ ./linage-demo <filename ["linage-demo.cob"]>
* $ cat -n mini-report
*****************************************************************
IDENTIFICATION DIVISION.
PROGRAM-ID. linage-demo.
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
select optional data-file assign to file-name
organization is line sequential
file status is data-file-status.
select mini-report assign to "mini-report".
DATA DIVISION.
FILE SECTION.
FD data-file.
01 data-record.
88 endofdata value high-values.
02 data-line pic x(80).
FD mini-report
linage is 16 lines
with footing at 15
lines at top 2
lines at bottom 2.
01 report-line pic x(80).
WORKING-STORAGE SECTION.
01 command-arguments pic x(1024).
01 file-name pic x(160).
01 data-file-status pic 99.
01 lc pic 99.
01 report-line-blank.
02 filler pic x(18) value all "*".
02 filler pic x(05) value spaces.
02 filler pic x(34)
VALUE "THIS PAGE INTENTIONALLY LEFT BLANK".
02 filler pic x(05) value spaces.
02 filler pic x(18) value all "*".
01 report-line-data.
02 body-tag pic 9(6).
02 line-3 pic x(74).
01 report-line-header.
02 filler pic x(6) VALUE "PAGE: ".
02 page-no pic 9999.
02 filler pic x(24).
02 filler pic x(5) VALUE " LC: ".
02 header-tag pic 9(6).
02 filler pic x(23).
02 filler pic x(6) VALUE "DATE: ".
02 page-date pic x(6).
01 page-count pic 9999.
PROCEDURE DIVISION.
accept command-arguments from command-line end-accept.
string
command-arguments delimited by space
into file-name
end-string.
if file-name equal spaces
move "linage-demo.cob" to file-name
end-if.
open input data-file.
read data-file
at end
display
"File: " function trim(file-name) " open error"
end-display
go to early-exit
end-read.
open output mini-report.
write report-line
from report-line-blank
end-write.
move 1 to page-count.
accept page-date from date end-accept.
move page-count to page-no.
write report-line
from report-line-header
after advancing page
end-write.
perform readwrite-loop until endofdata.
display
"Normal termination, file name: "
function trim(file-name)
" ending status: "
data-file-status
end-display.
close mini-report.
* Goto considered harmful? Bah! :)
early-exit.
close data-file.
exit program.
stop run.
****************************************************************
readwrite-loop.
move data-record to report-line-data
move linage-counter to body-tag
write report-line from report-line-data
end-of-page
add 1 to page-count end-add
move page-count to page-no
move linage-counter to header-tag
write report-line from report-line-header
after advancing page
end-write
end-write
read data-file
at end set endofdata to true
end-read
.
*****************************************************************
* Commentary
* LINAGE is set at a 20 line logical page
* 16 body lines
* 2 top lines
* A footer line at 15 (inside the body count)
* 2 bottom lines
* Build with:
* $ cobc -x -Wall -Wtruncate linage-demo.cob
* Evaluate with:
* $ ./linage-demo
* This will read in linage-demo.cob and produce mini-report
* $ cat -n mini-report
*****************************************************************
END PROGRAM linage-demo.
With those samples, along with Gilbert's answer, you should have enough to tackle your problem, with the caveat that these examples are shy on proper error handling, so be careful is this is homework or a paid assignment. For an example of standard input/output or by filename depending on command line arguments (or lack thereof), see the ocdoc.cob program in the OpenCOBOL FAQ.
Offtopic: Output of an ocdoc pass over ocdoc.cob itself can be seen at http://opencobol.add1tocobol.com/ocdoc.html (Why mention it? The COBOL lexicon highlighter for Pygments has just been accepted into main. Any Pygments pulled after version 1.6 will allow for COBOL (context free) lexical highlighting.)

Answer 3

You write an ordinary Cobol program that reads a file.
The first byte (character) of the record is either 0, 1, 2, or 9.
Define a Working-Storage area (01 level) for each of the 4 record types. Then, after you read the record, you move it from the input area to the appropriate Working-Storage area for the record.
Then you process the record how you wish from one of the 4 Working-Storage areas.