In this file declaration:
select bank-file assign to f-bank-file
organization line sequential
file status is wx-fstat.
Variable f-bank-file must be PIC X. Is there any way to make it PIC N? I need to open file with chinese chars in name.
Related
Every time I'm writing to an output file, there will always be an end-of-proof symbol (□).
Consider the program below:
IDENTIFICATION DIVISION.
PROGRAM-ID. HEY.
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT OUTFILE ASSIGN TO "alpha.txt".
DATA DIVISION.
FILE SECTION.
FD OUTFILE.
01 OUTREC PIC X(10).
PROCEDURE DIVISION.
OPEN OUTPUT OUTFILE
MOVE "ABCDEFGHIJ" TO OUTREC
WRITE OUTREC
CLOSE OUTFILE
STOP RUN.
The contents of alpha.txt is
ABCDEFJHIJ
□
I'm using Realia because that is what our school requires us to use. I'm also aware that if I run the same code above using some other compiler such as OpenCobol, the output is just fine, i.e., without the the end-of-proof symbol.
So, how do I remove the end-of-proof symbol?
There is likely no end-of-proof symbol in the file, instead the symbol you see is used for the non-printable character which is in there (or a character without a symbol in the used font; or, as Rick pointed out the end-of-file marker).
From the "txt" extension it looks like you want a text file but as you did not specify anything you end up with a sequential file.
I'm not 100% sure about the support for the (up to COBOL 202x non-standard) ORGANIZATION IS LINE SEQUENTIAL in Realia COBOL, but I suggest to give it a try:
SELECT OUTFILE ASSIGN TO "alpha.txt"
ORGANIZATION IS LINE SEQUENTIAL.
It is almost certainly an end-of-file mark (Cntl-Z or 0x1A). On my system (Win 10) the symbol is displayed as elongated (tall) rather than square. Pasted to this post it is square.
ABCDEFJHIJ
[The square shows in preview and edit; but later disappears.]
See also this answer and this Wikipedia article, End-of-file, for more information.
How to remove end-of-proof symbol?
Reading files in Realia COBOL is not a problem. It may not be a problem with GNUCobol. However, a character by character copy of the file, stopping at the eof-of-file mark, can be done in COBOL or any other language.
Reusing the file definition of file1 for another i.e file2 with out opening /closing file1.
I have a monthly file and my requirement is generate a daily file similar to monthly file.The record length and the File definition of both the files are same.
Can i make use of monthly file's FD to hold data and do some validation using monthly file data items for my Daily file and write my daily file later?
lot of validations are being done using monthly file's FD variables and I cannot do same validations using daily file's variables as i need to make changes/rewrite lot of code in many places in multiple programs.
Note: I am doing my daily file processing & generation after the monthly processing is over at the end .
Also ,we are using UNIX environment.
Please suggest me how can i achieve this if above mentioned method is not possible.
The record area for a file is not available until the file is opened. The SAME RECORD AREA clause may be used to provide an alias.
identification division.
program-id. srac.
environment division.
input-output section.
file-control.
select optional monthly-file assign "monthly.dat"
file status monthly-stat.
select daily-file assign "daily.dat"
file status daily-stat.
i-o-control.
same record area monthly-file daily-file.
data division.
file section.
fd monthly-file.
1 monthly-id pic 999.
fd daily-file.
1 daily-id pic 999.
working-storage section.
1 monthly-stat pic x(2).
1 daily-stat pic x(2).
procedure division.
open input monthly-file daily-file
display "monthly-stat:" space monthly-stat
display "daily-stat: " space daily-stat
read daily-file
display "monthly-id:" space monthly-id
if monthly-id < 10
add 100 to monthly-id
end-if
display "daily-id: " space daily-id
close monthly-file daily-file
stop run
.
monthly-stat: 05
daily-stat: 00
monthly-id: 001
daily-id: 101
In this example, the monthly and daily files have identical record description entries; but, the monthly file does not exist. Yet, one may use the data-names for the monthly file for validation, modification, or whatever.
I have file defined
select bankd-file assign to f-bankd-file
file status is wx-fstat
organization line sequential.
fd bankd-file.
****************
01 bankd-rec pic x(80).
And I am writing to this file line by line, just simple with write command. And on one line I need also write form-feed character. This character I have defined as
01 w-ff pic x value x'0C'.
But in output file, I have before form-feed char NUL char. Please how can I get rid of this NUL char? Other chars are written without any problems.
The question does not specify the used COBOL compiler therefore we can only guess. Different compilers include a x'00' before "non-text-data" to make sure it can be read in correctly (this is mainly done if someone tries to write a COMP item which may contain line breaks and/or form feeds).
This may not be possible with your compiler but normally you would do:
WRITE bankd-rec FROM SPACES BEFORE ADVANCING PAGE
(no need for the FROM SPACES when you do this BEFORE/AFTER the record you actually want to have the form-feed in)
For Micro Focus COBOL, you can turn off x"00" before non-test-data by using the INSERTNULL=OFF in the extfh.cfg
I'm using OpenCobolIDE 4.7.4 (it's based on GnuCOBOL) on Windows 10 and trying to compile this program opening a file for reading:
IDENTIFICATION DIVISION.
PROGRAM-ID. HELLO.
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT STUDENT ASSIGN TO 'input.txt'
ORGANIZATION IS LINE SEQUENTIAL.
DATA DIVISION.
FILE SECTION.
FD STUDENT.
01 STUDENT-FILE.
05 STUDENT-ID PIC 9(5).
05 NAME PIC A(25).
WORKING-STORAGE SECTION.
01 WS-STUDENT.
05 WS-STUDENT-ID PIC 9(5).
05 WS-NAME PIC A(25).
01 WS-EOF PIC A(1).
PROCEDURE DIVISION.
OPEN INPUT STUDENT.
PERFORM UNTIL WS-EOF='Y'
READ STUDENT INTO WS-STUDENT
AT END MOVE 'Y' TO WS-EOF
NOT AT END DISPLAY WS-STUDENT
END-READ
END-PERFORM.
CLOSE STUDENT.
STOP RUN.
The input.txt is in the same directory as the source coude, yet I'm still getting the following error:
Main.cob:24: libcob: File does not exist (STATUS = 35) File : 'input.txt'
What am I doing wrong?
OCIDE has a setting for the output directory, the default is "bin" (relative to the source file). Effectively it just passes this setting to the compiler cobc source.cob -o bin\source.exe
You can change this behaviour in settings Menu Preferences -> Compiler:
Output directory
This option let you chose where to put the binaries, by default binaries will be placed into a bin folder next to the source file. You can define another relative or absolute directory if you want.
In any case you can set the actual name in the environment, check GC FAQ - How to map a file name to an external name.
As an alternative you can set the data directory with the environment var COB_FILE_PATH.
Both environment options can be set in settings Menu Preferences -> Run.
Most IDE for other languages happen to run the executable from another directory (where it is built for example).
A simple test is to write a test program, opening a file for writing.
You'll quickly see what happens.
Perhaps better will be to write the full path in the select clause.
select STUDENT ASSIGN TO '/xpto/folder1/input.txt'
I have fixed-point numbers in file, one in each line, in this format S9(6)V9(2) but when they are actually read, I'm getting non numeric errors while trying to put them in math operations. What is more, when I try to display them in program, a number that is written in file as 567123.45 is saved in variable as +567123.04. And for example the number from file 123.45 is saved in variable as +123.45.00 and it provokes the following error 'WS-VALUE' not numeric: '123.45 0' during a math operation. Why is that?
I'm using OpenCobolIDE 4.7.4 for Windows.
EDIT:
File has records of the following form separated by new lines (read by READ operation record after record):
01 WS-OPERATION.
05 WS-ID PIC A(2).
05 WS-CLIENT PIC 9(5).
05 WS-COUNTRY PIC A(4).
05 WS-VALUE PIC S9(6)V9(2).
The reason is that you try to un-edit a field. 567123.45 in the data is not conforming to PIC S9(6)V9(2) but to -9(6).9(2). - internal stored data vs. print-data.
Simply changing the definition and use MOVE WS-VALUE TO WS-VALUE-INTERNAL (which is defined like you want to) may work with a specific compiler (and specific data) but I'd go a different route:
I'd suggest to always validate the data before doing something with it (the file may be broken or external edited).
At least check the simple numeric data like WS-CLIENT for IS NUMERIC and either do a full validation on the data field WS-VALUE or at least use MOVE FUNCTION NUMVAL(WS-VALUE) TO WS-VALUE-INTERNAL.