In the FORTRAN code, before any writing commands are done, insert the line:

OPEN(UNIT=7,FILE='PROB.OUT',FORM='FORMATTED',STATUS='UNKNOWN')

This command assigns the unit number (=7) to which you will write the formatted information. Use a CLOSE command before you end your program; otherwise, the compiler may or may not close the file for you at the end of execution.

Remember to close the unit before your program terminates; i.e.: CLOSE(7)

Sometimes it is useful to check whether the program was able to successfully open the file. This is done with the IOSTAT parameter in the OPEN statement. For example:

OPEN(UNIT=8,FILE='TEST.OUT', STATUS='UNKNOWN',IOSTAT=IOTST)
! check for open error, if iotst does not = 0, then it did not work, print message
IF(IOTST.NE.0) THEN
PRINT *,'COULD NOT OPEN TEST.OUT'
END IF

• The FORMAT command
  • where you specify the way in which you would like the information to be printed
• The PRINT command
  • sends output to the default output
  • example
                PRINT 5, t
    5         FORMAT (E12.2)
  • another ex.
                PRINT *, t
• The WRITE command
  • sends output to the file you have specified in an OPEN statement
  • example
                WRITE(7,5) t
    5           FORMAT (E12.2)
  • another ex. (makes 10 columns)
                WRITE(7,8) (A(J),J=1,10)
    8           FORMAT (10F8.2)

• "A" format: used for text strings. e.g. A10 prints 10 characters.
• "F" format: used for floating point numbers. e.g. F7.2 allows 4 numbers before the decimal and 2 after the decimal. Note that the decimal "period" is counted as one of the 7 places reserved by F7.2
  If the variable is negative, a space for the minus signe is needed, leaving only 3 numbers before the decimal.
• "I" format: used for integers. e.g. I5 reserves 5 spaces for an integer number if positive. If negative, 4 spaces.
• What if your number exceeds the available spaces? Most compilers will display a bunch of asterisks: i.e. ****.** or *****
  Note: A common situation is to neglect to include the space needed to print the minus sign for a negative number.
• "E" format: used for scientific notation. Note: need to leave space for the "e" (meaning power of 10), the sign of the power, and the size of the power. The amount of space needed varies with the compiler & operating system. e.g. E10.3 often is just enough spaces (10) to have the sign, one place left of the decimal and 3 spaces to the right plus all the power of 10 information.
• "G" format: the computer chooses a combination of F, E and sometimes I formats based on the data.
  Note: while G sounds like a simple solution, you often use formated writes to create columns of fixed width or other consistent ordering. Since G format may vary which of E or F it uses, you may not get consistent ordering when it is used.

1. most often used with the print command. e.g. PRINT *,t
2. useful if you want a quick print for debugging purposes
3. useful if you don't know the magnitude of the item to be printed (and don't wish to use E format)
4. very simple and thus easy to use -- you don't have a FORMAT statement
5. undesirable if you want to organize your output neatly. e.g. into tidy columns.

• "T" is "tab" and it does not format data but can be used to space over e.g. T15 goes over 15 spaces.
• "X" is very similar to tab: 15x goes over 15 spaces
• "/" is used to indicate carriage return.

• "H" is used to print text using hollerith format: e.g. 10hPrint this will print the letters "Print this". Notice that space(s) are counted as well. Obviously, counting the number of letters and spaces is quite tedious, so...
• " 'character string ' " enclosing the characters you want printed within single quotes works like the hollerith command without having to count and specify the number of characters. e.g. 'Print this' will print the letters "Print this".

            WRITE(7,3)(A(I),I=1,10)
3           FORMAT(1H0,5E12.2)
This prints out the first 10 elements of array A in scientific notation, the first 5 on the first line; the next 5 on the next line.

for printing out every other element of the first 9 elements of array A, try

            WRITE(7,4)(A(I),I=1,10,2)
4           FORMAT(1H0,5E12.2)

A(1) A(3) A(5) A(7) and A(9) are printed on one line

            WRITE(7,6)
            WRITE(7,8)(I,A(I),I=1,10)
6           FORMAT(2X,'I',5X,'A VALUES',//)
8           FORMAT(I4,E12.2)

You can loop several arrayes at a time

            WRITE(7,5)(I,A(I),B(I),I=1,10)
5           FORMAT(I4,2E12.2)

Here is an illustrative example using FORTRAN 90. Note that the variable used must be declared as a specific sized character string.
            CHARACTER (len=8) :: CHRS
            CHARACTER (10) :: CHRt
            WRITE(CHRS,'(8hPSI SB P)')
            WRITE(CHRt,'(4ht = ,f6.2)') t
            CALL PLCHHQ(.5,0.05,CHRS,12.,0.,0.)
            CALL PLCHHQ(.5,0.1,CHRt,12.,0.,0.)

The first WRITE call encodes a character string of unchanging text.
The second WRITE call encodes a character string including a variable t. Note the use of formatting; you can use F, I, A, etc.

Note that to use the NCAR graphics routine PLCHHQ, one needs to start GKS, have SET ranges defined, etc. If you are unfamiliar, consult RG's NCAR graphics documentation page. Example usage can be found in the ATM150 software library.

Note that the syntax of the Character statement is different. (The FORTRAN 77 version of the character statement is obsoleted with FORTRAN 95.) The rest of the code piece is unchanged.

Back to ATM 150 homepage