Programmers Guide to F

1
Programmers Guide to F

• Chapter 9
• Formatted Input Output
• Version 2.0
• Last Update May 15, 2006

2
Simple INPUT and OUTPUT (I/O)

• The input and output facilities of any
  programming language are extremely important,
  because communication between the user and the
  program is carried out through these features.
• F provides facilities for input and output at two
  quite different levels.
• The so-called list-directed input and output
  statements that we have been using so far
  provide the capability for simple input from the
  keyboard and simple output to the screen or
  printer. Eg
• integer n
• print , Enter a positive integer n !
  Prints on the screen.
• read , n ! Reads from the
  keyboard
• These list-directed I/O - statements, allow the
  user very little control over the source or
  layout of the input data.

3
More Control overINPUT and OUTPUT (I/O)

• F allows the programmer to specify
• from which of the available input units the data
  will be read
• in which format the data will be presented at the
  input unit
• to which of the available output units the
  results will be sent
• in which format the results are to be displayed.

4
(I/O) Statements (Summary)

• input
• read (ltcontrol_listgt) ltlistgt
• read (unit 1, fmt (i4,f3.2)) a, b
• output
• write (ltcontrol_listgt) ltlistgt
• write (unit 6, fmt (i4,f3.2)) a, b
• print (ltformat_specifiergt) ltlistgt
• print , a, b

5
I/O operations in F

• read
• print
• write
• open
• close
• inquire
• backspace
• endfile
• rewind

actual data transfer
connection between an I/O unit and a file
find out things about an I/O unit or a file
affect the position of the file
I/O operations deal with files. A file is a
collection of data records, which will be
discussed next.
6
Records

• A data record is a sequence of values.
• These values may be represented either formatted
  or unformatted.
• F may read and write three types of data records
  
• (a) formatted, (b) unformatted,
  (c) end-file records
• A formatted record contains only formatted data.
• A formatted record is composed of characters
  only.
• These characters include letters, digits, and
  special symbols from the ASCII character set
• The statement
• write(, (i1,a,i2)) 6,,,11
• would produce a record, which can be
  schematically represented as

7
Records

• Unformatted data records consist of values
  represented just as they are stored in computer
  memory.
• The structure of unformatted records is
  processor-dependent, since the amount of space
  required on the external unit depends on the
  details of the data representation on that
  processor.
• If integers are stored using an eight bit binary
  representation,
• write(9) 6, 11
• would produce an unformatted record such as
• Unformatted records contain only unformatted data.

8
endfile
The third kind of data records is the endfile
record, which, at least conceptually, has no
values and has no length. There can be only one
endfile record in a file and, it must be the last
record of a file. An endfile record may be
written explicitly by the programmer using the
endfile statement.
9

• Consider the following 9-digit input data
• 1 2 3 4 5 6 7 8 9
• There are an enormous number of possible
• interpretations
• 1 2 3 4 5 6 7 8 9
• 1, 2, 3, 4, 5, 6, 7, 8, and 9
• 123, 456, 789
• 12345 . 6789
• 1.23, .45, 67, 8900

10
FORMAT and EDIT DESCRIPTORS

• An input (read) statement must contain three
  distinct
• types of information
• where the data is currently residing (sitting).
• how it is to be interpreted.
• where it will be stored in the computers memory
  after it is read by the computer.
• Similarly, an output (print or write) statement
  must define
• where the results are currently stored in the
  computers memory
• where they are to be sent
• in what form they are to be displayed.

11
INPUT Editing for Integers
Simplest edit descriptor is for whole numbers,
which are to be stored in an integer
variable iw If you wished to read the
line 123456789 as a single integer to be stored
in integer variable n read (i9), n as three
separate integers (123, 456, 789) then read
(i3,i3,i3), n1, n2, n3
12
List of input edit descriptors
Descriptor
Meaning iw Read the next w characters as
integers. fw.d Read the next w characters as
a real number with d digits after the
decimal point. aw Read the next
w characters as characters. Lw
Read the next w characters as a logical
value. tc Next character to be
read is at position c counting from the very
left. tln Next character to be read
is n characters before the current
position. trn Next character to be read is n
characters after the current position.
13
INPUT Editing for Real Numbers

• The next edit descriptor is for real numbers,
  which
• are to be stored in a real variable
• fw.d
• w characters are read as real numbers, and d
• shows decimal point. Your character is again
• 123456789
• read (f9.4), real_n
• will cause the first 9 characters to be read as
• 12345.6789

14
Example

• Example for formatting of real numbers using the
  fedit descriptor depending on two different
  INPUT DATA sets
• read (f3.1, f2.2, f3.0), s1,s2,s3
• If the data to be read contains a decimal point
  like in DATA 2 the second part of the edit
  descriptor is ignored.
• DATA 1 DATA 2
• 123456789 .23.56.89
• s1 contains 12.3 0.23
• s2 contains 0.45 0.5
• s3 contains 678.0 6.8

15
The exponential format using e, E, d, D

• A real number can be written in many different
• ways, e.g. 361.764 can be written in all of the
  following
• forms
• 361.764
• 3.61764e2
• 0.361764d3
• 0.0361764e4
• 3617.64d-1
• 3.61764E2
• 361.764D0

16
INPUT Editing for Character Data

• The third major data edit descriptor is the a
  edit descriptor which is used to control the
  editing of character data
• aw
• If w lt len extra blank characters will be added
  at the end
• If w gt len the rightmost len characters of the
  input character string will be stored

17
Example

• Example for formatting of an input character type
  using the
• aedit descriptor .
• character ( len 10 ) ch1
• character ( len 6 ) ch2
• character ( len 15 ) ch3
• When the above declarations are used, the
  following two
• statements have the same effect
• read (a10, a6, a15) , ch1, ch2, ch3
• read (a, a, a), ch1, ch2, ch3

18
Logical data - input editing

• The edit descriptor which is used with logical
• data takes the following form Lw
• Note that the upper case L is used instead of
  l to avoid potential confusion with the digit
  1.
• w characters are used to derive either a
• .true. value, or a .false. value or an error.
• The input field for a logical value consists of
  any number of blanks, followed by an optional
  period, followed by t, or T, or f, or F, followed
  by anything.

19
Logical data - input editing

• Any of the following are acceptable as
  representing .true.
• t
• true
• .T.
• .true.
• .trabzon
• Turkiye
• while the following will all be interpreted as
  false
• F
• False
• .f.
• .false.
• .fadil.
• FAZIL

20
OUTPUT Editing

• The edit descriptors used for output are same as
  those
• used for input.
• However, there are some additional ones for
  output.

21
List of output edit descriptors

• Descriptor Meaning
• iw Output an integer in the next
  w characters.
• fw.d Output a real number in the next
  
• w characters with
  d decimal places.
• aw Output a character string in
  the next w characters.
• a Output a character string with
  no trailing blanks.
• Lw Output w-1 blanks, followed by T or
  F as logical value.
• tc Output the next item starting at position c
  starting from
• the very left.
• tln Output the next item starting n
  character positions before
• the current
  position.
• trn Output the next item starting n character
  positions after
• the current
  position.

22
Tn causes subsequent output to start at column
n.Trn causes a shift to the right by n
columns.Tln causes a shift to the left by n
columns (however, it will not move the
position to the left of column 1).

• total24
• ave2.43
• dev0.32
• write "(t3,a4,i5,t15,a8,f4.2,t30,a10,f4.2)",
• sum",total,average",ave, deviation,dev
• is the same as
• write "(t3,a4,i5,tr4,a8,f4.2,tr4,a10,f4.2)",
• sum",total,average",ave, deviation,dev
• The print out will be as follows
• _ _ sum_ _ _ 24_ _ _ average2.43_ _ _
  deviation0.32

23
OUTPUT EDITING i-edit descriptorfor integers

• The i edit descriptor (iw) causes an integer to
  be output using
• w character positions
• Example
• tom 23
• dick 715
• harry -12
• write (i5,i5,i5) , tom, dick, harry
• This statement will produce the following line of
  output
• (where the symbol represents a space )
• 23715-12

24
OUTPUT EDITING f-edit descriptor for real numbers

• The f edit descriptor (fw.d) causes a real number
  to be
• output using w characters with d digits after the
  decimal
• point.
• Example
• x -3.14159
• y 8275.3024
• z 12.9999
• write (f10.3,f10.3,f10.3) , x, y, z
• This statement will produce the following line of
  output
• (where the symbol represents a space )
• -3.1428275.30213.000
• Rounded up Rounded down Rounded up

25
Example Program

• program tabular_output
• real, parameter third 1.0 / 3.0
• real x
• integer i
• do i 1, 10
• x i
• write (f15.4, f15.4,f15.4), x,
  sqrt(x), xthird
• !write (3f15.4), x, sqrt(x), xthird
• !would do the same job.
• end do
• end program tabular_output

26
Output of the Program

• 1.0000 1.0000 1.0000
• 2.0000 1.4142 1.2599
• 3.0000 1.7321 1.4422
• 4.0000 2.0000 1.5874
• 5.0000 2.2366 1.7100
• 6.0000 2.4495 1.8171
• 7.0000 2.6458 1.9129
• 8.0000 2.8284 2.0000
• 9.0000 3.0000 2.0801
• 10.0000 3.1623 2.1544

27
OUTPUT Editing

• A number, called a repeat count, may be placed
  before the
• i, f, a, L edit descriptors to indicate how
  many times they are
• to be repeated.
• Thus, the output format could be written in a
  more compact form
• print (i5,i5,i5,f6.2,f6.2,f6.2,f6.2),
  x, y, z, d, e, f, g
• Using the repeat count this time
• print (3i5, 4f6.2), x, y, z, d, e, f, g

28
read, write and print statements

• A more general form of the read statement
• read (ltci_listgt) ltinput_listgt
• Note that ltci_listgt stands for the control
  information list
• consisting of the input unit and the format.
• unit specifier External unit to/from data is
  transferred
• unit ltio_unitgt
• unit 5 ! default input device
• unit ! default input device
• format specifier Explicit specification of the
  conversion
• between internal and external representations of
  data.
• fmt ltformatgt

29
read, write and print statements

• Here are two input statements
• read (unit5, fmt) a, b, c
• read (unit, fmt(3f6.3)) a, b, c
• Output is essentially the same, but you use print
  or write
• statements instead of the read statement
• write (unit6, fmt) a, b, c
• print (unit, fmt) a, b, c
• print , d, e, f

30
How to Use More Powerful Format?

• Using the forward slash, /, in read statements
• If you have the following record
• 12345678
• 4567891011
• then the statement
• read (3f2.1,/,3i1), a, b, c, p, q, r
• will read three real numbers from the first line
  and three
• integers from the second.

31
Control Information for Data Transfer

• Data transfer characteristics are specified by
  the following items, called control information
• The direction of data transfer (input or output).
• The external device (unit) to or from which data
  is to be transferred.
• The format description that specifies the
  conversion between the computer-oriented internal
  form of the data and its representation as a
  character string.

32
The unit specifier

• To transfer data to or from an external file, the
  file must be
• connected to a unit.
• read(unit 5) a
• read , b
• The unit specifies a processor-dependent unit
  number. On input, it is the same unit number that
  the processor would use if a read statement
  appeared without the unit number. On output/ it
  is the same unit number that the processor would
  use if a print statement appeared without the
  unit number.
• A unit number identifies one and only one unit in
  an F program. That is, a unit number is global to
  an entire program a particular file may be
  connected to unit 9 in one procedure and referred
  to through unit 9 in another procedure.

33
The fmt specifier

• LIST-DIRECTED FORMATTING
• An asterisk format () specifies list-directed
  formatting, which is
• adequate for input in most situations and for
  output in applications
• where the precise appearance of the results is
  not important.
• read(unit5,fmt) a, b
• write(unit6,fmt) a, b

34
Explicit format control

• The following specification describes the
  arrangement of five fields within a record
• (3f12.0, 2i3)
• An input field can be specified by the following
  format code
• read (unit, fmt(3e12.0, 2i3)) A, B, C, I, J
• Each format code in a format description includes
  the following information
• 1) a count, indicating the number of consecutive
  fields to which it corresponds
• 2) a conversion mode, indicating by letters such
  as f, e, i, L or a.
• 3) a field width
• 4) a decimal position (optional)
• F format code for integer input
• read(unit , fmt (i3, i4, 2i6, i8,
  i3))I,J,K,L,M,N
• external appearance field width
  format code internal value
• 123 3
  i3
  123
• 123 4
  i4
  123
• -123 6
  i6 -123
• -1203 6 i6
  -1203
• -12300 8
  i8 -12300

35
Explicit format control

• F format code for real output
• write(unit ,fmt (f10.3 f6.0,f8.2,f5.2,
  f6.3),A,B,B,C,H
• real variable internal value format code
  external appearance
• A -897.6577 f10.3 _ _
  -897.658
• B 234. f6.0 _ _
  234.
• B 234. f8.2 _ _
  234.00
• C -0.12 f5.2
  -0.12
• H 0. f6.3 _
  0.000

36
Repetitive I/O

• Inum11 Jnum1231
• Inum22 Jnum2345
• read (unit1, fmt(4i3)) Inum1,Jnum1,Inum2,Jnum2
  
• write(unit2,fmt2(i3,i5))Inum1,Jnum1,Inum2,Jnum
  2
• fmt2(i3,i5) is the same as fmt(i3,i5,i3,i5)
• fmt(4i3) is the same as fmt(i3,i3,i3,i3)
• ----------------
• 1 231 2 345

37
Format longer than list

• read (unit1,fmt(3i3)) Inum1,Jnum1,Inum2
• write (unit2,fmt2(i3,i5))Inum1,Jnum1,Inum2
• output ex
• ----------------
• 1 123 2

38
List longer than format

• read (unit1,fmt(3i3)) Inum1,Jnum1,Inum2,Jnum2
• write (unit2,fmt2(i3,i5))
• Inum1,Jnum1,Inum2,Jnum2,
• Inum3,Jnum3, Inum4,Jnum4
• format rescan
• Output ex
• -----------------
• 1 123 2 34
• 3 22 4 443

39
Unformatted record

• An unformatted record consists of a sequence of
  values (in processor-dependent form)
• An unformatted record can only be read by an
  unformatted input statement
• read(unit9) d,e,f
• write(unit9) A,B,C
• produces unformatted records for the values
  of A, B and C.

40
Connecting an external file to aprogram and
disconnecting it

• For any information to be transferred between a
  file and a
• program the file must be connected to a unit.
  
• This connection is initiated by means of an open
  statement
• open(ltopen_specifier_listgt)
• The ltopen_specifier_listgt consists of
• unit ltunit_numbergt
• file ltfile_namegt
• status ltfile_statusgt
• fmt ltformat_modegt
• action ltallowed_actionsgt
• position ltfile_positiongt

41
The status specifier

• The status specifier is not optional.
• It specifies what the status of the file is
  before the file is opened by the program
• status
  ltfilestatusgt
• ltfilestatusgt is a character expression which is
  one of old, new,
• replace or scratch.
• Eg statusold statusnew
• If the status is old the file must already exist,
  whereas if it is new then it must not already
  exist. If new is specified then a new file is
  created. If it is replace and the file already
  exists, then it is deleted and an attempt is made
  to create a new file with the same name. If this
  is successful, the status is changed to old. If
  it is replace and the file doesnt exist, the
  action will be the same as if new had been
  specified. If ltfilestatusgt is scratch then a
  special unnamed file is created for use by the
  program until when the program ceases execution.
  Such a file can be used as a temporary file for
  the duration of execution.

42
The action specifier

• As well as specifying the initial status of the
  file, it must also be
• specified what types of I/O operations are
  allowed with the file.
• The action specifier is used for this purpose
• action ltallowed_actionsgt
• ltallowed_actionsgt is a character expression which
  must take
• exactly one of read, write or readwrite.
• If ltallowed_actionsgt is read then file is to be
  treated as a
• read-only file.
• If ltallowed_actionsgt is write then file is to be
  treated as an
• output file.
• If ltallowed_actionsgt is readwrite then all I/O
  are allowed.

43
The access specifier

• The access specifier is optional.
• The access specifier specifies the access type
  that is permitted to the file
• access ltaccess_typegt
• where ltaccess_typegt is a character expression
  which must take one of the two values sequential
  or direct.
• If ltaccess_typegt is not specified, then the
  default is (it is understood to be) sequential.
• If ltaccess_typegt is specified to be sequential, a
  position
• specifier must be included to instruct the open
  statement where
• the file is to be initially positioned. This is
  discussed next.

44
The position specifier

• The position specifier is optional
• position ltfile_positiongt
• where ltfile_positiongt is a character expression
  which
• must take one of the values rewind or
  append
• If the file did not previously exist and
  ltfile_positiongt
• is rewind then the file is positioned at its
  initial point.
• If the file does exist and ltfile_positiongt is
  rewind
• then the file is positioned at its initial
  point and read statements
• read the first record in the file, and write
  statements write a new
• first record.
• If the file exists and ltfile_positiongt is append
  then the file is
• positioned just before the endfile record.

45
files

• Files normally have a name by which they are
  known to the computer
• system
• file ltfile_namegt
• where ltfile_namegt is a character expression
  which must obey the
• rules for of a file name for the particular
  computer system.
• Thus, if the name of the file is given as semih,
  the statement to connect to
• this file can be given as
• open(unit9, filesemih, statusold,
• actionread, position rewind)

46
The open statement

• The open statement establishes a connection
  between a unit and an external file and
  determines the connection properties.
• After this is done, the file can be used for data
  transfer (reading and writing) using the unit
  number. Eg
• open (unit6,fileMy_Data.bin,statusold,
• action read, form unformatted)
• open (unit8,fileMy_Input.txt,status old,
• action read)
• open(unit31, fileX23, status old,
• action read)

47
The close statement

• Execution of a close statement terminates the
  connection of a file to a unit.
• Any connections not closed explicitly by a close
  statement are closed by the operating system when
  the program terminates.
• The form of the close statement is
• close(ltclose_spec_listgt)Eg
• close(unit9)