The snippets are under the CC-BY-SA license.

Creative Commons Attribution-ShareAlike 3.0


  • The snippets are under the CC-BY-SA license.
  • Please consider keeping a bookmark
  • (instead of printing)
Print a literal string on standard output
print *, 'Hello, world!'
Loop to execute some code a constant number of times
program main
  implicit none
  integer :: i
  do i=1,10
    write (*,'(A)') "Hello"
  end do
end program main
Like a function which doesn't return any value, thus has only side effects (e.g. Print to standard output)
module foo
  implicit none
  subroutine bar
    print *,"Hello"
  end subroutine bar
end module foo
Create a function which returns the square of an integer
module foo
  implicit none
  function square(i) result(res)
    integer, intent(in) :: i
    integer :: res
    res = i * i
  end function square
end module foo
Declare a container type for two floating-point numbers x and y
type point
  real :: x
  real :: y
end type point
Alternative implementation:
module x
  type point
  real :: x, y
  end type point
end module x
Do something with each item x of the list (or array) items, regardless indexes.
elemental subroutine foo(x)
  real, intent(in) :: x
end subroutine foo

call foo(x)
Print each index i with its value x from an array-like collection items
do i=1, size(items)
  print *,i, items(i)
end do
The structure must be recursive because left child and right child are binary trees too. A node has access to children nodes, but not to its parent.
type binary_tree
    integer :: value
    type(binary_tree), allocatable :: left
    type(binary_tree), allocatable :: right
end type binary_tree
Generate a random permutation of the elements of list x
module M_shuffle
function scramble( number_of_values ) result(array)
   integer,intent(in)    :: number_of_values
   integer,allocatable   :: array(:)
   n=1; m=number_of_values
   do k=1,2
    do i=1,m
     call random_number(u)
     j = n + FLOOR((m+1-n)*u)
     itemp=array(j); array(j)=array(i); array(i)=itemp
end function scramble
end module M_shuffle
use M_shuffle, only : scramble
The list x must be non-empty.

call random_number (a)
x(lbound(x) + int(a*ubound(x))
Check if the list contains the value x.
list is an iterable finite container.
if (findloc (list, x, 1) != 0) then
Alternative implementation:
if (any(x == list)) ...
Pick a random number greater than or equals to a, strictly inferior to b. Precondition : a < b.
call random_number(c)
d = a + (b-a) * c
Pick a random integer greater than or equals to a, inferior or equals to b. Precondition : a < b.
real :: c
integer :: res

call random_number(c)
res = int((b-a+1)*c)
Call a function f on every node of binary tree bt, in depth-first infix order
module x
  type tree
     type (tree), pointer :: left, right
     procedure, pass:: trav
  end type tree
  recursive subroutine trav (t, f)
    class (tree), intent(inout) :: t
       subroutine f(t)
         class (tree), intent(inout) :: t
       end subroutine f
    end interface
    if (associated (t%left)) call trav (t%left, f)
    call f(t)
    if (associated (t%right)) call trav (t%right, f)
  end subroutine trav
end module x
The structure must be recursive. A node may have zero or more children. A node has access to its children nodes, but not to its parent.
type node_t
  integer:: value
  type(node_t), pointer :: next_sibling;
  type(node_t), pointer :: first_child;
end type node_t
Call a function f on every node of a tree, in depth-first prefix order
  recursive subroutine depth_first (node, f)
    type (tr), pointer :: node
       subroutine f(node)
         type(tr), pointer :: node
       end subroutine f
    end interface
    if (associated(node%left)) call depth_first (node, f)
    if (associated(node%right)) call depth_first (node, f)
    call f(node)
  end subroutine depth_first
Reverse the order of the elements of the list x.
This may reverse "in-place" and destroy the original ordering.
  a = a(ubound(a,dim=1)::-1)
Implement a function search which looks for item x in a 2D matrix m.
Return indices i, j of the matching cell.
Think of the most idiomatic way in the language to return the two values at the same time.
function search(m,x)
  integer, dimension(:,:), intent(in) :: m
  integer, intent(in) :: x
  integer, dimension(2) :: search
  search = findloc(x,m)
end function search
Swap the values of the variables a and b
tmp = a
a = b
b = tmp
Alternative implementation:
subroutine swap(a, b)
  integer, intent(inout) :: a, b
  integer :: temp
  temp = a
  a = b
  b = temp
end subroutine swap
Extract the integer value i from its string representation s (in radix 10)
read (unit=s,fmt=*) i
Given a real number x, create its string representation s with 2 decimal digits following the dot.
  write (unit=s,fmt="(F20.2)") x
Declare a new string s and initialize it with the literal value "ネコ" (which means "cat" in japanese)
  use, intrinsic :: iso_fortran_env
  implicit none
  integer, parameter :: u = selected_char_kind('ISO_10646')

  character(kind=u,len=2) :: cat
  cat = u_"ネコ"
Share the string value "Alan" with an existing running process which will then display "Hello, Alan"
str[1] = "Alan"
sync all
if (this_image() == 1) then
  print *,"Hello, ", str
end if
Declare and initialize a matrix x having m rows and n columns, containing real numbers.
  real, dimension(m,n) :: x
Declare and initialize a 3D array x, having dimensions boundaries m, n, p, and containing real numbers.
  real, dimension(:,:,:), allocatable :: x

  allocate (x(m,n,p))
Remove i-th item from list items.
This will alter the original list or return a new list, depending on which is more idiomatic.
Note that in most languages, the smallest valid value for i is 0.
integer, allocatable, dimension(:) :: items
items = [items(:i-1), items(i+1:)]
Launch the concurrent execution of procedure f with parameter i from 1 to 1000.
Tasks are independent and f(i) doesn't return any value.
Tasks need not run all at the same time, so you may use a pool.
  integer :: tasks, n, t, i
  tasks = 1000
  n = num_images()
  t = this_image()
  do i = t, tasks, n
     call f(i)
  end do
  sync all
Create the recursive function f which returns the factorial of the non-negative integer i, calculated from f(i-1)
module x
  implicit none
  recursive function f (i) result (res)
    integer, intent(in) :: i
    integer :: res
    if (i <= 0) then
       res = 1
       res = f(i-1) * i
    end if
  end function f
end module x
Create function exp which calculates (fast) the value x power n.
x and n are non-negative integers.
function exp(x,n) result(I)
end function exp
Assign to the variable x the new value f(x), making sure that no other thread may modify x between the read and the write.
x = f(x)
Alternative implementation:
integer, dimension[*] :: x

  x = f(x)
end critical
Find substring t consisting in characters i (included) to j (excluded) of string s.
Character indices start at 0 unless specified otherwise.
Make sure that multibyte characters are properly handled.
character(len=:), allocatable :: t

  t = s(i:j-1)
Set the boolean ok to true if the string word is contained in string s as a substring, or to false otherwise.
ok = index(string, word) /= 0
Create string t containing the same characters as string s, in reverse order.
Original string s must remain unaltered. Each character must be handled correctly regardless its number of bytes in memory.
  character(len=:), allocatable :: t
  integer :: i, n

  allocate (t, mold=s)
  n = len(s)
  do i = 0, n - 1
     t(n-i : n-i) = s(i+1 : i+1)
  end do
Print each item v of list a which is not contained in list b.
For this, write an outer loop to iterate on a and an inner loop to iterate on b.
  outer: do i=1,size(a)
     do j=1,size(b)
        if (a(i) == b(j)) cycle outer
     end do
     print *,a(i)
  end do outer
Look for a negative value v in 2D integer matrix m. Print it and stop searching.
  outer: do j=1, size(m,2)
     do i=1, size(m,1)
        v = m(i,j)
        if (v < 0) then
           print *,v
           exit outer
        end if
     end do
  end do outer
Insert the element x at position i in the list s. Further elements must be shifted to the right.
integer, dimension(:), allocatable :: s

s = [s(1:i-1), x, s(i:)]
Sleep for 5 seconds in current thread, before proceeding with the next instructions.
module M_time
subroutine system_sleep(wait)
use,intrinsic :: iso_c_binding, only: c_int
integer,intent(in) :: wait
integer(kind=c_int):: waited
 function c_usleep(msecs) bind (C,name="usleep")
  integer(c_int) :: c_usleep
  integer(c_int),intent(in),VALUE :: msecs
 end function c_usleep
end interface
end subroutine system_sleep
end module M_time
program x
use M_time
call system_sleep(5000000)
Create the string t consisting of the 5 first characters of the string s.
Make sure that multibyte characters are properly handled.
 character(len=5) :: t
 t = s(1:5)
Create string t consisting in the 5 last characters of string s.
Make sure that multibyte characters are properly handled.
character(len=5) :: t
t = s(len(s)-4:)
Assign to variable s a string literal consisting in several lines of text, including newlines.
  s = "Hello &
Build list chunks consisting in substrings of the string s, separated by one or more space characters.
program main
use stringifor_string_t !
   implicit none
   type( string ) ::  string1
   type( string ), allocatable :: substrings( : )
   integer :: i

   string1 =  " Build list _chunks consisting in substrings of input string _s  separated by one or more space characters"
   call string1%split(sep=' ', tokens=substrings )
   do i=1,size(substrings, dim=1)
   write(*,*) substrings(i)
end program main
Write a loop that has no end clause.
end do
Concatenate elements of string list x joined by the separator ", " to create a single string y.
  write (unit=y,fmt='(*(A,:,", "))') x

Calculate the sum s of the integer list or array x.
   s = sum(x)
Create the string representation s (in radix 10) of the integer value i.
  write (unit=s,fmt=*) i
Fork-join : launch the concurrent execution of procedure f with parameter i from 1 to 1000.
Tasks are independent and f(i) doesn't return any value.
Tasks need not run all at the same time, so you may use a pool.
Wait for the completion of the 1000 tasks and then print "Finished".
  integer :: tasks, n, t, i
  tasks = 1000
  n = num_images()
  t = this_image()
  do i = t, tasks, n
     call f(i)
  end do
  sync all
  print *,"Finished"
Create the list y containing the items from the list x that satisfy the predicate p. Respect the original ordering. Don't modify x in-place.
y = pack(x,mask=p(x))
Create the string lines from the content of the file with filename f.
program p
   character(len=:),allocatable :: s
   allocate(character(len=i) :: s)
end program p
Print the message "x is negative" to standard error (stderr), with integer x value substitution (e.g. "-2 is negative").
program write_to_stderr
   use iso_fortran_env, only : stderr=>ERROR_UNIT   
   implicit none
   integer :: x=-2
   write(stderr,'(i0," is negative")') x
end program write_to_stderr
Assign to x the string value of the first command line parameter, after the program name.
  character(len=:), allocatable :: x
  integer :: n
  call get_command_argument (1, length=n)
  allocate (character(n):: x)
  call get_command_argument (1, x)
Assign to the variable d the current date/time value, in the most standard type.
integer, dimension(8) :: d

call date_and_time (values=d)

Set i to the first position of string y inside string x, if exists.

Specify if i should be regarded as a character index or as a byte index.

Explain the behavior when y is not contained in x.
Assign to x2 the value of string x with all occurrences of y replaced by z.
Assume occurrences of y are not overlapping.
  character(len=:), allocatable :: x
  character(len=:), allocatable :: x2
  character(len=:), allocatable :: y,z
  integer :: j, k
     j = index(x(k:),y)
     if (j==0) then
        x2 = x2 // x(k:)
     end if
     if (j>1) then
        x2 = x2 // x(k:j+k-2)
     end if
     x2 = x2 // z
     k = k + j + len(y) - 1
  end do
From the real value x in [0,1], create its percentage string representation s with one digit after decimal point. E.g. 0.15625 -> "15.6%"
write (*,'(F5.1,"%")') x*100.
Calculate binom(n, k) = n! / (k! * (n-k)!). Use an integer type able to handle huge numbers.
integer, parameter :: i8 = selected_int_kind(18)
integer, parameter :: dp = selected_real_kind(15)
n = 100
k = 5
print *,nint(exp(log_gamma(n+1.0_dp)-log_gamma(n-k+1.0_dp)-log_gamma(k+1.0_dp)),kind=i8)
Use seed s to initialize a random generator.

If s is constant, the generator output will be the same each time the program runs. If s is based on the current value of the system clock, the generator output will be different each time.
  call random_seed(size = n)
  ! ...
  call random_seed(put=seed)
Get the current datetime and provide it as a seed to a random generator. The generator sequence will be different at each run.
  call random_seed (size=k)
  allocate (seed(k))
  call date_and_time (values=val)
  seed = [(173*i**2+4567,i=1,k)]
  m = min(8,k)
  seed(1:m) = seed(1:m) + val(m:1:-1)
  call random_seed (put=seed)
Basic implementation of the Echo program: Print all arguments except the program name, separated by space, followed by newline.
The idiom demonstrates how to skip the first argument if necessary, concatenate arguments as strings, append newline and print it to stdout.
program x
  implicit none
  character (len=:), allocatable :: a
  integer :: n, i, l
  n = command_argument_count()
  a = ''
  do i=1,n
     call get_command_argument(i, a, l)
     if (l > len(a)) then
        deallocate (a)
        allocate (character(len=l) :: a)
        call get_command_argument(i, a)
     end if
     write (unit=*,fmt='(A)', advance="no") a
     if (i < n) then
        write (unit=*,fmt='(" ")', advance="no")
        write (unit=*,fmt='()')
     end if
  end do
Compute the greatest common divisor x of big integers a and b. Use an integer type able to handle huge numbers.
function gcd(m,n) result(answer)
implicit none
integer(kind=int64),intent(in)  :: m, n
integer(kind=int64)             :: answer,irest,ifirst
         irest = mod(ifirst,answer)
         if(irest == 0)  exit
         ifirst = answer
         answer = irest
      answer= iabs(answer)
end function gcd
Create the string s of integer x written in base 2.

E.g. 13 -> "1101"
write (unit=s,fmt='(B0)') x
Alternative implementation:
write (unit=s,fmt='(B32)') x
Declare a complex x and initialize it with value (3i - 2). Then multiply it by i.
  complex :: x
  x = (-2,3)
  x = x * (0,1)
Execute a block once, then execute it again as long as boolean condition c is true.
  call do_something
  if (.not. c) exit
end do
Declare the floating point number y and initialize it with the value of the integer x .
real :: y
y = x
Declare integer y and initialize it with the value of floating point number x . Ignore non-integer digits of x .
Make sure to truncate towards zero: a negative x must yield the closest greater integer (not lesser).
integer :: y
real  :: x
Declare the integer y and initialize it with the rounded value of the floating point number x .
Ties (when the fractional part of x is exactly .5) must be rounded up (to positive infinity).
integer :: y

y = nint(x)
Find how many times string s contains substring t.
Specify if overlapping occurrences are counted.
  lt = len(t) - 1
  k = 1
  num = 0
     print *,s(k:)
     i = index(s(k:),t)
     if (i==0) exit
     num = num + 1
     k = k + i + lt
  end do
  print *,num
Count number c of 1s in the integer i in base 2.

E.g. i=6 → c=2
c = popcnt(i)
Write boolean function addingWillOverflow which takes two integers x, y and return true if (x+y) overflows.

An overflow may be above the max positive value, or below the min negative value.
  logical function adding_will_overflow (x,y) result(res)
    integer, intent(in) :: x, y
    res = (x > 0 .and. y > huge(x) - x) .or. (x < 0 .and. y < huge(x) - x)
  end function adding_will_overflow
Write the boolean function multiplyWillOverflow which takes two integers x, y and returns true if (x*y) overflows.

An overflow may reach above the max positive value, or below the min negative value.
logical function multiply_will_overflow (x, y) result(res)
    integer, intent(in) :: x, y
    integer, parameter :: ik = selected_int_kind (int(digits(y)*log10(2.)*2))

    res = int(x,kind=ik) * int(y,kind=ik) > huge(x)
end function multiply_will_overflow
Exit immediately.
If some extra cleanup work is executed by the program runtime (not by the OS itself), describe it.
Create a new bytes buffer buf of size 1,000,000.
  integer(kind=c_int8_t), dimension(:), allocatable :: a

  allocate (a(10**6))
You've detected that the integer value of argument x passed to the current function is invalid. Write the idiomatic way to abort the function execution and signal the problem.
  if (x > largest_value) error stop "Illegal value in function."
Expose a read-only integer x to the outside world while being writable inside a structure or a class Foo.
module x
  implicit none
  type foo
     integer, private :: x
     procedure :: readx
  end type foo
  integer function readx(f)
    class(foo) :: f
    readx = f%x
  end function readx
end module x
Implement the procedure control which receives one parameter f, and runs f.
module x
  implicit none
  subroutine control(f)
       subroutine f()
       end subroutine f
    end interface
    call f
  end subroutine control
end module x
Assign to variable x the length (number of bytes) of the local file at path.
program xx
implicit none
character(len=256) :: message
character(len=4096) :: filename
integer :: ios,x
logical :: foundit
   write(*,*)'file ',trim(filename),' not found'
   write(*,*)'size =',x,'bytes'
end program xx
Set the boolean b to true if string s starts with prefix prefix, false otherwise.
  logical :: b
  b = index (string, prefix) == 1
Set boolean b to true if string s ends with string suffix, false otherwise.
  b = (s(len(s)-len(suffix)+1:) == suffix)
Assign to the string x the value of the fields (year, month, day) of the date d, in format YYYY-MM-DD.
program p
integer, dimension(8) :: d
character(len=10) :: x
call date_and_time (values=d)
print *,'DATE=',x
end program p
Assign to the string s the name of the currently executing program (but not its full path).
program p
implicit none
character(len=:),allocatable :: s
integer                      :: stat,l,i,j,k
 call get_command_argument (0,length=l)
 allocate(character(len=l) :: s)
 call get_command_argument (0,s,status=stat)
 if (stat == 0) then
  print *, "The program's name is " // trim (s)
end program p
Assign to string dir the path of the working directory.
(This is not necessarily the folder containing the executable itself)
     function c_getcwd (buf, size) bind(C,name="getcwd") result(r)
       type(c_ptr) :: r
       character(kind=c_char), dimension(*), intent(out) :: buf
       integer(kind=c_size_t), value :: size
     end function c_getcwd
  end interface

    if (c_associated(c_getcwd (buf, size(buf,kind=c_size_t)))) then
       n = findloc(buf,achar(0),1)
       allocate (character(len=n-1) :: dir)
       dir(1:n-1) = transfer(buf(1:n-1),dir(1:n-1))
    end if
Print the value of variable x, but only if x has been declared in this program.
This makes sense in some languages, not all of them. (Null values are not the point, rather the very existence of the variable.)
implicit none

print *,x
Set n to the number of bytes of a variable t (of type T).
n = c_sizeof(i)
Set the boolean blank to true if the string s is empty, or null, or contains only whitespace ; false otherwise.
blank = s == ''
From current process, run program x with command-line parameters "a", "b".
program p
   integer :: i

   call execute_command_line("x a b", exitstat=i)
   print *, "Exit status of x was ", i

   call execute_command_line("x a b", wait=.false.)
   print *, "running x in the background"
end program p 
Remove all occurrences of string w from string s1, and store the result in s2.
  character(len=:), allocatable :: s1
  character (len=:), allocatable :: s2

  character(len=:), allocatable :: w
  integer :: i, j, k, tc

  allocate (s2, mold=s1)
  i = 1
  j = 1
     k = index (s1(i:), w)
     if (k == 0) then
        s2(j:j+len(s1)-i) = s1(i:)
        s2 = s2(:j+len(s1)-i)
        tc = k - 1
        if (tc > 0) s2(j:j+tc-1) = s1(i:i+tc-1)
        i = i + tc + len(w)
        j = j + tc
     end if
  end do
Set n to the number of elements of the list x.
n = size(x)
Read an integer value from the standard input into the variable n
   integer :: n
   read (*,*) n
Create an enumerated type Suit with 4 possible values SPADES, HEARTS, DIAMONDS, CLUBS.
enum, bind(c)
  enumerator :: spades, hearts, diamonds, clubs
end enum
Verify that predicate isConsistent returns true, otherwise report assertion violation.
Explain if the assertion is executed even in production environment or not.
if (.not. isconsistent) stop "Inconsistent state"
Write the function binarySearch which returns the index of an element having the value x in the sorted array a, or -1 if no such element exists.
   integer function binarysearch(a,x) result(l)
      do while(l<=r)
         end if
         end if
      end do
   end function binarysearch
measure the duration t, in nanoseconds, of a call to the function foo. Print this duration.
integer, parameter :: i8 = selected_int_kind(15)
integer (kind=i8) :: count, count_rate, count_2
x = 1.2
call system_clock (count, count_rate)
call foo
call system_clock (count_2)
if (count_rate == 10**9) then
  print *,"The call to foo used ", (count_2-count), "Nanoseconds"
  print *,"The call to foo used ", (count_2-count)/real(count_rate)*1e9, "Nanoseconds"
end if
Write a function foo that returns a string and a boolean value.
subroutine foo(c, b)
  character(len=:), allocatable, intent(out) :: c
  logical, intent(out) :: b
  c = 'string'
  b = .false.
end subroutine foo
Import the source code for the function foo body from a file "foobody.txt".
function foo()
include "foobody.txt"
end function foo
Execute f1 if condition c1 is true, or else f2 if condition c2 is true, or else f3 if condition c3 is true.
Don't evaluate a condition when a previous condition was true.
if (c1) then
  call f1
else if (c2) then
  call f2
else if (c3) then
  call f3
end if
Run the procedure f, and return the duration of the execution of f.
integer, parameter :: i8 = selected_int_kind(15)
integer(kind=i8) :: start, finish, rate

call system_clock (count_rate=rate)
call system_clock (count=start)
call f()
call system_clock (count=finish)
print *,(finish-start)/real(rate)
Set boolean ok to true if string word is contained in string s as a substring, even if the case doesn't match, or to false otherwise.
 function u_i(string, substr)
  character (len=*), intent(in) :: string, substr
  integer :: i,j, c1, c2, u_i
  u_i = 0
  out: do i=1,len(string)-len(substr)+1
   c1 = iachar(string(i:i))
   if (c1 >= iachar('a') .and. c1 <= iachar('z')) c1 = c1 - 32
   do j=0,len(substr)-2
     c2 = iachar(substr(j+1:j+1))
     if (c2 >= iachar('a') .and. c1 <= iachar('z')) c2 = c2 - 32
     if (c1 /= c2) cycle out
   end do
   u_i = i
 end do out
end function u_i

ok = u_i(string, word) /= 0
Declare and initialize a new list items, containing 3 elements a, b, c.
integer, dimension(3) :: items
items = [a,b,c]
Remove at most 1 item from list items, having the value x.
This will alter the original list or return a new list, depending on which is more idiomatic.
If there are several occurrences of x in items, remove only one of them. If x is absent, keep items unchanged.
integer, dimension(:), allocatable :: items

i = findloc(items, x)
if (i /= 0) items = [items(1:i-1), items(i+1:)]
Remove all occurrences of the value x from list items.
This will alter the original list or return a new list, depending on which is more idiomatic.
items = pack (items,items != x)
Set the boolean b to true if the string s contains only characters in the range '0'..'9', false otherwise.
b = .true.
do i=1, len(s)
  if (s(i:i) < '0' .or. s(i:i) > '9') then
    b = .false.
  end if
end do
Create a new temporary file on the filesystem.
open (newunit=u,status="scratch")
Iterate in sequence over the elements of the list items1 then items2. For each iteration print the element.
print *,items1,items2
Assign to string s the hexadecimal representation (base 16) of integer x.

E.g. 999 -> "3e7"
write (*,'(Z8.8)') x
Iterate alternatively over the elements of the lists items1 and items2. For each iteration, print the element.

Explain what happens if items1 and items2 have different size.
do i=1, min(size(items1), size(items2))
  print *,items1(i), items2(i)
end do
Set boolean b to true if file at path fp exists on filesystem; false otherwise.

Beware that you should never do this and then in the next instruction assume the result is still valid, this is a race condition on any multitasking OS.
inquire (file=fp,exist=b)
Print message msg, prepended by current date and time.

Explain what behavior is idiomatic: to stdout or stderr, and what the date format is.
integer :: value(8)
msg = "asdf"
call date_and_time (values=value)
write (unit=*,fmt='(I4,"-",I2.2,"-",I2.2," ",I2.2,":",I2.2,".",I4.4,": ",A)') value(1:3), value(5:7), msg
Extract floating point value f from its string representation s
Alternative implementation:
read (unit=s,fmt=*) f
Create string t from string s, keeping only ASCII characters
n = 0
do i=1,len(s)
   if (iachar(s(i:i)) <= 127) n = n + 1
end do
allocate (character(len=n) :: t)
j = 0
do i=1,len(s)
   if (iachar(s(i:i)) <= 127) then
      j = j + 1
      t(j:j) = s(i:i)
   end if
end do
Read a list of integer numbers from the standard input, until EOF.
integer :: I,j,k ,l(3)
read(*,*) I, j, k, l
Remove the last character from the string p, if this character is a forward slash /
program x
character(len=:),allocatable :: string
integer :: ii
string='my string/'
string=trim(merge(string(:ii-1)//' ',string,string(ii:ii).eq.'/'))
end program x
Create string s containing only the character c.
character(len=:),allocatable :: s
character(len=1) :: c
Create the string t as the concatenation of the string s and the integer i.
write (unit=t,fmt='(A,I0)') s, i
Find color c, the average between colors c1, c2.

c, c1, c2 are strings of hex color codes: 7 chars, beginning with a number sign # .
Assume linear computations, ignore gamma corrections.
character (len=7) :: c, c1, c2
integer, dimension(3) :: v, v1, v2

read (unit=c1,fmt='(X,3Z2)') v1
read (unit=c2,fmt='(X,3Z2)') v2
v = (v1 + v2)/2
write (unit=c,fmt='("#",3Z2)') v
Delete from filesystem the file having path filepath.
open (10,file=filepath,status="old", iostat=ierr)
if (ierr == 0) close (10,status="delete")
Assign to the string s the value of the integer i in 3 decimal digits. Pad with zeros if i < 100. Keep all digits if i1000.
write (unit=s,fmt='(I0.3)') i
Initialize a constant planet with string value "Earth".
  character(len=*), parameter :: planet = "Earth"
Alternative implementation:
character(*), parameter :: planet = "Earth"
Create a new list y from randomly picking exactly k elements from list x.

It is assumed that x has at least k elements.
Each element must have same probability to be picked.
Each element from x must be picked at most once.
Explain if the original ordering is preserved or not.
allocate (sample(k))
do i=1,k
   sample(i) = x(i)
end do
do i=k+1,n
  call random_number(a)
  j = 1 + int(i*a)
  if (j .le. k) sample(j) = x(i)
end do
Define a Trie data structure, where entries have an associated value.
(Not all nodes are entries)
  type trie_p
     type(trie), pointer :: p => NULL()
  end type trie_p
  type trie
     class(*), allocatable :: value
     type(trie_p), dimension(:), allocatable :: nodes
  end type trie
Execute f32() if platform is 32-bit, or f64() if platform is 64-bit.
This can be either a compile-time condition (depending on target) or a runtime detection.
program main
  use iso_c_binding
  implicit none
  type(c_ptr) :: x
  logical, parameter :: is_64 = c_sizeof(x) == 8
  if (is_64) then
     call f64
     call f32
  end if
end program main
Multiply all the elements of the list elements by a constant c
elements = elements * c
execute bat if b is a program option and fox if f is a program option.
  do i=1, command_argument_count ()
     call get_command_argument (i, length=length)
     if (length > len(opt)) then
        deallocate (opt)
        allocate (character(length) :: opt)
     end if
     call get_command_argument (i, opt)
     if (opt(1:1) /= '-') exit
     do j=2, length
        select case (opt(j:j))
        case ('b')
           print *,"bat"
        case ('f')
           print *,"fox"
        end select
     end do
  end do
Print all the list elements, two by two, assuming list length is even.
write (*,'(2I8,:," ")') list
Assign to the variable x the last element of the list items.
x = items(ubound(items,1))
Create the list ab containing all the elements of the list a, followed by all the elements of the list b.
real, dimension(:), allocatable :: ab

ab = [a, b]
Create the string t consisting of the string s with its prefix p removed (if s starts with p).
  if (index(s,p) == 1) then
     t = s(len(p)+1:)
     t = s
  end if
Create string t consisting of string s with its suffix w removed (if s ends with w).
  i = index(s,w,back=.true.) - 1
  if (i == len(s) - len(w) ) then
     allocate (t, source=s(:i))
     allocate (t, source=s)
  end if
Assign to the integer n the number of characters of the string s.
Make sure that multibyte characters are properly handled.
n can be different from the number of bytes of s.
n = len(s)
Append the element x to the list s.
real, allocatable, dimension(:) :: s

s = [s, x]
From array a of n bytes, build the equivalent hex string s of 2n digits.
Each byte (256 possible values) is encoded as two hexadecimal characters (16 possible values per digit).
  character(len=:), allocatable :: s
  allocate (character(len=2*size(a)) :: s)
  write(unit=s,fmt='(*(Z2.2))') a
From hex string s of 2n digits, build the equivalent array a of n bytes.
Each pair of hexadecimal characters (16 possible values per digit) is decoded into one byte (256 possible values).
  integer(kind=c_int8_t), dimension(:), allocatable :: a

  allocate (a(len(s)/2))
  read(unit=s,fmt='(*(Z2))') a
Set boolean b to true if if the point with coordinates (x,y) is inside the rectangle with coordinates (x1,y1,x2,y2) , or to false otherwise.
Describe if the edges are considered to be inside the rectangle.
logical :: b
b = (x1 < x) .and. (x < x2) .and. (y1 < y) .and. (y < y2)
Return the center c of the rectangle with coördinates(x1,y1,x2,y2)
real :: center(2)
center = [(x1 + x2)/2, (y1 + y2)/2]
Output the source of the program.
! See original attribution URL
Exit a program cleanly indicating no error to OS
Perform matrix multiplication of a real matrix a with nx rows and ny columns, a real matrix b with ny rows and nz columns and assign the value to a real matrix c with nx rows and nz columns.
c = matmul (a,b)
Produce a new list y containing the result of the function T applied to all elements e of the list x that match the predicate P.
y = pack (t(x), mask=p(x))
Declare an external C function with the prototype

void foo(double *a, int n);

and call it, passing an array (or a list) of size 10 to a and 10 to n.

Use only standard features of your language.
module x
  use iso_c_binding
     subroutine foo (a, n) bind(c)
       real(kind=c_double), dimension(*) :: a
       integer (kind=c_int), value :: n
     end subroutine foo
  end interface
  subroutine call_foo
    real(kind=c_double) :: a(10)
    call foo(a,10)
  end subroutine call_foo
end module x
Given a one-dimensional array a, check if any value is larger than x, and execute the procedure f if that is the case
if (any(a > x)) call f
Declare a real variable a with at least 20 digits; if the type does not exist, issue an error at compile time.
  integer, parameter :: qp = selected_real_kind(20)
  real(kind=qp) :: a
Declare two two-dimensional arrays a and b of dimension n*m and m*n, respectively. Assign to b the transpose of a (i.e. the value with index interchange).
real :: a(n,m), b(m,n)

b = transpose(a)
Given an array a, set b to an array which has the values of a along its second dimension shifted by n. Elements shifted out should come back at the other end.
b = cshift(a,n,dim=2)
Pass an array a of real numbers to the procedure (resp. function) foo. Output the size of the array, and the sum of all its elements when each element is multiplied with the array indices i and j (assuming they start from one).
subroutine foo(a)
  implicit none
  real, dimension(:,:) :: a
  real :: s
  integer :: i,j
  print *,size(a,1), size(a,2)
  s = 0
  do j=1,size(a,2)
    do i=1,size(a,1)
      s = s + a(i,j) * i * j
    end do
  end do
  print *,s
end subroutine foo
  call foo(a)
Given an integer array a of size n, pass the first, third, fifth and seventh, ... up to the m th element to a routine foo which sets all these elements to 42.
! Caller:
  integer, dimension(n) :: a
  call foo(a(1:m:2))

! Callee:

  subroutine foo(a)
    integer, dimension(:), intent(inout) :: a
    a = 42
  end subroutine foo
Abort program execution with error condition x (where x is an integer value)
Compute the hypotenuse h of the triangle where the sides adjacent to the square angle have lengths x and y.
h = hypot(x,y)
Calculate n, the Euclidean norm of data (an array or list of floating point values).
n = norm2( data )
Calculate the sum of squares s of data, an array of floating point values.
s = sum( data**2 )
Calculate the mean m and the standard deviation s of the list of floating point values data.
real, allocatable :: data(:)
real :: m, s
m = sum( data ) / size( data )
s = sqrt( sum( data**2 ) / size( data ) - m**2 )
Given a real number a, print the fractional part and the exponent of the internal representation of that number. For 3.14, this should print (approximately)

0.785 2
a = 3.14
print *,fraction(a), exponent(a)
Read an environment variable with the name "FOO" and assign it to the string variable foo. If it does not exist or if the system does not support environment variables, assign a value of "none".
  call get_environment_variable ("FOO", length=n, status=st)
  if (st /= 0) then
    foo = "none"
    allocate (character(len=n) :: foo)
    call get_environment_variable ("FOO", value=foo)
  end if 
Execute different procedures foo, bar, baz and barfl if the string str contains the name of the respective procedure. Do it in a way natural to the language.
  select case (str)
  case ("foo")
     call foo
  case ("bar")
     call bar
  case ("baz")
     call baz
  case ("barfl")
     call barfl
  end select
Allocate a list a containing n elements (n assumed to be too large for a stack) that is automatically deallocated when the program exits the scope it is declared in.
  integer, dimension(:), allocatable :: a
  allocate (a(n))
Given the arrays a,b,c,d of equal length and the scalar e, calculate a = e*(a+b*c+cos(d)).
Store the results in a.
a = e*(a+b*c+cos(d))
Declare a type t which contains a string s and an integer array n with variable size, and allocate a variable v of type t. Allocate v.s and v.n and set them to the values "Hello, world!" for s and [1,4,9,16,25], respectively. Deallocate v, automatically deallocating v.s and v.n (no memory leaks).
  type t
    character(len=:), allocatable :: c
    integer, dimension(:), allocatable :: n
  end type t
  type(t), allocatable :: v

  allocate (v)
  v%s = 'Hello, world!'
  v%n = [1,4,9,16,25]

  deallocate (v)
Assign, at runtime, the compiler version and the options the program was compilerd with to variables version and options, respectively, and print them. For interpreted languages, substitute the version of the interpreter.

Example output:

GCC version 10.0.0 20190914 (experimental)
-mtune=generic -march=x86-64
  character(len=:), allocatable :: version, options
  version = compiler_version()
  options = compiler_options()
  print *,version
  print *,options
Append extra character c at the end of string s to make sure its length is at least m.
The length is the number of characters, not the number of bytes.
Prepend extra character c at the beginning of string s to make sure its length is at least m.
The length is the number of characters, not the number of bytes.
repeat(x,max(n-len(s),0)) // s
Add the extra character c at the beginning and ending of string s to make sure its length is at least m.
After the padding the original content of s should be at the center of the result.
The length is the number of characters, not the number of bytes.

E.g. with s="abcd", m=10 and c="X" the result should be "XXXabcdXXX".
  i = (m-len(s))/2
  j = (m - len(s)) - (m-len(s))/2
  s = repeat(c,i) // s // repeat(c,j)
Set i to the first index in list items at which the element x can be found, or -1 if items does not contain x.
i = findloc(items, x)
if (i == 0) i = -1
Loop through list items checking a condition. Do something else if no matches are found.

A typical use case is looping through a series of containers looking for one that matches a condition. If found, an item is inserted; otherwise, a new container is created.

These are mostly used as an inner nested loop, and in a location where refactoring inner logic into a separate function reduces clarity.
  found = .false.
  do i=1,size(items)
    if (items(i) == "asdf") then
      found = .true.
    end if
  end do
  if (.not. found) call do_something
Insert the element x at the beginning of the list items.
items = [x, items]
Declare an optional integer argument x to procedure f, printing out "Present" and its value if it is present, "Not present" otherwise
subroutine f(x)
  integer, optional :: x
  if (present(x)) then
    print *,"Present", x
    print *,"Not present"
  end if
end subroutine f
Remove the last element from the list items.
items = items(1:size(items)-1)
Create the new list y containing the same elements as the list x.

Subsequent modifications of y must not affect x (except for the contents referenced by the elements themselves if they contain pointers).
type (foo), allocatable, dimension(:) :: y
y = x
Copy the file at path src to dst.
character, dimension(n_buff) :: buffer
open (newunit=u_r,file="src", action="read", form="unformatted", &
open(newunit=u_w,file="dst", action="write", form="unformatted",&
do while (sz > 0)
   n_chunk = min(sz, n_buff)
   read (unit=u_r) buffer(1:n_chunk)
   write (unit=u_w) buffer(1:n_chunk)
   sz = sz - n_chunk
end do
Assign to c the result of (a xor b)
c = ieor(a,b)
Write in a new byte array c the xor result of byte arrays a and b.

a and b have the same size.
integer(kind=int8), dimension(:) :: a, b, c
! Assign values to a and b
c = ieor(a,b)
Print the contents of the list or array a on the standard output.
write (*,*) a
Remove all the elements from list x that don't satisfy the predicate p, without allocating a new list.
Keep all the elements that do satisfy p.

For languages that don't have mutable lists, refer to idiom #57 instead.
x = pack(x,.not.p())
Construct the "double precision" (64-bit) floating point number d from the mantissa m, the exponent e and the sign flag s (true means the sign is negative).
  if (s) then
    d = -m * radix(d)**e
    d =  m * radix(d)**e
  end if

Define variables a, b and c in a concise way.
Explain if they need to have the same type.
integer :: a, b, c
Extract integer value i from its binary string representation s (in radix 2)
E.g. "1101" -> 13
character (len=:), allocatable :: s
integer :: i  
s = '1101'
read (s,'(B4)') i

Assign to the variable x the string value "a" if calling the function condition returns true, or the value "b" otherwise.
if (condition())
  x = a
  x = b
end if
Replace all exact occurrences of "foo" with "bar" in the string list x
  where (x == "foo")
     x = "bar"
Print the numbers 5, 4, ..., 0 (included), one line per number.
do i=5,0,-1
  print *,i
end do
Print each index i and value x from the list items, from the last down to the first.
do i=size(items),1,-1
  print *,i,items(i)
end do
Convert the string values from list a into a list of integers b.
  allocate (b(size(a)))
  read (unit=a,fmt=*) b
Declare a new list items of string elements, containing zero elements
character(len=:), allocatable, dimension(:) :: items
allocate (items(0),source="")
Assign to the string x the value of fields (hours, minutes, seconds) of the date d, in format HH:MM:SS.
integer, dimension(8) :: d
character (len=8) :: x
call date_and_time (values=d)
write (unit=x,fmt='(I2.2,":",I2.2,":",I2.2)') d(5), d(6), d(7)
Assign to t the number of trailing 0 bits in the binary representation of the integer n.

E.g. for n=112, n is 1110000 in base 2 ⇒ t=4
t = trailz(n)
Write two functions log2d and log2u, which calculate the binary logarithm of their argument n rounded down and up, respectively. n is assumed to be positive. Print the result of these functions for numbers from 1 to 12.
program main
  implicit none
  integer :: i
  do i=1,12
     print *,i,log2d(i),log2u(i)
  end do
  integer function log2d (n)
    integer, intent(in) :: n
    log2d = bit_size(n) - 1 - leadz(n)
  end function log2d

  integer function log2u (n)
    integer, intent(in) :: n
    log2u = bit_size(n) - leadz(n-1)
  end function log2u
end program main
Pass a two-dimensional integer array a to a procedure foo and print the size of the array in each dimension. Do not pass the bounds manually. Call the procedure with a two-dimensional array.
module x
  subroutine foo(a)
    integer, dimension(:,:) :: a
    print *,size(a,1), size(a,2)
  end subroutine foo
end module x

program main
  use x
  integer, dimension(5,10) :: a
  call foo(a)
end program main
Calculate the parity p of the integer variable i : 0 if it contains an even number of bits set, 1 if it contains an odd number of bits set.
print *,poppar(i)
Assign to the string s the value of the string v repeated n times, and write it out.

E.g. v="abc", n=5 ⇒ s="abcabcabcabcabc"
character(len=:), allocatable :: s,v
integer :: n
s = repeat(v,20)
write (*,'(A)') s
Declare an argument x to a procedure foo that can be of any type. If the type of the argument is a string, print it, otherwise print "Nothing."

Test by passing "Hello, world!" and 42 to the procedure.
program main
  call foo("Hello, world!")
  call foo(42)
  subroutine foo(x)
    class(*), intent(in) :: x
    select type(x)
    type is (character(len=*))
       write (*,'(A)') x
    class default
       write (*,'(A)') "Nothing."
    end select
  end subroutine foo
end program main
Define a type vector containing three floating point numbers x, y, and z. Write a user-defined operator x that calculates the cross product of two vectors a and b.
module vect
  type, public:: vector
     real :: x,y,z
  end type vector
  public:: operator(.x.)
  interface operator(.x.)
     procedure vector_cross
  end interface operator(.x.)
  function vector_cross(a,b) result(c)
    type(vector), intent(in) :: a,b
    type(vector) :: c
    c%x = a%y*b%z - a%z*b%y
    c%y = a%z*b%x - a%x*b%z
    c%z = a%x*b%y - a%y*b%x
  end function vector_cross
end module vect
Given a floating point number r1 classify it as follows:
If it is a signaling NaN, print "This is a signaling NaN."
If it is a quiet NaN, print "This s a quiet NaN."
If it is not a NaN, print "This is a number."
  character (len=:), allocatable :: msg
  if (ieee_support_nan(r1)) then
     if (ieee_class(r1) == ieee_quiet_nan) then
        msg = "This s a quiet NaN."
     else if (ieee_class(r1) == ieee_signaling_nan) then
        msg = "This is a signaling NaN."
        msg = "This is a number."
     end if
     msg = "NaNs are not supported."
  end if
  write (*,'(A)') msg
If a variable x passed to procedure tst is of type foo, print "Same type." If it is of a type that extends foo, print "Extends type." If it is neither, print "Not related."
  subroutine tst (x)
    class (*), intent(in) :: x
    type (foo) :: y
    if (same_type_as (x,y)) then
       write (*,'(A)') "Same type."
    else if (extends_type_of (x,y)) then
       write (*,'(A)') "Extends type."
       write (*,'(A)') "Not related."
    end if
  end subroutine tst
Fizz buzz is a children's counting game, and a trivial programming task used to affirm that a programmer knows the basics of a language: loops, conditions and I/O.

The typical fizz buzz game is to count from 1 to 100, saying each number in turn. When the number is divisible by 3, instead say "Fizz". When the number is divisible by 5, instead say "Buzz". When the number is divisible by both 3 and 5, say "FizzBuzz"
  integer :: i
  logical :: fizz, buzz
  do i=1,100
     fizz = mod(i,3) == 0
     buzz = mod(i,5) == 0
     if (fizz) write (*,'(A)',advance="no") 'Fizz'
     if (buzz) write (*,'(A)',advance="no") 'Buzz'
     if (.not. fizz .and. .not. buzz) write (*,'(I0)',advance="no") i
     write (*,'(A)',advance="no") ', '
  end do
  write (*,'()')
end program
From the string s consisting of 8n binary digit characters ('0' or '1'), build the equivalent array a of n bytes.
Each chunk of 8 binary digits (2 possible values per digit) is decoded into one byte (256 possible values).
  subroutine to_s (s, a)
    use iso_fortran_env, only: int8
    character (len=*), intent(in) :: s
    integer (kind=int8), allocatable, intent(out), dimension(:) :: a
    allocate (a(len(s)/8))
    read (unit=s,fmt='(*(B8.8))') a
  end subroutine to_s
Remove the element e from the set x.

Explains what happens if e was already absent from x.
!remove 6
Remove all the elements from the map m that don't satisfy the predicate p.
Keep all the elements that do satisfy p.

Explain if the filtering happens in-place, i.e. if m is reused or if a new map is created.
  real, dimension(:), allocatable :: a
  a = pack(a,p(a))
  elemental logical function p(x)
    real, intent(in) :: x
    p = x > 0.7
  end function p

Create a new list a (or array, or slice) of size n, where all elements are integers initialized with the value 0.
integer, dimension(:), allocatable :: a

allocate(a(n),source = 0)
Given two floating point variables a and b, set a to a to a quiet NaN and b to a signalling NaN. Use standard features of the language only, without invoking undefined behavior.
 a = ieee_value (a, IEEE_QUIET_NAN)
 b = ieee_value (b, IEEE_SIGNALING_NAN)
Print a line "Char i is c" for each character c of the string s, where i is the character index of c in s (not the byte index).

Make sure that multi-byte characters are properly handled, and count for a single character.
  integer, parameter :: ucs4  = selected_char_kind ('ISO_10646')
  character(kind=ucs4,  len=30) :: s
open (output_unit, encoding='UTF-8')

do i=1,len(s)
  print *,"Char ", i, " is ", s(i:i)
end do
Set the boolean b to true if the set x contains the element e, false otherwise.
b = x(e)
Create the string s by concatenating the strings a and b.
s = a // b
Delete all the elements from index i (included) to index j (excluded) from the list items.
items = [items(:,i-1), items(j:)]
Write "Hello World and 你好" to standard output in UTF-8.
  integer, parameter :: ucs4  = selected_char_kind ('ISO_10646')
  character(kind=ucs4,  len=30) :: hello_world
  hello_world = ucs4_'Hello World and ' &
                // char (int (z'4F60'), ucs4)     &
                // char (int (z'597D'), ucs4)

  open (output_unit, encoding='UTF-8')
  write (*,*) trim (hello_world)
Given an array a containing the three values 1, 12, 42, print out
"1, 12, 42" with a comma and a space after each integer except the last one.
  integer, dimension(:), allocatable :: a
  a = [1,12,42]
  write (*,'(*(I0:", "))') a
Assign to x2 the value of string x with the last occurrence of y replaced by z.
If y is not contained in x, then x2 has the same value as x.
  character(len=:), allocatable :: x, x2, y, z

  k = index(x,y,back=.true.)
  if (k > 0) then
     x2 = x(1:k-1) // z // x(k+len(y):)
     x2 = x
  end if
Write a line of comments.

This line will not be compiled or executed.
! This is a comment.
Compute the Fibonacci sequence of n numbers using recursion.

Note that naive recursion is extremely inefficient for this task.
  recursive function f (n) result(res)
    integer, value, intent(in) :: n
    integer :: res
    if (n == 0) then
       res = 0
    else if (n <= 2) then
       res = 1
       if (mod(n,2) == 1) then
          res = f((n + 1)/2)**2 + f((n - 1)/2)**2
          res = f(n/2) * (f(n/2-1) + f(n/2 + 1))
       end if
    end if
  end function f
Given the integer x = 8, assign to the string s the value "Our sun has 8 planets", where the number 8 was evaluated from x.
  write (unit=s,fmt='(*(G0))') "Our sun has ",x," planets."
Declare an array a of integers with six elements, where the first index is 42 and consecutive elements have the indices 43, 44, 45, 46, 47.
  integer, dimension(42:47) :: a
Compute and print a^b, and a^n, where a and b are floating point numbers and n is an integer.
  print *,a**b, a**n
Create the new 2-dimensional array y containing a copy of the elements of the 2-dimensional array x.

x and y must not share memory. Subsequent modifications of y must not affect x.
integer, allocatable, dimension(:,:) :: y

y = x
Fill the byte array a with randomly generated bytes.
  real, dimension(100) :: b
  integer(int8), dimension(100) :: a
  call random_number(b)
  a = b*256 - 128
Set b to true if the lists p and q have the same size and the same elements, false otherwise.
  if (size(p,1) /= size(q,1)) then
    b = false
    b = all (p == q)
  end if
Set all the elements in the array x to the same value v
x = v
Determine the number c of elements in the list x that satisfy the predicate p.
Assign to c the value of the i-th character of the string s.

Make sure to properly handle multi-byte characters. i is the character index, which may not be equal to the byte index.
 c = s(i:i)
Compute x = b

b raised to the power of n is equal to the product of n terms b × b × ... × b
x = b ** n