The syntax of the Difmap interface language.

LOOPING CONSTRUCTS
------------------

 The DO loop
 -----------

 do VARIABLE = START, END, INCREMENT
   STATEMENTS
 end do

 Where VARIABLE must name an integer or float variable. START, END and
 INCREMENT are scalar arithmetic expressions. If INCREMENT is omitted,
 1 is substituted. The loop performs N iterations, where:

  N = MAX(0, INT((START-END+INCREMENT)/INCREMENT))

 At the start of the n'th iteration, where n=0..N-1, VARIABLE is
 assigned the value:

  VARIABLE = START + n * INCREMENT

 After the final iteration, VARIABLE has the value:

  VARIABLE = END + INCREMENT

 You may use the "break" statement to abort the loop before the final
 iteration, or use the "continue" statement to start the next
 iteration before all statements in the loop have been executed.

 DO loop example.
 ----------------

  0>float angle
  0>do angle=0,360,45
  1>  print "sin(", angle, ") =", sin(angle*pi/180)
  1>end do
  sin( 0 ) = 0 
  sin( 45 ) = 0.707107 
  sin( 90 ) = 1 
  sin( 135 ) = 0.707107 
  sin( 180 ) = -8.74228e-08 
  sin( 225 ) = -0.707107 
  sin( 270 ) = -1 
  sin( 315 ) = -0.707107 
  sin( 360 ) = 1.74846e-07 

  WHILE loops.
  -----------
 

DECLARATIONS.
------------
 Variables can be declared at any time. An existing variable can be
 re-declared as long as the new declaration doesn't change its type.
 All variables are global. There is no way to declare a local variable.

 Variable declarations take the form:

  Type Variable_Name
  Type Variable_Name(dim_1, dim_2...)

 Where type is one of:

  integer
  float
  string

 and where Variable_Name may be any unique combination of underscore
 and alphanumeric characters. However, the first character must be
 alphabetic.

 and where dim_n is the dimension of the variable along axis n. There
 may be between 0 and 3 dimensions.

 For example:

  integer i
  float f
  float grid(4,5)

 The varlist command allows you to see how existing variables are
 currently declared.

 The dimensions of existing array variables may be changed by
 re-declaration.

 For example in,

  float image_data(400)
  float image_data(2)

 The second line changes the image_data() variable from having 400
 elements to having just 2 elements.

 The dimensions of array variables can also be changed by assignment,
 as shown below.

ASSIGNMENT
----------
 Assigning the value of an expression to a variable is done with the
 '=' operator.

 For example:

  x = sin(pi/2) + cos(sqrt(2.5))

 If the expression on the right hand side of the assignment is an
 array expression, then the variable being assigned must have the same
 number of dimensions as that expression. However the variable on the
 left hand side of the expression will be re-declared if possible to
 have the same number of elements on each dimension as the expression
 that is being assigned to it.

 For example:

  0>float my_array(10)
  0>print my_array
  0 0 0 0 0 0 0 0 0 0
  0>my_array = 1,2,3,4,5
  0>print my_array
  1 2 3 4 5

 However, if the variable expression on the left hand side of the
 assignment cites an explicit sub-range of indexes, the number of
 elements in the expression on the right hand side must match the
 number of elements in the sub-range:

 For example:

  0>float my_array(10)
  0>my_array(3:5) = 1,2
  Illegal assignment due to differing array bounds.
  0>my_array(3:5) = 1,2,3
  print my_array
  0 0 1 2 3 0 0 0 0 0 
  0>

DO LOOPS
--------

DO variable=start,end,increment
  statements
END DO

do i=1,10
  print i
end do

x=2.3
while(x < 100)
  print f
end while