[unix-history] / usr / doc / ratfor / m1

.nr PS 9
.nr VS 11
.if t .2C
.if n .ls 2
.NH
INTRODUCTION
.PP
Most programmers will agree that Fortran is
an unpleasant language to program in,
yet there are many occasions when they are forced to use it.
For example, Fortran is often the only language
thoroughly supported on the local computer.
Indeed, it is the closest thing to a universal programming language
currently available:
with care it is possible to write large, truly portable
Fortran programs[1].
Finally, Fortran is often the most ``efficient'' language
available, particularly for programs requiring much computation.
.PP
But Fortran 
.ul
is
unpleasant.
Perhaps the worst deficiency is in
the control flow
statements
_ conditional branches and loops _
which express the logic of the program.
The conditional statements in Fortran are primitive.
The Arithmetic 
.UC IF
forces the user into at least two statement numbers and
two (implied) 
.UC GOTO 's;
it leads to unintelligible code, and is eschewed by good programmers.
The Logical
.UC IF
is better, in that the test part can be stated clearly,
but hopelessly restrictive because the statement
that follows the
.UC IF
can only be one Fortran statement
(with some
.ul
further
restrictions!).
And of course there can be no
.UC ELSE
part to a Fortran
.UC IF :
there is no way to specify an alternative action if the
.UC IF
is not satisfied.
.PP
The Fortran
.UC DO
restricts the user to going forward in an arithmetic progression.
It is fine for ``1 to N in steps of 1 (or 2 or ...)'',
but there is no direct way to go backwards,
or even (in ANSI Fortran[2]) to go from 1 to
.if n N-1.
.if t N\(mi1.
And of course the
.UC DO
is useless if one's problem doesn't map into an arithmetic progression.
.PP
The result of these failings is that Fortran programs
must be written with numerous labels and branches.
The resulting code is
particularly difficult to read and understand,
and thus hard to debug and modify.
.PP
When one is faced with an unpleasant language,
a useful technique is to define
a new language that overcomes the deficiencies,
and to translate it into the unpleasant one
with a preprocessor.
This is the approach taken with 
Ratfor.
(The preprocessor idea is of course not new,
and preprocessors for Fortran are especially popular
today.
A recent listing [3] of preprocessors 
shows more than 50, of which at least half a dozen are widely available.)
Commit	Line	Data
8340f87c BJ	1	.nr PS 9
	2	.nr VS 11
	3	.if t .2C
	4	.if n .ls 2
	5	.NH
	6	INTRODUCTION
	7	.PP
	8	Most programmers will agree that Fortran is
	9	an unpleasant language to program in,
	10	yet there are many occasions when they are forced to use it.
	11	For example, Fortran is often the only language
	12	thoroughly supported on the local computer.
	13	Indeed, it is the closest thing to a universal programming language
	14	currently available:
	15	with care it is possible to write large, truly portable
	16	Fortran programs[1].
	17	Finally, Fortran is often the most ``efficient'' language
	18	available, particularly for programs requiring much computation.
	19	.PP
	20	But Fortran
	21	.ul
	22	is
	23	unpleasant.
	24	Perhaps the worst deficiency is in
	25	the control flow
	26	statements
	27	_ conditional branches and loops _
	28	which express the logic of the program.
	29	The conditional statements in Fortran are primitive.
	30	The Arithmetic
	31	.UC IF
	32	forces the user into at least two statement numbers and
	33	two (implied)
	34	.UC GOTO 's;
	35	it leads to unintelligible code, and is eschewed by good programmers.
	36	The Logical
	37	.UC IF
	38	is better, in that the test part can be stated clearly,
	39	but hopelessly restrictive because the statement
	40	that follows the
	41	.UC IF
	42	can only be one Fortran statement
	43	(with some
	44	.ul
	45	further
	46	restrictions!).
	47	And of course there can be no
	48	.UC ELSE
	49	part to a Fortran
	50	.UC IF :
	51	there is no way to specify an alternative action if the
	52	.UC IF
	53	is not satisfied.
	54	.PP
	55	The Fortran
	56	.UC DO
	57	restricts the user to going forward in an arithmetic progression.
	58	It is fine for ``1 to N in steps of 1 (or 2 or ...)'',
	59	but there is no direct way to go backwards,
	60	or even (in ANSI Fortran[2]) to go from 1 to
	61	.if n N-1.
	62	.if t N\(mi1.
	63	And of course the
	64	.UC DO
65	is useless if one's problem doesn't map into an arithmetic progression.
66	.PP
67	The result of these failings is that Fortran programs
68	must be written with numerous labels and branches.
69	The resulting code is
70	particularly difficult to read and understand,
71	and thus hard to debug and modify.
72	.PP
73	When one is faced with an unpleasant language,
74	a useful technique is to define
75	a new language that overcomes the deficiencies,
76	and to translate it into the unpleasant one
77	with a preprocessor.
78	This is the approach taken with
79	Ratfor.
80	(The preprocessor idea is of course not new,
81	and preprocessors for Fortran are especially popular
82	today.
83	A recent listing [3] of preprocessors
84	shows more than 50, of which at least half a dozen are widely available.)