Discussion:
Slow Loop (alternatives in lisp?)
(too old to reply)
B. Pym
2024-07-22 18:29:17 UTC
Permalink
(loop while (foo-p) do .... )
Gauche Scheme

(while (read) (print "True enough."))
2
True enough.
#t
True enough.
'yes
True enough.
#f

(while (read) => x (print x " is truer than you think."))
"why"
why is truer than you think.
88
88 is truer than you think.
0
0 is truer than you think.
#f




Paul Graham:

I consider Loop one of the worst flaws in CL, and an example
to be borne in mind by both macro writers and language designers.

[In "ANSI Common Lisp", Graham makes the following comments:]

The loop macro was originally designed to help inexperienced
Lisp users write iterative code. Instead of writing Lisp code,
you express your program in a form meant to resemble English,
and this is then translated into Lisp. Unfortunately, loop is
more like English than its designers ever intended: you can
use it in simple cases without quite understanding how it
works, but to understand it in the abstract is almost
impossible.
....
the ANSI standard does not really give a formal specification
of its behavior.
....
The first thing one notices about the loop macro is that it
has syntax. A loop expression contains not subexpressions but
clauses. The clauses are not delimited by parentheses;
instead, each kind has a distinct syntax. In that, loop
resembles traditional Algol-like languages. But the other
distinctive feature of loop, which makes it as unlike Algol as
Lisp, is that the order in which things happen is only
loosely related to the order in which the clauses occur.
....
For such reasons, the use of loop cannot be recommended.


Dan Weinreb, one of the designers of Common Lisp:

... the problem with LOOP was that it turned out to be hard to
predict what it would do, when you started using a lot of
different facets of LOOP all together. This is a serious problem
since the whole idea of LOOP was to let you use many facets
together; if you're not doing that, LOOP is overkill.


Barry Margolin:

My recommendation is based on seeing many question in the past
of the form "What happens if you use both XXX and YYY in the
same LOOP?" The unfortunate fact is that when we were writing
the standard we didn't have time to nail down all the possible
interactions between different LOOP features, so many of these
are not well specified. And even if we did get it right in
the standard, it's likely to be difficult to find them and I
wouldn't trust that all implementors got it right (many of
those questions were probably from implementors, trying to
figure out what they were supposed to do). And even if they
all got it right, someone reading your code may not be able to
figure it out.

So, with all those potential problems, my feeling is that if
you have to ask, it's probably better to use something other
than LOOP.
3. Loop is very powerful, granted, and many people are trying to
argue that "you can do so much with loop that it's unreadable."
This is not an argument.
But it is! Because any use of LOOP has the potential to be
unreadable, the reader must read it carefully to verify that
it's just one of the cases that doesn't require careful
reading!


John Foderaro:

I'm not trying to join a debate on loop. I just wanted to present
the other side of [the issue so that] the intelligent people can
then weigh the arguments on both sides.

I'm not suggesting that loop can be fixed either by adding
parenthesis or coming up with ways of indenting it to make it
understandable. It's a lost cause.

...

Another great example from kmp:

=== from kmp

For example, you might think
(loop with i = (random 100) for x from 1 to 10 do (print (list i x)))
and
(loop for i = (random 100) for x from 1 to 10 do (print (list i x)))
meant the same in English, [but they don't do the same thing in loop]

=== end kmp

loop lulls you into thinking that you understand the program since
you understand English. Make no mistake about it, loop is its
own language. If you use it you condem everyone who reads the
code to also learn the loop language.
B. Pym
2024-07-22 18:29:50 UTC
Permalink
(loop while (foo-p) do .... )
Gauche Scheme

(while (read) (print "True enough."))
2
True enough.
#t
True enough.
'yes
True enough.
#f

(while (read) => x (print x " is truer than you think."))
"why"
why is truer than you think.
88
88 is truer than you think.
0
0 is truer than you think.
#f




Paul Graham:

I consider Loop one of the worst flaws in CL, and an example
to be borne in mind by both macro writers and language designers.

[In "ANSI Common Lisp", Graham makes the following comments:]

The loop macro was originally designed to help inexperienced
Lisp users write iterative code. Instead of writing Lisp code,
you express your program in a form meant to resemble English,
and this is then translated into Lisp. Unfortunately, loop is
more like English than its designers ever intended: you can
use it in simple cases without quite understanding how it
works, but to understand it in the abstract is almost
impossible.
....
the ANSI standard does not really give a formal specification
of its behavior.
....
The first thing one notices about the loop macro is that it
has syntax. A loop expression contains not subexpressions but
clauses. The clauses are not delimited by parentheses;
instead, each kind has a distinct syntax. In that, loop
resembles traditional Algol-like languages. But the other
distinctive feature of loop, which makes it as unlike Algol as
Lisp, is that the order in which things happen is only
loosely related to the order in which the clauses occur.
....
For such reasons, the use of loop cannot be recommended.


Dan Weinreb, one of the designers of Common Lisp:

... the problem with LOOP was that it turned out to be hard to
predict what it would do, when you started using a lot of
different facets of LOOP all together. This is a serious problem
since the whole idea of LOOP was to let you use many facets
together; if you're not doing that, LOOP is overkill.


Barry Margolin:

My recommendation is based on seeing many question in the past
of the form "What happens if you use both XXX and YYY in the
same LOOP?" The unfortunate fact is that when we were writing
the standard we didn't have time to nail down all the possible
interactions between different LOOP features, so many of these
are not well specified. And even if we did get it right in
the standard, it's likely to be difficult to find them and I
wouldn't trust that all implementors got it right (many of
those questions were probably from implementors, trying to
figure out what they were supposed to do). And even if they
all got it right, someone reading your code may not be able to
figure it out.

So, with all those potential problems, my feeling is that if
you have to ask, it's probably better to use something other
than LOOP.
3. Loop is very powerful, granted, and many people are trying to
argue that "you can do so much with loop that it's unreadable."
This is not an argument.
But it is! Because any use of LOOP has the potential to be
unreadable, the reader must read it carefully to verify that
it's just one of the cases that doesn't require careful
reading!


John Foderaro:

I'm not trying to join a debate on loop. I just wanted to present
the other side of [the issue so that] the intelligent people can
then weigh the arguments on both sides.

I'm not suggesting that loop can be fixed either by adding
parenthesis or coming up with ways of indenting it to make it
understandable. It's a lost cause.

...

Another great example from kmp:

=== from kmp

For example, you might think
(loop with i = (random 100) for x from 1 to 10 do (print (list i x)))
and
(loop for i = (random 100) for x from 1 to 10 do (print (list i x)))
meant the same in English, [but they don't do the same thing in loop]

=== end kmp

loop lulls you into thinking that you understand the program since
you understand English. Make no mistake about it, loop is its
own language. If you use it you condem everyone who reads the
code to also learn the loop language.
Kaz Kylheku
2024-07-23 01:24:37 UTC
Permalink
Post by B. Pym
(loop while (foo-p) do .... )
Gauche Scheme
(while (read) (print "True enough."))
2
True enough.
#t
True enough.
'yes
True enough.
#f
(while (read) => x (print x " is truer than you think."))
[ ... ]
Post by B. Pym
I consider Loop one of the worst flaws in CL, and an example
to be borne in mind by both macro writers and language designers.
But the above => cruft is exactly like LOOP clause syntax.

It might as well be

(while (read) with x do (print ...))

More Lispy way:

(whilet ((x (read)))
...)
steve g
2024-08-10 18:09:54 UTC
Permalink
"B. Pym" <***@here-nor-there.org> writes:

< > * there is already a WHILE in Common Lisp. No need to invent a new one:
< >
< > (loop while (foo-p) do .... )
Post by B. Pym
Gauche Scheme
(while (read) (print "True enough."))
2
True enough.
#t
True enough.
'yes
True enough.
#f
(while (read) => x (print x " is truer than you think."))
"why"
why is truer than you think.
88
88 is truer than you think.
0
0 is truer than you think.
#f
....
the ANSI standard does not really give a formal specification
of its behavior.
....
not an ANSI issue; it is a machine issue. you cannot add new registers
to a CPU.
Post by B. Pym
The first thing one notices about the loop macro is that it
has syntax. A loop expression contains not subexpressions but
clauses. The clauses are not delimited by parentheses;
instead, each kind has a distinct syntax. In that, loop
resembles traditional Algol-like languages. But the other
distinctive feature of loop, which makes it as unlike Algol as
Lisp, is that the order in which things happen is only
loosely related to the order in which the clauses occur.
....
i hope you are not talking about order of evaluation - this is a scheme
problem; not with common lisp.
Post by B. Pym
For such reasons, the use of loop cannot be recommended.
it works, it is repeatable, and it is predictable. I typically use loop
that came from MIT lisp machine.
Post by B. Pym
... the problem with LOOP was that it turned out to be hard to
predict what it would do,
?
Post by B. Pym
when you started using a lot of
different facets of LOOP all together. This is a serious problem
since the whole idea of LOOP was to let you use many facets
together; if you're not doing that, LOOP is overkill.
My recommendation is based on seeing many question in the past
of the form "What happens if you use both XXX and YYY in the
same LOOP?" The unfortunate fact is that when we were writing
the standard we didn't have time to nail down all the possible
interactions between different LOOP features, so many of these
are not well specified. And even if we did get it right in
the standard, it's likely to be difficult to find them and I
wouldn't trust that all implementors got it right (many of
those questions were probably from implementors, trying to
figure out what they were supposed to do). And even if they
all got it right, someone reading your code may not be able to
figure it out.
So, with all those potential problems, my feeling is that if
you have to ask, it's probably better to use something other
than LOOP.
< > 3. Loop is very powerful, granted, and many people are trying to
< > argue that "you can do so much with loop that it's unreadable."
< > This is not an argument.
Post by B. Pym
But it is! Because any use of LOOP has the potential to be unreadable,
have you ever tried programming in PERL or assembler?
Post by B. Pym
I'm not trying to join a debate on loop. I just wanted to present
the other side of [the issue so that] the intelligent people can
then weigh the arguments on both sides.
I would consider the condition system and format macros to be more
complicated. You need to use a quick cheat sheet. not too hard.
Post by B. Pym
I'm not suggesting that loop can be fixed either by adding
parenthesis or coming up with ways of indenting it to make it
understandable. It's a lost cause.
1986 was a long time ago. why not reinvent a language that has call/cc!
Post by B. Pym
...
=== from kmp
For example, you might think
(loop with i = (random 100) for x from 1 to 10 do (print (list i x)))
and
(loop for i = (random 100) for x from 1 to 10 do (print (list i x)))
meant the same in English, [but they don't do the same thing in loop]
using `with' in a loop clause allows one to use variables.
Post by B. Pym
=== end kmp
loop lulls you into thinking that you understand the program since
you understand English. Make no mistake about it, loop is its
own language.
If you use it you condem everyone who reads the
code to also learn the loop language.
why the hatred of iteration? i can already tell that you are a scheme
programmer. anything that is not tail recursive is evil for some reason.
I did understand this.


the biggest problem with loop is implementing it.
B. Pym
2024-08-17 09:30:28 UTC
Permalink
(defun distribution1 (items values test)
(let ((table (make-hash-table :test test)))
(loop for item in items
for value in values
do (incf (gethash item table 0) value))
(let ((items-list nil))
(maphash (lambda (item sum-value)
(push (cons item sum-value) items-list))
table)
(sort items-list #'> :key #'cdr))))
CL-USER 58 > (distribution1 '("a" "b" "c" "b" "a" "f" "e" "g"
"h" "k" "z" "k" "r" "u" "f")
'(1 5 8 7 14 8 3 7 9 4 3 21 5 7 9)
#'equal)
(("k" . 25) ("f" . 17) ("a" . 15) ("b" . 12) ("h" . 9) ("c" . 8)
("g" . 7) ("u" . 7) ("r" . 5) ("e" . 3) ("z" . 3))
newLISP

Let's simply use an association list.

(macro (ainc! Alist Key Value Function Deflt)
(local (E-Message Val Func Def)
(setq Func Function)
(if (true? Func)
(setq Val Value)
(begin (setq Func +) (setq Val (or Value 1))))
(setq Def Deflt)
(if (= nil Def) (setq Def 0))
(unless
(catch
(setf (assoc Key Alist)
(list ($it 0) (Func Val ($it 1))))
'E-Message)
(setf Alist (cons (list Key (Func Val Def)) Alist)))))

(define (distribution1 items vals)
(let (table '())
(dolist (it items) (ainc! table it (pop vals)))
(sort table (fn (a b) (> (a 1) (b 1))))))

(distribution1
'("a" "b" "c" "b" "a" "f" "e" "g" "h" "k" "z" "k" "r" "u" "f")
'(1 5 8 7 14 8 3 7 9 4 3 21 5 7 9))

===>
(("k" 25) ("f" 17) ("a" 15) ("b" 12) ("h" 9) ("c" 8) ("g" 7)
("u" 7) ("r" 5) ("e" 3) ("z" 3))

Loading...