A lisp macro virgin tells all
I finished my first lisp macro, and I want to tell the world.
I’ll talk about what a lisp macro is, and what makes it unique in the world of programming, how it’s a technique only possible in lisp. I’ll then take you through an example.
So firstly, what’s a lisp macro, and why would you want to write one?
So, you may have seen lisp programs before, and you’ll recognise them instantly — Larry Wall, the inventor of Perl, said they had all the aesthetic appeal of a bowl of porridge mixed with toenail clippings;
(defun accumulate (combiner lst initial)
(let ((accum initial))
(dolist (i lst)
(setf accum (funcall combiner accum i)))
accum))
He has a point. They are butt-ugly. But hell, the best he came up with is Perl, so he can $_@++ right off. (I’m pretty sure that’s valid Perl, too
)
It’s ugly, in an aesthetic way, but it’s amazingly practical. It’s got an engineering beauty to it. If you look at that snippet above, you’ll notice that the whole program is made out of exactly three types of symbols;
- open parenthesis:
( - close parenthesis:
) - symbols, like
defun,accumandsetf
All simple lisp programs are like this. Just brackets to group stuff together, and stuff that needs grouping. Compare that with C#, where you might find;
- parenthesis for;
- function calls;
print("hello") - special forms;
using(OdbcConnection con = ...)
- function calls;
- semi-colon to end statements;
int x = 1; - curly brackets for;
- code blocks;
{ /* code block */ } - array initialisers;
string[] words = { “hello”, “world” };
- code blocks;
- square brackets for array indexing;
x[3] = 4;
and the list goes on. I gave up because there are too many to list.
So lisp has this seriously small syntactic footprint. You can have a thing, or a group of things in brackets. It’s simple. It’s so simple that you can start doing crazy stuff in lisp that you just can’t do otherwise. That crazy stuff goes by the name of macros.
I can write a program that takes a chunk of lisp (remember, just a thing or a list of things), cuts it up, and reassembles it. That creates new lisp code.
So imagine you do a lot of work on three-dimensional arrays. You find yourself, over and over, writing nested loops that say;
for x in range(100):
for y in range(100):
for z in range(100):
# do something to matrix[x,y,z]
And frankly, you’re bored of typing it over and over. What you really want to do is something like;
for {x 100, y 100, z 100}:
# do something to matrix[x,y,z]
You want a brand new bit of syntax for multiple-value looping. Can you add it to python? Nope. C? Nope. Java? Nope.
But now look at the lisp version;
I could, theoretically, write this
(domanytimes (x 100 y 100 z 100)
body)
and, because it’s just a list of stuff, I can chop and change that into this new bit of lisp;
(dotimes (x 100)
(dotimes (y 100)
(dotimes (z 100)
body)))
I’ll show you how in a second, but notice what’s possible — I can write my own looping construct (domanytimes) and lisp will rewrite it into many simpler looping construct (the built-in dotimes).
Is that particularly special? Well, yeah. I’ve written new syntax. I’ve defined a new way of looping that is no different from the standard loops. I’ve basically added something new to the language. Lisp is now better at dealing with multi-dimensional loops. Try adding a new loop to ruby, or javascript. Make python understand
for x in range(100), y in range(100), z in range(100):
# body here
and you’ll find you can’t.
So I’ve made my version of lisp a bit better at handling loops. If I were writing database code, I could make lisp better at writing SQL statements or data access layers. C# recently got built-in DAL logic with LINQ, and it’s great, but only the C# team can write it. Whereas a lisper could write this sort of code;
(sql-select (ID NAME) from PROJECT where (DUEDATE > TODAY))
and it’s do basically the same thing as LINQ.
So that’s the why’s and wherefores. Here’s the how of the domanytimes macro.
domanytimes takes two parts; the loop variables (x 100 y 100 z 100) and whatever body you want to execute. We’re going to write a program that skims two elements from the front of the loop variables (say, x and 100) and uses them to write a built-in dotimes loop; so a program which converts
(domanytimes (x 100 y 100) body)
into
(dotimes (x 100)
(domanytimes (y 100) body))
and then again to give you
(dotimes (x 100)
(dotimes (y 100)
body))
Here’s the domanytimes macro, in all it’s eye-bleeding horror;
(defmacro domanytimes (loop-list &body body)
"allows you to write (domanytimes (x 10 y 10) ...)
instead of (dotimes (x 10) (dotimes (y 10)) body ))"
(if (eq (length loop-list) 0)
;; we have our form to execute
`(progn ,@body)
;; we have more loops to arrange
(let ((fst (car loop-list))
(snd (cadr loop-list))
(rst (cddr loop-list)))
`(dotimes (,fst ,snd)
(domanytimes ,rst ,@body)))))
There. Wasn’t that fun?
It looks nasty, I know. All lisp looks nasty. But it’s actually created something new in the language. As far as I understand it, lisp has survived for fifty years basically because the macro system lets you write macros which can add any new kind of syntax you like. You can write knock up a set of macros to implement OO, and suddenly lisp is OO. You can know up macros for manipulating lazy lists, and suddenly lisp has a lazy evaluation. You can knock up data access layer macros, and it’s got a version of LINQ. There seems to be nothing you can’t hack lisp into being.
And if you want to know how the hell that works, I’d recommend Practical Common Lisp, which is online and free.
April 14th, 2008 at 4:47 am
I can’t speak for Python, but it is possible to implement a dotimes without macros. Here’s my version:
http://blog.jrock.us/articles/Macros%20are%20probably%20not%20necessary.pod
And BTW, lisp is beautiful. The macro you wrote is not ugly at all, it just takes some mental adjustment from years of C-style-syntax braindamage
April 14th, 2008 at 5:19 am
in ruby:
domanytimes (x => 100, y => 100, z => 100) { #body }
I know htis isn’t new syntax, but I’m wondering how many few cases there are where macros really give you something better than what you can do with meta programming in ruby.
April 14th, 2008 at 9:27 am
Erm, wrong.
in python (might be a bit shorter way, but this took me 20 seconds to puzzle out)
for tup in ((xx,yy,zz) for xx in range(10) for yy in range(10) for zz in range(10)): print tup
As a bonus, this one is a generator.
April 14th, 2008 at 10:43 am
Jonathan Rockway: Clearly not the same thing.
First the cosmetic stuff: in your perl solution you have the bounds of the iteration after the code. That’s harder to understand.
But there’s another bug. You made the x, y and z variables dinamically bound (at least I think that’s what “our” does in perl). So you can’t capture the lexical environment in a lambda and just send it to the outside. That’s a big no-no.
April 14th, 2008 at 12:10 pm
@cory: ——–
So how does this really do what’s needed? First, you’ve got a fixed number of inner loops — in this case, three. How do you extend it to arbitrary length? The macro I gave takes any number of pairs. So you can use domanytimes for 2-, 3-, and 4-dimensional matrices by calling;
Also, how limited is the code you can pass in? Could you pass in variables declared outside the call? eg
In this case, the ‘caption’ variable is used as a parameter. How do you
def:something in python that will be able to use that variable?So the solution needs be a python construct called domanytimes (say, a function) which
@jay: ——–
So this is a function which takes a dictionary of loop variables and executes #body? Can the body include plain old references to the loop variables? that is, can you write;
}
@fil: ——–
I think that’s also the issue with cory’s python code (see the charity calendar code at the top of this comment.)
April 14th, 2008 at 12:38 pm
Nice post Steve! Clear and well-written.
April 14th, 2008 at 1:55 pm
What are you using to edit your lisp code? I ask because its indented a bit strangely and lisp indentation is pretty standard.
April 14th, 2008 at 3:35 pm
“C# recently got built-in DAL logic with LINQ, and it’s great, but only the C# team can write it. “
Here is my LINQ for Scheme syntax I am working on.
http://xacc.wordpress.com/2008/04/13/linq-for-r6rs-scheme-take-5/
Works like a charm
April 14th, 2008 at 4:41 pm
@Ben:
Thanks! I’m trying to document my way through both to clarify my own thinking, and because it’s harder to find beginner lisp stuff.
@David:
I’m using emacs (Peter Siebel’s LispBox distro) to develop, then I assembled my post in sublime text, which I’d recommend as a very pretty python-extensible text editor.
One of the examples in the comments was just handwritten, so is probably a bit funny.
@leppie:
Very nice! And thanks for turning me on to IronLisp/IronScheme! Now all I have to do is convince my colleagues that they want to work in lisp, and I should make rapid lisp progress.
April 14th, 2008 at 6:55 pm
@Steve:
I think that your chances of getting us to use ironLisp at work are rather less than slim!
April 14th, 2008 at 8:15 pm
@Spencer:
yeah, I think the odds are pretty long…
April 15th, 2008 at 5:24 am
this actually works
I had to separate the variable names from the values. The implementation is a little longer because DoManyTimes is a class and maintains state to pass to the block.
DoManyTimes.run([3,3,3]) do |x,y,z| puts “#{x} #{y} #{z}” end
You can use an arbitrary number of indices. DoManyTimes([10,10,10,10,10]) do {x,y,z,a,b| … end
I’m sure there is a better way to do this in ruby, but I got sidetracked because I realized it would be a lot easier in javascript, I think
April 15th, 2008 at 7:26 am
@Steve:
Even I have to use it secretly at work!
April 15th, 2008 at 5:09 pm
Your macro would be slightly prettier (and a little more Lisp-idiomatic) like this:
(defmacro domanytimes (loop-list &body body) (if (null loop-list)(progn ,@body) (destructuring-bind (fst rst . tail) loop-list(dotimes (,fst , rst) (domanytimes ,tail ,@body)))))Untested, and I hope I got the markup right…
April 15th, 2008 at 7:19 pm
@bob
Thanks, Bob!
I haven’t got to destructuring-bind yet…
May 13th, 2008 at 7:05 pm
Try this one, which is more in the spirit of dotimes. If there are zero loop-list items, it runs the body once and returns nil. Else it works like yours except if there’s an extra single expression at the end of the loop-list it evaluates and returns it afterwards; else it returns nil.
May 13th, 2008 at 7:24 pm
@Foo:
Thanks for taking the time to code this, Foo. I appreciate it.
June 24th, 2008 at 11:03 am
(….(DUEDATE > TODAY))
???
(….(> DUEDATE TODAY))
Excellent work, otherwise.
June 25th, 2008 at 12:53 am
timbo: –
I actually meant it this way round, since this is the form you’d see in an SQL statement. Inside a macro, you can rewrite the list
into your second form, within the body of the macro. SQL programmers don’t get as thrown.
June 26th, 2008 at 9:03 pm
can’t u just do the same thing with a function that takes a closure as an arg. You would have the loop in this function call the closure.
June 27th, 2008 at 12:35 am
jeff: –
I don’t think so in all cases, but it certainly works for some.
Also, it’s just prettier to write
than it is to write
(domanytimes (x 10 y 10) (lambda (a b) (print (* a b))))