Tutorial

Welcome to the Hy tutorial!

In a nutshell, Hy is a Lisp dialect, but one that converts its structure into Python ... literally a conversion into Python's abstract syntax tree! (Or to put it in more crude terms, Hy is lisp-stick on a Python!)

This is pretty cool because it means Hy is several things:

  • A Lisp that feels very Pythonic
  • For Lispers, a great way to use Lisp's crazy powers but in the wide world of Python's libraries (why yes, you now can write a Django application in Lisp!)
  • For Pythonistas, a great way to start exploring Lisp, from the comfort of Python!
  • For everyone: a pleasant language that has a lot of neat ideas!

Basic intro to Lisp for Pythonistas

Okay, maybe you've never used Lisp before, but you've used Python!

A "hello world" program in Hy is actually super simple. Let's try it:

(print "hello world")

See? Easy! As you may have guessed, this is the same as the Python version of:

print "hello world"

To add up some super simple math, we could do:

(+ 1 3)

Which would return 4 and would be the equivalent of:

1 + 3

What you'll notice is that the first item in the list is the function being called and the rest of the arguments are the arguments being passed in. In fact, in Hy (as with most Lisps) we can pass in multiple arguments to the plus operator:

(+ 1 3 55)

Which would return 59.

Maybe you've heard of Lisp before but don't know much about it. Lisp isn't as hard as you might think, and Hy inherits from Python, so Hy is a great way to start learning Lisp. The main thing that's obvious about Lisp is that there's a lot of parentheses. This might seem confusing at first, but it isn't so hard. Let's look at some simple math that's wrapped in a bunch of parentheses that we could enter into the Hy interpreter:

(setv result (- (/ (+ 1 3 88) 2) 8))

This would return 38. But why? Well, we could look at the equivalent expression in python:

result = ((1 + 3 + 88) / 2) - 8

If you were to try to figure out how the above were to work in python, you'd of course figure out the results by solving each inner parenthesis. That's the same basic idea in Hy. Let's try this exercise first in Python:

result = ((1 + 3 + 88) / 2) - 8
# simplified to...
result = (92 / 2) - 8
# simplified to...
result = 46 - 8
# simplified to...
result = 38

Now let's try the same thing in Hy:

(setv result (- (/ (+ 1 3 88) 2) 8))
; simplified to...
(setv result (- (/ 92 2) 8))
; simplified to...
(setv result (- 46 8))
; simplified to...
(setv result 38)

As you probably guessed, this last expression with setv means to assign the variable "result" to 38.

See? Not too hard!

This is the basic premise of Lisp. Lisp stands for "list processing"; this means that the structure of the program is actually lists of lists. (If you're familiar with Python lists, imagine the entire same structure as above but with square brackets instead, any you'll be able to see the structure above as both a program and a data structure.) This is easier to understand with more examples, so let's write a simple Python program, test it, and then show the equivalent Hy program:

def simple_conversation():
    print "Hello!  I'd like to get to know you.  Tell me about yourself!"
    name = raw_input("What is your name? ")
    age = raw_input("What is your age? ")
    print "Hello " + name + "!  I see you are " + age + " years old."

simple_conversation()

If we ran this program, it might go like:

Hello!  I'd like to get to know you.  Tell me about yourself!
What is your name? Gary
What is your age? 38
Hello Gary!  I see you are 38 years old.

Now let's look at the equivalent Hy program:

(defn simple-conversation []
   (print "Hello!  I'd like to get to know you.  Tell me about yourself!")
   (setv name (raw-input "What is your name? "))
   (setv age (raw-input "What is your age? "))
   (print (+ "Hello " name "!  I see you are "
              age " years old.")))

(simple-conversation)

If you look at the above program, as long as you remember that the first element in each list of the program is the function (or macro... we'll get to those later) being called and that the rest are the arguments, it's pretty easy to figure out what this all means. (As you probably also guessed, defn is the Hy method of defining methods.)

Still, lots of people find this confusing at first because there's so many parentheses, but there are plenty of things that can help make this easier: keep indentation nice and use an editor with parenthesis matching (this will help you figure out what each parenthesis pairs up with) and things will start to feel comfortable.

There are some advantages to having a code structure that's actually a very simple data structure as the core of Lisp is based on. For one thing, it means that your programs are easy to parse and that the entire actual structure of the program is very clearly exposed to you. (There's an extra step in Hy where the structure you see is converted to Python's own representations ... in "purer" Lisps such as Common Lisp or Emacs Lisp, the data structure you see in the code and the data structure that is executed is much more literally close.)

Another implication of this is macros: if a program's structure is a simple data structure, that means you can write code that can write code very easily, meaning that implementing entirely new language features can be very fast. Previous to Hy, this wasn't very possible for Python programmers ... now you too can make use of macros' incredible power (just be careful to not aim them footward)!

Hy is a Lisp-flavored Python

Hy converts to Python's own abstract syntax tree, so you'll soon start to find that all the familiar power of python is at your fingertips.

You have full access to Python's data types and standard library in Hy. Let's experiment with this in the hy interpreter:

=> [1 2 3]
[1, 2, 3]
=> {"dog" "bark"
... "cat" "meow"}
{'dog': 'bark', 'cat': 'meow'}
=> (, 1 2 3)
(1, 2, 3)
=> #{3 1 2}
{1, 2, 3}
=> 1/2
Fraction(1, 2)

Notice the last two lines: Hy has a fraction literal like Clojure.

If you start Hy like this (a shell alias might be helpful):

$ hy --repl-output-fn=hy.contrib.hy-repr.hy-repr

the interactive mode will use hy-repr instead of Python's native repr function to print out values, so you'll see values in Hy syntax rather than Python syntax:

=> [1 2 3]
[1 2 3]
=> {"dog" "bark"
... "cat" "meow"}
{"dog" "bark" "cat" "meow"}

If you are familiar with other Lisps, you may be interested that Hy supports the Common Lisp method of quoting:

=> '(1 2 3)
(1L 2L 3L)

You also have access to all the built-in types' nice methods:

=> (.strip " fooooo   ")
"fooooo"

What's this? Yes indeed, this is precisely the same as:

" fooooo   ".strip()

That's right---Lisp with dot notation! If we have this string assigned as a variable, we can also do the following:

(setv this-string " fooooo   ")
(this-string.strip)

What about conditionals?:

(if (try-some-thing)
  (print "this is if true")
  (print "this is if false"))

As you can tell above, the first argument to if is a truth test, the second argument is the body if true, and the third argument (optional!) is if false (ie. else).

If you need to do more complex conditionals, you'll find that you don't have elif available in Hy. Instead, you should use something called cond. In Python, you might do something like:

somevar = 33
if somevar > 50:
    print "That variable is too big!"
elif somevar < 10:
    print "That variable is too small!"
else:
    print "That variable is jussssst right!"

In Hy, you would do:

(setv somevar 33)
(cond
 [(> somevar 50)
  (print "That variable is too big!")]
 [(< somevar 10)
  (print "That variable is too small!")]
 [True
  (print "That variable is jussssst right!")])

What you'll notice is that cond switches off between a statement that is executed and checked conditionally for true or falseness, and then a bit of code to execute if it turns out to be true. You'll also notice that the else is implemented at the end simply by checking for True -- that's because True will always be true, so if we get this far, we'll always run that one!

You might notice above that if you have code like:

(if some-condition
  (body-if-true)
  (body-if-false))

But wait! What if you want to execute more than one statement in the body of one of these?

You can do the following:

(if (try-some-thing)
  (do
    (print "this is if true")
    (print "and why not, let's keep talking about how true it is!"))
  (print "this one's still simply just false"))

You can see that we used do to wrap multiple statements. If you're familiar with other Lisps, this is the equivalent of progn elsewhere.

Comments start with semicolons:

(print "this will run")
; (print "but this will not")
(+ 1 2 3)  ; we'll execute the addition, but not this comment!

Hashbang (#!) syntax is supported:

#! /usr/bin/env hy
(print "Make me executable, and run me!")

Looping is not hard but has a kind of special structure. In Python, we might do:

for i in range(10):
    print "'i' is now at " + str(i)

The equivalent in Hy would be:

(for [i (range 10)]
  (print (+ "'i' is now at " (str i))))

You can also import and make use of various Python libraries. For example:

(import os)

(if (os.path.isdir "/tmp/somedir")
  (os.mkdir "/tmp/somedir/anotherdir")
  (print "Hey, that path isn't there!"))

Python's context managers (with statements) are used like this:

(with [f (open "/tmp/data.in")]
  (print (.read f)))

which is equivalent to:

with open("/tmp/data.in") as f:
    print f.read()

And yes, we do have List comprehensions! In Python you might do:

odds_squared = [
  pow(num, 2)
  for num in range(100)
  if num % 2 == 1]

In Hy, you could do these like:

(setv odds-squared
  (list-comp
    (pow num 2)
    (num (range 100))
    (= (% num 2) 1)))
; And, an example stolen shamelessly from a Clojure page:
; Let's list all the blocks of a Chessboard:

(list-comp
  (, x y)
  (x (range 8)
   y "ABCDEFGH"))

; [(0, 'A'), (0, 'B'), (0, 'C'), (0, 'D'), (0, 'E'), (0, 'F'), (0, 'G'), (0, 'H'),
;  (1, 'A'), (1, 'B'), (1, 'C'), (1, 'D'), (1, 'E'), (1, 'F'), (1, 'G'), (1, 'H'),
;  (2, 'A'), (2, 'B'), (2, 'C'), (2, 'D'), (2, 'E'), (2, 'F'), (2, 'G'), (2, 'H'),
;  (3, 'A'), (3, 'B'), (3, 'C'), (3, 'D'), (3, 'E'), (3, 'F'), (3, 'G'), (3, 'H'),
;  (4, 'A'), (4, 'B'), (4, 'C'), (4, 'D'), (4, 'E'), (4, 'F'), (4, 'G'), (4, 'H'),
;  (5, 'A'), (5, 'B'), (5, 'C'), (5, 'D'), (5, 'E'), (5, 'F'), (5, 'G'), (5, 'H'),
;  (6, 'A'), (6, 'B'), (6, 'C'), (6, 'D'), (6, 'E'), (6, 'F'), (6, 'G'), (6, 'H'),
;  (7, 'A'), (7, 'B'), (7, 'C'), (7, 'D'), (7, 'E'), (7, 'F'), (7, 'G'), (7, 'H')]

Python has support for various fancy argument and keyword arguments. In Python we might see:

>>> def optional_arg(pos1, pos2, keyword1=None, keyword2=42):
...   return [pos1, pos2, keyword1, keyword2]
...
>>> optional_arg(1, 2)
[1, 2, None, 42]
>>> optional_arg(1, 2, 3, 4)
[1, 2, 3, 4]
>>> optional_arg(keyword1=1, pos2=2, pos1=3, keyword2=4)
[3, 2, 1, 4]

The same thing in Hy:

=> (defn optional-arg [pos1 pos2 &optional keyword1 [keyword2 42]]
...  [pos1 pos2 keyword1 keyword2])
=> (optional-arg 1 2)
[1 2 None 42]
=> (optional-arg 1 2 3 4)
[1 2 3 4]

If you're running a version of Hy past 0.10.1 (eg, git master), there's also a nice new keyword argument syntax:

=> (optional-arg :keyword1 1
...              :pos2 2
...              :pos1 3
...              :keyword2 4)
[3, 2, 1, 4]

Otherwise, you can always use apply. But what's apply?

Are you familiar with passing in *args and **kwargs in Python?:

>>> args = [1 2]
>>> kwargs = {"keyword2": 3
...           "keyword1": 4}
>>> optional_arg(*args, **kwargs)

We can reproduce this with apply:

=> (setv args [1 2])
=> (setv kwargs {"keyword2" 3
...              "keyword1" 4})
=> (apply optional-arg args kwargs)
[1, 2, 4, 3]

There's also a dictionary-style keyword arguments construction that looks like:

(defn another-style [&key {"key1" "val1" "key2" "val2"}]
  [key1 key2])

The difference here is that since it's a dictionary, you can't rely on any specific ordering to the arguments.

Hy also supports *args and **kwargs. In Python:

def some_func(foo, bar, *args, **kwargs):
  import pprint
  pprint.pprint((foo, bar, args, kwargs))

The Hy equivalent:

(defn some-func [foo bar &rest args &kwargs kwargs]
  (import pprint)
  (pprint.pprint (, foo bar args kwargs)))

Finally, of course we need classes! In Python, we might have a class like:

class FooBar(object):
    """
    Yet Another Example Class
    """
    def __init__(self, x):
        self.x = x

    def get_x(self):
        """
        Return our copy of x
        """
        return self.x

And we might use it like:

bar = FooBar(1)
print bar.get_x()

In Hy:

(defclass FooBar [object]
  "Yet Another Example Class"

  (defn --init-- [self x]
    (setv self.x x))

  (defn get-x [self]
    "Return our copy of x"
    self.x))

And we can use it like:

(setv bar (FooBar 1))
(print (bar.get-x))

Or using the leading dot syntax!

(print (.get-x (FooBar 1)))

You can also do class-level attributes. In Python:

class Customer(models.Model):
    name = models.CharField(max_length=255)
    address = models.TextField()
    notes = models.TextField()

In Hy:

(defclass Customer [models.Model]
  [name (models.CharField :max-length 255})
   address (models.TextField)
   notes (models.TextField)])

Macros

One really powerful feature of Hy are macros. They are small functions that are used to generate code (or data). When program written in Hy is started, the macros are executed and their output is placed in the program source. After this, the program starts executing normally. Very simple example:

=> (defmacro hello [person]
...  `(print "Hello there," ~person))
=> (hello "Tuukka")
Hello there, Tuukka

The thing to notice here is that hello macro doesn't output anything on screen. Instead it creates piece of code that is then executed and prints on screen. This macro writes a piece of program that looks like this (provided that we used "Tuukka" as parameter):

(print "Hello there," "Tuukka")

We can also manipulate code with macros:

=> (defmacro rev [code]
...  (setv op (last code) params (list (butlast code)))
...  `(~op ~@params))
=> (rev (1 2 3 +))
6

The code that was generated with this macro just switched around some of the elements, so by the time program started executing, it actually reads:

(+ 1 2 3)

Sometimes it's nice to be able to call a one-parameter macro without parentheses. Sharp macros allow this. The name of a sharp macro must be only one character long, but since Hy operates well with Unicode, we aren't running out of characters that soon:

=> (defsharp ↻ [code]
...  (setv op (last code) params (list (butlast code)))
...  `(~op ~@params))
=> #↻(1 2 3 +)
6

Macros are useful when one wishes to extend Hy or write their own language on top of that. Many features of Hy are macros, like when, cond and ->.

What if you want to use a macro that's defined in a different module? The special form import won't help, because it merely translates to a Python import statement that's executed at run-time, and macros are expanded at compile-time, that is, during the translate from Hy to Python. Instead, use require, which imports the module and makes macros available at compile-time. require uses the same syntax as import.

=> (require tutorial.macros)
=> (tutorial.macros.rev (1 2 3 +))
6

Hy <-> Python interop

Using Hy from Python

You can use Hy modules in Python!

If you save the following in greetings.hy:

(defn greet [name] (print "hello from hy," name))

Then you can use it directly from Python, by importing Hy before importing the module. In Python:

import hy
import greetings

greetings.greet("Foo")

Using Python from Hy

You can also use any Python module in Hy!

If you save the following in greetings.py in Python:

def greet(name):
    print("hello, %s" % (name))

You can use it in Hy (see import):

(import greetings)
(.greet greetings "foo")

More information on Hy <-> Python interop.

Protips!

Hy also features something known as the "threading macro", a really neat feature of Clojure's. The "threading macro" (written as ->) is used to avoid deep nesting of expressions.

The threading macro inserts each expression into the next expression's first argument place.

Let's take the classic:

(loop (print (eval (read))))

Rather than write it like that, we can write it as follows:

(-> (read) (eval) (print) (loop))

Now, using python-sh, we can show how the threading macro (because of python-sh's setup) can be used like a pipe:

=> (import [sh [cat grep wc]])
=> (-> (cat "/usr/share/dict/words") (grep "-E" "^hy") (wc "-l"))
210

Which, of course, expands out to:

(wc (grep (cat "/usr/share/dict/words") "-E" "^hy") "-l")

Much more readable, no? Use the threading macro!