Clojure

Differences with other Lisps

This information is provided for programmers familiar with Common Lisp or Scheme.

  • Clojure is case sensitive

  • Clojure is a Lisp-1

  • () is not the same as nil

  • The reader is side-effect free

  • Keywords are not Symbols

  • Symbols are not storage locations (see Var)

  • nil is not a Symbol

  • t is not syntax, use true

  • The read table is not accessible to user programs

  • let binds sequentially

  • do is not a looping construct

  • There is no tail-call optimization, use recur.

  • syntax-quote does symbol resolution, so `x is not the same as 'x.

  • ` has auto-gensyms.

  • ~ is unquote , is whitespace

  • There is reader syntax for maps, vectors, and sets

  • cons, first and rest manipulate sequence abstractions, not concrete cons cells

  • Most data structures are immutable

  • lambda is fn, and supports overloading by arity

  • = is the equality predicate

  • Global Vars can be dynamically rebound (if declared dynamic) without interfering with lexical local bindings. No special declarations are necessary to distinguish between dynamic and lexical bindings. Since Clojure is a Lisp-1, (global) functions can be dynamically rebound (if they are marked as dynamic).

  • No letrec, labels or flet - use (fn name [args]…​) for self-reference, letfn for mutual reference.

  • In Clojure nil means 'nothing'. It signifies the absence of a value, of any type, and is not specific to lists or sequences.

  • Empty collections are distinct from nil. Clojure does not equate nil and '().

  • false means one of the two possible boolean values, the other being true

  • There is more to collections than lists. You can have instances of empty collections, some of which have literal support ([], {}, and ()). Thus there can be no sentinel empty collection value.

  • Coming from Scheme, nil may map most closely to your notion of #f.

  • A big difference in Clojure, is sequences. Sequences are not specific collections, esp. they are not necessarily concrete lists. When you ask an empty collection for a sequence of its elements (by calling seq) it returns nil, saying "I can’t produce one". When you ask a sequence on its last element for the rest it returns another logical sequence. You can only tell if that sequence is empty by calling seq on it in turn. This enables sequences and the sequence protocol to be lazy.

  • Some of the sequence functions correspond to functions from Scheme and CL that there manipulated only pairs/conses ('lists') and returned sentinel values ('() and nil) that represented 'empty' lists. The Clojure return values differ in not returning specific empty collections, but rather another logical sequence. Some of the sequence functions have no counterpart in Scheme/CL, and map to Haskell/ML-like functions. Some of those functions return infinite or calculated sequences, where the analogy to concrete data-structures like Scheme/CL lists is tenuous at best.

  • It helps to distinguish collections/data-structures and seqs/iteration. In both CL and Scheme they are conflated, in Clojure they are separate.

Clojure Common Lisp Scheme Java

Has nil?

nil - means 'nothing'

nil - means false or empty list

-

null

Has true?

true

-

#t

true (primitive)

Has false?

false

-

#f

false (primitive)

Conditionals distinguish:

nil or false/ everything else

nil/non-nil

#f/non-#f

false/true

List/sequence library manipulates distinguished concrete type(s)?

No - seq abstraction with many collection implementations

Yes - cons and vector

Yes - pair

No - Iterator abstraction with many collection implementations

Singleton empty-list value?

No - can have distinct empty values of concrete collection types

nil

'()

No

End-of-sequence returns:

a logical sequence for which seq returns nil

nil

'()

false

Host null:

nil

NA

NA

NA

Host true:

true (boxed)

NA

NA

NA

Host false:

false (boxed)

NA

NA

NA