Trains

From NARS2000
Revision as of 19:32, 15 April 2018 by Sudleyplace (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.
Z←  (f g) R is called a Monadic HookZ ≡ R f g R
Z←L (f g) R is called a Dyadic HookZ ≡ L f g R
Z←  (f g h) R is called a Monadic ForkZ ≡ (f R) g h R
Z←L (f g h) R is called a Dyadic ForkZ ≡ (L f R) g L h R
Z←  (f g h ...) R
Z←L (f g h ...) R
is also defined for longer Trains
L and R are arbitrary arrays, f, g, h, etc, are arbitrary functions of any type: primitive, user-defined, system, anonymous, and/or derived.


This clever idea from the designers of J is called Trains where a parenthesized sequence of functions (which normally would signal a SYNTAX ERROR) can be interpreted as per the above descriptions. Very nicely, they fit into and extend the spectrum of function definition syntax from user-defined to dynamic to trains to operator expressions. They are another and very interesting form of functional programming.

Note that the spacing between functions is for visual purposes only — it has no effect on the interpretation.

For example,

(,⍎)'2+3'
←→ '2+3',⍎'2+3'
←→ '2+3',5
2+3 5

avg←(+⌿ ÷ ≢) defines a function that computes the average of a numeric vector.
avg 1 2 3 4
←→ (+⌿ ÷ ≢) 1 2 3 4
←→ (+⌿1 2 3 4) ÷ ≢1 2 3 4
←→ 10 ÷ 4
2.5

Longer Trains are defined as follows:

(e f g h) ←→ (e (f g h))
(d e f g h) ←→ (d e (f g h))

and in general

Even length: (a b c ...) ←→ (a (b c ...))
Odd length: (a b c ...) ←→ (a b (c ...))

For more applications of this concept, see the discussion in the Learning J manual.

There is also a series of tables of common function expressions and their corresponding Train.