fext

1.0.7 • Public • Published

fext.js

Fast explicit tail calls in JavaScript [live page] [Node.js package]

Classically, tail calls are automatically optimized (LISP...) . In JavaScript, programmers may well not know which "return" statements are optimized by the engine, and which not. This would lead to practical issues when debugging/optimizing (2009).

As of 2018, progress is "slow" and the Chrome team has already removed the automatic self-recursion optimization from its JavaScript engine.

Another possibility is to explicitly mark the tail calls that will be optimized. Several JavaScript extensions were proposed that add new keywords to the language (2013, 2016).

It turns out that we can do this with today's JavaScript, without extending the language.

fext.js demonstrates this.

API

Two entry points:

  • mfun(...) returns an optimized function,
  • meth(...) returns an optimized method.

Debugging

Just append a "D" character:

  • turn mfun() into mfunD().
  • turn mret() into mretD().

...to turn off the optimizations. You can then use logging and breakpoints.

Getting started

Wrap the whole function with mfun(...) and use return mret(<expr>) to mark the tail calls to be optimized.

Self-recursion example:

<script src="fext.js"></script>
<script>
  var gcd = mfun(
         (a, b) => a > b  ?  mret( mself, a-b, b )
             :     a < b  ?  mret( mself, b-a, a )
             :     a
  );
  console.log( gcd( 2*3*5*17, 3*5*19 ) );  // 15 (3*5)
</script> 

Mutual recursion

var groupkey = {}  // whatever object (won't be modified)
 
,   isOdd = mfun( groupkey
                 , "isOdd"
                 , n => n < 0    ?  mret( mself, -)
                 :      n === 0  ?  false
                 :      mret( isEven, n-1 )
               )
,  isEven = mfun( groupkey
                 , "isEven"
                 , n => n < 0    ?  mret( mself, -)
                 :      n === 0  ?  true
                 :      mret( isOdd, n-1 )
               )
;
console.log( isOdd( 84327681 ) );   // true (no call stack issue)
console.log( isEven( 84327681 ) );  // false (no call stack issue)

groupkey is only used as a key to determine "groups" of function that know each other.

You can conveniently omit groupkey:

// The default `groupkey` is the returned function
// `var isOdd` in this case
var isOdd = mfun( n => n < 0    ?  mret( mself, -)
                  :    n === 0  ?  false
                  :    mret( isEven, n-1 )
                )
,  isEven = mfun( isOdd  // <<< `isOdd` is used here as `groupkey`
                  , "isEven"
                  , n => n < 0    ?  mret( mself, -)
                  :      n === 0  ?  true
                  :      mret( isOdd, n-1 )
                )
;
console.log( isOdd( 84327681 ) );   // true (no call stack issue)
console.log( isEven( 84327681 ) );  // false (no call stack issue)

Namespace alternative

In some situations the globals mfun, meth, mfunD, etc. may feel annoying. Solution: use the fext.* namespace. Below, an example in the Node.js context:

var fx = require( '../fext' ).fext;
            
var isOdd = fx.mfun( function isOdd( n ) {
    return n > 0  ?  fx.mret( isEven, n-1 )
        :  n < 0  ?  fx.mret( fx.mself, -)
        :  false
    ;
})
,  isEven = fx.mfun( isOdd, function isEven( n ) {
    return n > 0  ?  fx.mret( isOdd, n-1 )
        :  n < 0  ?  fx.mret( fx.mself, -)
        :  true
    ;
})
console.log( isOdd( 8951531 ) ); // true ; no call stack issue

Shortcut

Note that fext === fext.mfun as a shortcut for the simplest cases, especially in the Node.js context:

var fx = require( '../fext' ).fext;
 
var self_recursive_fun = fx( ... );

...and in the IE11 context, when relying one the fext namespace, this gives easy access to arrow functions:

var sum_two = fext('(x,y) => x+y');

Methods

Self-recursion example:

var o = {
   gcd : meth( "gcd"
               , (that, a, b) => a > b  ?  mret( that.mself, a-b, b )
               :                 a < b  ?  mret( that.mself, b-a, a )
               :                 a
             )
};
console.log( o.gcd( 2*3*5*17, 3*5*19 ) );  // 15 (3*5)

No need for groupkey here. Moreover, instead of:

mfun( (a,b,c) => ... )

use:

meth( "methodname", (that,a,b,c) => ... )
  • "methodname" MUST be the name of the method,
  • The first parameter MUST be that,
  • Inside the method, you MUST use that (and not this). Reason: .bind() slower in Firefox 60.
  • Use that. in the mret calls, as in:
return mret( that.mself, ...)

or

return mret( that.otherMethod, ... )

The mret( that.<methodName> ) call syntax

  • makes clear we are calling a method, not just a simple function,
  • AND permits easy debugging through methD.

Methods: Mutual recursion example

var o = {
    isOdd : meth( "isOdd"
                  , (that, n) => n < 0  ?  mret( that.mself, -)
                  :            n === 0  ?  false
                  :            mret( that.isEven, n-1 )
                )
    , isEven : meth( "isEven"
                     , (that, n) => n < 0  ?  mret( that.mself, -)
                     :            n === 0  ?  true
                     :            mret( that.isOdd, n-1 )
                   )
};
console.log( o.isOdd( 84327681 ) );  // true (no call stack issue)
console.log( o.isEven( 84327681 ) ); // false (no call stack issue)

Prototype methods

Pretty much the same as above.

function A() {}
A.prototype.isOdd = meth( "isOdd"
                          , (that, n) => n < 0  ?  mret( that.mself, -)
                          :            n === 0  ?  false
                          :            mret( that.isEven, n-1 )
                        );
A.prototype.isEven = meth( "isEven"
                           , (that, n) => n < 0  ?  mret( that.mself, -)
                           :            n === 0  ?  true
                           :            mret( that.isOdd, n-1 )
                         );
var o = new A;
console.log( o.isOdd( 84327681 ) );  // true (no call stack issue)
console.log( o.isEven( 84327681 ) ); // false (no call stack issue)

Speed test

The higher, the better. 100.0 is the highest in each row, (2.6) is a standard deviation, and [9.34e8] is the absolute speed in iterations per second.

Browser isOdd_mfun isOdd_meth isOdd_metaret isOdd_tailtramp
Firefox 60 95.5 (2.6) 100.0 (1.4) 81.2 (0.2) 0.1 (<0.1)
[9.34e+8] [9.78e+8] [7.93e+8] [1.25e+6]
---- ---- ---- ---- ----
Chromium 66 98.5 (0.3) 100.0 (4.9) 52.0 (0.1) 0.7 (<0.1)
[8.87e+8] [9.00e+8] [4.68e+8] [6.12e+6]
---- ---- ---- ---- ----
Chrome 67 99.8 (0.8) 100.0 (0.3) 62.2 (0.2) 0.8 (<0.1)
[7.49e+8] [7.51e+8] [4.67e+8] [6.18e+6]
  • isOdd_mfun and isOdd_meth: Proposed "explicit" approach. Uses standard JavaScript.
  • isOdd_metaret Another "explicit" approach (2013). Extends JavaScript with new keywords.
  • isOdd_tailtramp: The fastest trampoline implementation that I could find.

For more details, explanations, and run this test in your browser, either open the live instance or open the ./index.html file here.

Browser support

Modern browsers, and IE11, as of June 2018.

Unit tests

Either in the browser (live instance) or in the command-line (e.g. Nashorn).

An example: sortedSearch

Code: lib/sorted_search.js

Tests: lib/sorted_search_unittest.js

The code in that example shows an issue when having to pass too many parameters around: the code contains 5 times the same list of parameters, as in the code excerpt:

 
var improveFirst = mfun( improveFirst )
,   improveLast  = mfun( improveFirst, improveLast )
;   
 
function sortedSearch(sortedArray, x, /*?fun?*/less, /*?fun?*/equal)
{
   // ...
   return improveFirst(
       sortedArray, x, less, equal, isFirstFound, isLastFound
       , first_found, last_found, i, j, imax, jmin
   );     
}
 
function improveFirst(
    sortedArray, x, less, equal, isFirstFound, isLastFound
    , first_found, last_found, i, j, imax, jmin
)
{
    // ...
 
    return mret(
        improveLast
        , sortedArray, x, less, equal, isFirstFound, isLastFound
        , first_found, last_found, i, j, imax, jmin
    );
}
 
function improveLast(
    sortedArray, x, less, equal, isFirstFound, isLastFound
    , first_found, last_found, i, j, imax, jmin
)
{
    // ...
 
    return mret(
        improveFirst
        , sortedArray, x, less, equal, isFirstFound, isLastFound
        , first_found, last_found, i, j, imax, jmin
    );
}
 

The straightforward answer to this issue would be to put all parameters in an object, but this leads to a performance degradation (see "Performance cost of passing parameters through an object" further below).

An alternative is to declare the parameters only once, use closure and eval so that the generated code has access to the parameters through the closure.

For a complete example see:

Code: lib/sorted_search_closure.js

Tests: lib/sorted_search_closure_unittest.js

Excerpt:

 
    // First "declare" all functions so that co-dependencies can be
    // solved.
    //
    // Use `mfunD` to debug
    var improveFirst = mfun( improveFirst )
    ,   improveLast  = mfun( improveFirst, improveLast )
    ;
    
    // Now we can implement them, using evil access to the local
    // lexical scope so that we do not need to write parameters
    // repeatedly 5 times (see above `function sortedSearch(...)`).
    improveFirst = eval( improveFirst.getEvalImpl() );
    improveLast  = eval( improveLast.getEvalImpl() );
    
    var first_found
    ,    last_found
    ,             i
    ,             j
    ,          imax
    ,          jmin
    ;
    var sortedArray, x, less, equal;
    
    function sortedSearchClosure
    (in_sortedArray, in_x, /*?fun?*/in_less, /*?fun?*/in_equal)
    /*
      In a sorted array, search for first & last occurences of `x`.
      
      If `x` found, return `[ first_index, last_index ]` (integers).
      
      If `x` not found, return `null`.
    */
    {
        // ...
        
        return improveFirst();
    }
 
    function improveFirst()
    {
        // ...
 
        return mret( improveLast );
    }
 
    function improveLast()
    {
        // ...
        return mret( improveFirst );
    }

Performance cost of passing parameters through an object

To measure the extra overhead of wrapping parameters inside an object, we use the isOdd/isEven case, since each function does very little in itself. First the results, then the detailed implementations.

Browser isOdd_mfun isOdd_mfun_obj isOdd_mfun_obj_inplace
Firefox 60 89.0 (5.4) 24.8 (1.6) 19.3 (0.3)
[8.02e+8] [2.24e+8] [1.74e+8]
---- ---- ---- ----
Chromium 66 100.0 (2.3) 47.9 (0.7) 87.6 (0.8)
[8.35e+8] [4.00e+8] [7.31e+8]
Browser isOdd_meth isOdd_meth_obj isOdd_meth_obj_inplace
Firefox 60 94.0 (2.1) 26.3 (0.5) 18.7 (0.5)
[8.47e+8] [2.37e+8] [1.69e+8]
---- ---- ---- ----
Chromium 66 99.2 (1.7) 45.6 (0.2) 87.9 (0.9)
[8.28e+8] [3.80e+8] [7.34e+8]

To obtain the best performance, it is preferable to pass parameters separately: isOdd_mfun and isOdd_meth. If there are too many parameters, one can use closure, see sortedSearchClosure just above.

*_obj implementations:

For the full details, see the *_obj functions in test/fext_speedtest.js

Excerpt:

function isOdd_mfun_obj( niter )
{
    // The default `namespacekey` is the returned
    // function `var isOdd` in this case.
    var isOdd = mfun( function isOdd( o ) {
        let n = o.n;
        if (> 0)
        {
            o = { n : n-1 };
            return mret( isEven, o );
        }
        else if (< 0)
        {
            o = { n : -};
            return mret( mself, o );
        }
        else
        {
            return false;                
        }
    })
 
    ,  isEven = mfun( isOdd, function isEven( o ) {
        let n = o.n;
        if (> 0)
        {
            o = { n : n-1 };
            return mret( isOdd, o );
        }
        else if (< 0)
        {
            o = { n : -};
            return mret( mself, o );
        }
        else
        {
            return true;
        }
    })
    ;
    // Sanity check
    isOdd_isEven_check( isOdd, isEven, function (n) { return {n:n}; } );
    
    var result = isOdd( { n : niter } ); // <<< speedtest
 
    // Sanity check
    result === (niter % 2 !== 0)  ||  null.bug;
}
 
function isOdd_mfun_obj_inplace( niter )
{
    // The default `namespacekey` is the returned
    // function `var isOdd` in this case.
    var isOdd = mfun( function isOdd( o ) {
        let n = o.n;
        if (> 0)
        {
            o.n--;
            return mret( isEven, o );
        }
        else if (< 0)
        {
            o.n = -n;
            return mret( mself, o );
        }
        else
        {
            return false;                
        }
    })
 
    ,  isEven = mfun( isOdd, function isEven( o ) {
        let n = o.n;
        if (> 0)
        {
            o.n--;
            return mret( isOdd, o );
        }
        else if (< 0)
        {
            o.n = -n;
            return mret( mself, o );
        }
        else
        {
            return true;
        }
    })
    ;
    // Sanity check
    isOdd_isEven_check( isOdd, isEven, function (n) { return {n:n}; } );
    
    var result = isOdd( { n : niter } ); // <<< speedtest
 
    // Sanity check
    result === (niter % 2 !== 0)  ||  null.bug;
}

install

npm i fext

Downloadsweekly downloads

8

version

1.0.7

license

Boost

homepage

github.com

repository

Gitgithub

last publish

collaborators

  • avatar
Report a vulnerability