Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!feeder.erje.net!eu.feeder.erje.net!weretis.net!feeder1.news.weretis.net!news.solani.org!.POSTED!not-for-mail
From: Thomas 'PointedEars' Lahn <PointedEars@web.de>
Newsgroups: comp.lang.javascript
Subject: Re: function to get frequencies of a char array
Date: Tue, 27 Jan 2015 03:03:05 +0100
Organization: PointedEars Software (PES)
Lines: 68
Message-ID: <1859346.AqY2EoPZMB@PointedEars.de>
References: <dec04bcb-c8c2-4446-b82c-6c3fa7e05e48@googlegroups.com> <XnsA42B6242E020eejj99@194.109.133.133> <m9ti7r$sbg$1@solani.org> <26839541.MRbOrngyXZ@PointedEars.de> <87twzgl0aq.fsf@bsb.me.uk> <2697158.aGRKFDBAEn@PointedEars.de> <87lhkskr0r.fsf@bsb.me.uk> <4289032.9WYFBcEGMJ@PointedEars.de> <87twzdf1gu.fsf@bsb.me.uk>
Reply-To: Thomas 'PointedEars' Lahn <cljs@PointedEars.de>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8Bit
X-Trace: solani.org 1422324224 29758 eJwNwoERgEAIA7CVUNoK48AL+4+gl9B16TwQBe4PztOL4PSGK8YmmKrJWpuMGy96nWZR8P4AHlsRFA== (27 Jan 2015 02:03:44 GMT)
X-Complaints-To: abuse@news.solani.org
NNTP-Posting-Date: Tue, 27 Jan 2015 02:03:44 +0000 (UTC)
User-Agent: KNode/4.14.2
X-User-ID: eJwVyMERACEIBLCWRNgVywGB/ku4uTwDpfAdI2gYjJ7qLr8REiOdC34al657e9eWHD6DSC23mJT/PLCYpUH7AGURFbg=
Cancel-Lock: sha1:wZvWkFzATLcbGZdKCsB/b40RMvo=
X-NNTP-Posting-Host: eJwFwYERADAEBLCVKP4Zp1ztP0KTMCiGjoDHxs5t5tP7tDLMzYd7CN2Sdno3xEjp5NjB1gcXORB+
Xref: csiph.com comp.lang.javascript:26726

Ben Bacarisse wrote:

> Thomas 'PointedEars' Lahn <PointedEars@web.de> writes:
>> Ben Bacarisse wrote:
>>> Thomas 'PointedEars' Lahn <PointedEars@web.de> writes:
>>>> Ben Bacarisse wrote:
>>>>> Thomas 'PointedEars' Lahn <PointedEars@web.de> writes:
>>>>>>   var counts = {};
>>>>>>
>>>>>>   for (var i = 0, len = a.length; i < len; ++i)
>>>>>>   {
>>>>>>     /* NOTE: hasOwnProperty() may be required */
>>>>>>     var e = a[i];
>>>>>>     e ? ++counts[e] : (counts[e] = 1);
>>>>>>   }
>>>>> 
>>>>> I think you meant counts[e] ? ++counts[e] : (counts[e] = 1); here (and
>>>>> immediately below): […]
>>>>
>>>> No, but
>>>>
>>>>   if (e) { counts[e] ? ++counts[e] : (counts[e] = 1); }
>>> 
>>> Why?
>>
>> Because there are sparse arrays ([To]String(“undefined”) ===
>> "undefined").
> 
> Isn't
> 
>   if (a.hasOwnProperty(i)) { counts[e] ? ++counts[e] : (counts[e] = 1); }
> 
> a better way to deal with sparse arrays in such a loop?

In general, it is a better way.  However, since Array indexes are enumerable 
properties, you can also iterate over a filtered, sorted Object.keys(a), say

  var indexes = Object.keys(a).filter(function (e) {
    return +e >= 0 && +e < Math.pow(2, 32);
  }).sort(function (a, b) {
    return a - b;
  });

  for (var ii = 0, len = indexes.length; ii < len; ++ii)
  {
    var e = a[indexes[ii]];
    // …
  }

Which approach is more efficient depends on the sparseness of the array; the 
latter solves the problem that iteration order is unspecified with “for–in” 
and includes inherited properties.

That means you could do

  var reduced = Object.keys(a).filter(…).sort(…).map(…).reduce(…);

> Testing e seems rather heavy-handed, though it's not clear what the OP 
> might want in anything but the simple case given.  In a general counting 
> loop one would probably want to count zeros and NaNs and so on.

Yes, as so often, this is limited to the use-case.

-- 
PointedEars
FAQ: <http://PointedEars.de/faq> | SVN: <http://PointedEars.de/wsvn/>
Twitter: @PointedEars2 | ES Matrix: <http://PointedEars.de/es-matrix>
Please do not cc me. / Bitte keine Kopien per E-Mail.