<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

<html>

<head>

  <meta content="text/html; charset=ISO-8859-1"

 http-equiv="Content-Type">

</head>

<body bgcolor="#ffffff" text="#000000">

I think that covers all the known bugs in std.range and std.algorithm,

with the following exceptions:<br>

<br>

1.&nbsp; Enhancement requests, as opposed to "real" bugs.<br>

<br>

2.&nbsp; Pure documentation issues.<br>

<br>

3.&nbsp; Stuff related to static arrays.<br>

<br>

4.&nbsp; Bug 3067, which is really a compiler bug.<br>

<br>

How do we want to handle the static array case, long term?&nbsp; IMHO stuff

in std.range and std.algorithm should generally<b> not </b>support

static arrays because in addition to being a PITA to

implement,&nbsp;since&nbsp;static&nbsp;arrays&nbsp;are&nbsp;usually&nbsp;stack&nbsp;allocated,

using&nbsp;them&nbsp;as&nbsp;ranges&nbsp;is&nbsp;inherently&nbsp;unsafe.&nbsp;&nbsp;Even&nbsp;in&nbsp;cases&nbsp;where&nbsp;it&nbsp;seems&nbsp;safe,

the&nbsp;predicate&nbsp;may&nbsp;do&nbsp;weird&nbsp;things&nbsp;like&nbsp;escape&nbsp;the&nbsp;static&nbsp;array. In

non-generic code using [] to explicitly make the array into a range

isn't much of a problem.&nbsp; I&nbsp;think&nbsp;the

generic&nbsp;programming&nbsp;issue&nbsp;can&nbsp;be&nbsp;solved&nbsp;by&nbsp;simply&nbsp;having&nbsp;an&nbsp;unsafe&nbsp;function&nbsp;in

std.range.&nbsp; This would make it easy to make static arrays work with

generic programming but require making the conversion somewhat

explicit.&nbsp; It would look something like: <br>

<br>

auto&nbsp;toRange(R)(ref&nbsp;R&nbsp;stuff)&nbsp;if(isInputRange!R)&nbsp;{ <br>

&nbsp;&nbsp;&nbsp;&nbsp;static&nbsp;if(isStaticArray!R)&nbsp;{ <br>

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return&nbsp;stuff[]; <br>

&nbsp;&nbsp;&nbsp;&nbsp;}&nbsp;else&nbsp;{ <br>

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return&nbsp;stuff; <br>

&nbsp;&nbsp;&nbsp;&nbsp;} <br>

} <br>

<br>

Of&nbsp;course,&nbsp;I'd&nbsp;also&nbsp;create&nbsp;an&nbsp;overload&nbsp;w/o&nbsp;the&nbsp;ref&nbsp;so&nbsp;it&nbsp;works&nbsp;with&nbsp;rvalues.&nbsp;

Comments?<br>

<br>

On 8/16/2010 10:56 PM, dsource.org wrote:

<blockquote cite="mid:20100817025658.GA22449@dsource" type="cite">

  <pre wrap="">phobos commit, revision 1884

user: dsimcha

msg:

Bug 4408:  Ambiguity when using std.algorithm.splitter with generic ranges 

<a class="moz-txt-link-freetext" href="http://www.dsource.org/projects/phobos/changeset/1884">http://www.dsource.org/projects/phobos/changeset/1884</a>

_______________________________________________

phobos mailing list

<a class="moz-txt-link-abbreviated" href="mailto:phobos@puremagic.com">phobos@puremagic.com</a>

<a class="moz-txt-link-freetext" href="http://lists.puremagic.com/mailman/listinfo/phobos">http://lists.puremagic.com/mailman/listinfo/phobos</a>

  </pre>

</blockquote>

<br>

</body>

</html>