Re: [boost] GGL review (JTF)

Subject: Re: [boost] GGL review (JTF)
From: Jonathan Franklin (franklin.jonathan_at_[hidden])
Date: 2009-11-28 16:32:51

On Thu, Nov 26, 2009 at 5:21 AM, Barend Gehrels <barend_at_[hidden]> wrote:
> The reason for the assign with 4 functions
> is to assign the sphere completely, in one call.

Right.

> I do not understand exactly
> what you dislike, but probably it was something that seems to be not used or
> moved?

Simply that the 2- and 3-type overloads completely assign 2- and 3-d
geometries respectively.  I would prefer the 4-type version to
completely assign a 4-d geometry.  Perhaps the interface already
supports this (not the implementation), since the function is
overloaded on the geometry type.

>> The example code on the main page didn't compile, which isn't a
>> problem in and of itself.
>
> This is surprising, as all examples are extracted from real, compiling
> code... Be it, indeed, that headerfiles are not extracted (more reviewers
> have mentioned this and this will be addressed)

There was nothing wrong w/ the code itself.  The issue was that it
required headers, other than ggl.hpp, to compile.  It took a bit of
leg-work to figure out what I needed to include.

> We understand the flaws and drawbacks of the documentation, but for your
> information, the "modules" page lists all of the algorithms available....

I went back later, and was able to find the algorithm in question from
this page.  Thanks!

>> However, it is a shame that the library lacks a first-class spatial
>> index capability in it's present state (e.g. what is up for review). ...
>
> It is available as an extension, the reasons for non-inclusion were:
> - we want to improve performance
> - we want to enhance the interface

So will it be moved *into* the library-proper at some point, or will
it always remain an extension?

Thanks,

Jon