[LRUG] Comparable Null Objects

[Duncan Stuart]

Awesome - thank you (and everyone) for taking the time to explain.

On 16 July 2015 at 13:40, Tom Stuart <tom at codon.com> wrote:

> On 16 Jul 2015, at 13:03, Duncan Stuart <dgmstuart at gmail.com> wrote:
> > Because ordering on Dates in Ruby requires actual Dates and not
> Date-like things, forcing us into an undesirable implementation?
>
> Yes, I think that’s the underlying problem.
>
> However, many of the techniques for “eliminating” nil checks or
> instance-of checks don’t really eliminate them, only encapsulate them. The
> condition that drives the decision about which class to instantiate (or
> whatever) has to go somewhere. So, for me, the principle is about doing
> those checks sooner, so that the rest of the system can lean on
> polymorphism — which is just a different kind of conditional that gets
> hidden from the caller — instead of having to be directly aware of the
> condition.
>
> For example, you could “eliminate” the checks here by having two different
> classes of date-ranking object for events with & without dates, and
> designing their #<=> methods so that the two kinds of object interoperate
> nicely during a heterogeneous sort, etc. But you’d still need a condition
> somewhere to decide which of those classes to use for ranking a given event.
>
> So it is a trade-off, and the guiding principle is how tightly you want to
> contain the information that an event may be missing its date.
> _______________________________________________
> Chat mailing list
> Chat at lists.lrug.org
> Archives: http://lists.lrug.org/pipermail/chat-lrug.org
> Manage your subscription: http://lists.lrug.org/options.cgi/chat-lrug.org
> List info: http://lists.lrug.org/listinfo.cgi/chat-lrug.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lrug.org/pipermail/chat-lrug.org/attachments/20150716/04912339/attachment-0003.html>