Re: [boost] [Boost.Parameter] advice needed for Boost.svg_plot

From: Paul A Bristow (pbristow_at_[hidden])
Date: 2007-06-27 04:20:09

>-----Original Message-----
>From: boost-bounces_at_[hidden] 
>[mailto:boost-bounces_at_[hidden]] On Behalf Of Jake Voytko
>Sent: 27 June 2007 03:08
>To: boost_at_[hidden]
>Subject: [boost] [Boost.Parameter] advice needed for Boost.svg_plot

>As you can see below, I've gone a long way on letting users 
>customize plots.

>Mathias Gaunard suggested I look at "Boost.Parameter", and I 
>think that it
>could provide a lot of benefits to my library. The question I 
>have is how much should I let it influence my interface?

>There are functions that could clearly benefit from named parameters.

My opinion, FWIW, is that the layout you have is nice and intuitive.

I specially like the way the chaining allows you to group various features together in a way that makes it easy to read.  

Although named parameters might be nice - and would have been really nice if they had been included in the original ++ when C was
incremented, since the Standards people have not been persuaded to add names parameters to the language (because there are other
more important faults, and there is trouble with mixing with positional parameters and defaults), named parameters remain unfamiliar
to most users.

So I would not get diverted to this unless the benefits are *really* compelling, and I can't see that they are.  Your time is
limited and despite your excellent progress, there is much to do :-)

(There is also the potential downside of longer compile times and perhaps complications to get working with all compilers).

Paul

---
Paul A Bristow
Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB
[phone] & SMS, Mobile [phone] & SMS
pbristow_at_[hidden]