Boost logo

Boost :

From: Edward Diener (eddielee_at_[hidden])
Date: 2002-11-12 14:08:45


"Anthony Williams" <anthony.williamsNOSPAM_at_[hidden]> wrote in
message news:15824.50990.811000.938157_at_gargle.gargle.HOWL...
> Edward Diener writes:
> >I know that I am missing how the mechanism works which links
> > the property_map concept to the get(), put(), and operator[] functions.
>
> A property map is something which relates "keys" to "values", where there
"keys"
> are the "names" of the properties. Given a property map, you can get
and/or
> set the "value" associated with each "key" using the get() and put()
> functions.
>
> You can build a property map out of anything at all, just by defining the
> necessary interface. For example, you could build a property map that
returned
> the values of class members by name, or one that returned the n-th number
in a
> mathematical sequence, given n.
>
> The idea is that in your generic code, you say "give me a property map
which I
> can use with the following keys to return values of type T", and it is
then up
> to the client code how they implement that property map --- _anything_
which
> provides the necessary interface will do.
>
> operator[] is essentially just syntactic sugar for get()/put(), where the
> values are real, discernable objects, for which you can obtain a
> reference. Returning a reference means that you can store the reference
for
> later use, or pass it to another function, and is actually the key benefit
of
> using operator[] rather than plain get()/put().

Thanks. Your explanation makes much better sense to me than anything I have
been able to understand in the property map docs. Here are some simple
questions which, if you can take the time to answer them, might get me over
the hurdle of trying to understand and let me use the generic functions on
my own property maps:

1) Are the get(), put() functions somewhere, like in the boost namespace ?
Is there a doc for the parameters these functions take, or is it the intent
that if one wants to use them, one looks in the source code to figure them
out ?

2) What is the necessary interface to a property map which you mention
above. I see property map categories, property traits, property map types,
but I am missing an understanding from the documentation of the necessary
interface for creating a property map ?

Finally I would like a non-partisan answer from you to my next question and
I am perfectly willing to be wrong and accept the fact that my lack of
understanding of "concept" based explanations are at fault. Do you feel that
the answers to my 2 questions are readily apparent to most intelligent C++
programmers from reading the property map docs ?


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk