Re: [sigc] Proposal for standardization in C++ Library TR2

[Douglas Gregor <doug gregor gmail com>]

On Jul 26, 2005, at 6:37 AM, Murray Cumming wrote:
> This sounds great to me. Hopefully you can get the document started, as
> someone who's experienced with this?

Sure. Some mundanities:

1) We need version control. If you two send me your sourceforge IDs, we 
can use the boost-sandbox CVS repository at:

  http://sourceforge.net/projects/boost-sandbox/

2) Any documentation format preferences? LaTeX, HTML, reStructuredText? 
Plain HTML might just be easiest...

> Personally, I don't see anything in the boost::bind API that I dislike,
> though the "disconnecting all signals of a certain group" on that page
> is a little odd.

Everything that concerns "named slot groups" should be removed from the 
Boost.Signals API and should not make it into the proposal. Named slot 
groups have caused a *huge* amount of pain on the implementation side, 
and have caused Boost.Signals to be much less efficient than it should 
be.

> Do you know what platforms boost::bind currently runs on? That might be
> a practical consideration for the API. I've managed to port libsigc++ 
> to
> just about every compiler (latest versions) except HP-UX and the very
> latest SUN Forte (though I expect them to fix those bugs soon).

Amusingly enough, that's one thing that just doesn't matter for 
standardization :) It'll be at least two years from the time that we 
write our proposal until it settles down to more precise standard 
wording that a library vendor would actually implement. Moreover, since 
standard libraries tend to be released with new compilers, any bugs 
that prevent the libraries from working will end up fixed in the 
compilers before we need them.

Still, Boost.Signals is reasonably portable. The only thing that kills 
some common compilers is the

  signal<int(float, double)>

syntax. But, since that's already in several TR1 components, we don't 
have to feel guilty about reusing it.

	Doug