Re: [Boost-users] proto: analytical-only math functions

Subject: Re: [Boost-users] proto: analytical-only math functions
From: Eric Niebler (eric_at_[hidden])
Date: 2009-02-07 12:46:36

Hicham Mouline wrote:
> 
> The grammar of the LHS function definition seems correct.
> The compilation takes too long however (using here intel/msvc libs).
> See the main.cpp and includes below.
> 
> Any ideas how to speed compile time up?

<snip>

There are ways to speed up Proto grammars, but the result is less terse 
and fungible. I wouldn't worry too much about compile times right now. 
 From what I can tell, you're still in the early phases of your DSEL 
design. An expressive, agile DSEL definition will give you freedom to 
experiment.

Once things are working to your satisfaction, you can think about 
replacing some of the complicated composite transforms with custom 
primitive transforms, as described here:

http://boost-sandbox.sf.net/libs/proto/doc/html/boost_proto/users_guide/back_end/expression_transformation/primitives.html

HTH,

-- 
Eric Niebler
BoostPro Computing
http://www.boostpro.com