Re: [boost] GSOC Proposal for BIGINT - Long Mail

Subject: Re: [boost] GSOC Proposal for BIGINT - Long Mail
From: Scott McMurray (me22.ca+boost_at_[hidden])
Date: 2010-03-24 19:50:17

On 24 March 2010 19:20, swagat konchada <swagatata_at_[hidden]> wrote:
>
> 1. Is there sufficient interest for a BigDecimal proposal on the lines of
> the Java implementation of BigDecimal (or has there been some prior work
> that can be used as a foundation?)
>

I think proposing a BigDecimal is premature until a BigInteger has
made it into Boost.  Keep the scope small for now; we don't know
exactly what's needed just yet.

> 2. What are the deficiencies of bigint that have prevented its inclusion
> into the Boost distribution so as to address the same as part of my GSoC
> proposal.
>

Licensing is one issue; IIRC there is a significant portion for whom
any license other than a Boost-like one is unacceptable, thus making a
GMP-based, and thus LGPLed, BigInt insufficient.

On the other hand, optimizing all the algorithms involved is a huge
task, and probably not something the
single-implementor/sponsor/maintainer model generally used by Boost is
good at.  As such, a Boost-only library would probably not have the
performance demanded by a substantial portion of its users.

So as I've said before, I think a Boost project should be mostly an
expression-template system that provides a nice interface on top of an
externally-provided set of numeric operations.