Subject: Re: [boost] Library categories
From: Christophe Henry (christophe.j.henry_at_[hidden])
Date: 2011-11-24 18:16:06
>> Why to classify by subject when the subject is too narrow (like state
>> machines or parsing)?
>If you want I can merge State Machines on Domain Specific if you think the
>subject is too narrow. I have already merged Parsing with Strings and text
>processing. Any alternative suggestion?
Domain Specific seems a bit vague, at least State Machine says what it is.
>> For example, (proto-based) DSEL could be nice. I would see:
>> - proto (of course!)
>> - msm
>> - spirit
>> - phoenix
>> - xpressive
>> - whatever I have forgotten
>> Just a suggestion.
>Umm! Do you mean by this all the libraries that use Expression Templates?
>Would a library that defines its own DSEL with the preprocessor be
>considered a DSEL?
Yes for the first question, I'd say yes too for the second (though I confess
a minute of doubt). These libraries provide a solution to a domain-specific
problem but do it a "different" way or at least it seems so to the user, who
can then decide if he likes this programming style or not and see if he
wants to have a look at those libraries designed on the same principle.
Boost list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk