Re: [boost] [Boost-announce] [metaparse] Review period starts May 25th and ends June 7th

Subject: Re: [boost] [Boost-announce] [metaparse] Review period starts May 25th and ends June 7th
From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2015-05-31 09:36:59

On 31 May 2015 at 15:03, Abel Sinkovics wrote:

> >> - What is your evaluation of the implementation?
> > * Doesn't follow Boost naming conventions (macro names, namespace,
> > directory layout).
> Could you please be more specific about the naming convention issues? 
> The intention was to make the adoption easy (mostly search and replace: 
> mpllibs -> boost, MPLLIBS -> BOOST) in case of acceptance by using the 
> Boost naming convection with a different "master" library name (mpllibs 
> instead of boost).

I get that it's mostly find and replace, apart from the directory 
layout which will require include paths to be changed.

My problem for me, as a reviewer, is that I want to review an as-if 
finished library, one that could be accepted immediately by the 
community and immediately enter Boost right now, no additional 
changes, because that is the ideal that a library submitter is 
supposed to be aiming for.

That's the whole point of the communtiy review - to review Boost 
libraries. Not libraries which still need to be converted to Boost.

> > * Doesn't appear to actually be a standalone library, but some
> > internal sublibrary.
> It is part of a library with same structure Boost used to have (It looks 
> like this should have been pointed out in the announcement). This must 
> have caused your confusion about the directory structure (parts that 
> look similar).

If it's not part of the review, it shouldn't be in the git repo 
supplied to us.

> I'm happy to fix naming convention-related issues (keeping the 
> Boost/Mpllibs difference) if you point me to them. Moving the code to 
> its own git repository should be straight forward in case of acceptance.

I'm sorry, but I cannot vote in favour of a library not in a final 
form. Yours is a fine library, you've clearly put a ton of work into 
it, and the whole thing smells of polish and attention to detail, 
even maturity.

But I would like to give an immediate acceptance vote for any library 
I review. I don't like recommending conditional acceptance. Yours I 
cannot even recommend conditional acceptance to in its present form 
because its present form bears no resemblence to a Boost library.

Niall

-- 
ned Productions Limited Consulting
http://www.nedproductions.biz/ 
http://ie.linkedin.com/in/nialldouglas/