Boost logo

Boost :

Subject: Re: [boost] [modularization] Modularizing Boost (modularization)
From: John Maddock (john_at_[hidden])
Date: 2013-11-01 05:31:12


> If we again, for the sake of simplifying the core, treat
> (type_traits|mpl|detail|utility|smart_ptr|typeof) as a single node, the
> remainder of the strong graph is this:
>
>
> http://steveire.com/boost/nov1_strong_with_property_map_parallel_and_corelib.dot
>
> http://steveire.com/boost/nov1_strong_with_property_map_parallel_and_corelib.png
>
>
> There are obviously other worthwhile changes which could reduce this
> further, some of which I recommended in my last report.

Looking at my stuff:

Math->multiprecision
Math->regex

are both false dependencies: they come from a single header that is used for
code maintenance (generating new numeric constants). The header is however
documented and usable by end users to generate their own constants so it
logically belongs under include-path/boost/math even though it will only
ever appear in maintenance code, not production code.

Math->ConceptCheck

Again a false positive or "glue" header. The dependency is only there if
you really want to use those two libraries together (and for testing of
course).

As mentioned in a previous email, I don't believe cases like these should be
considered true dependencies, and there should be some way to prevent
whatever tools we're using to determine and check dependencies to ignore
cases like these.

John.


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk