Re: [boost] Re: Directory Structure

From: Beman Dawes (beman_at_[hidden])
Date: 2000-12-15 13:25:59

At 12:50 PM 12/15/2000 -0500, Jeff Squyres wrote:

 >This makes me thing about something Beman said in a previous post --
 >several of the boost core libraries all depend on each other.  And some 
of
 >the non-core boost libraries depend on one or two of the core libraries.
 >You can easily imagine a dependency graph here.

I'm still struggling with terminology.

Use of "core" will confuse C++ committee members, who use "core" as 
shorthand for "the core language".  They might think "core library" 
referred to clause 18 of the standard; the language support library. 
Perhaps "boost base libraries" would be a better description.

Then for what you called "non-core boost libraries", I've been thinking of 
as "boost domain libraries".  Currently we have two that seem to clearly 
fall into that category: graph and python.

Then base libraries and domain libraries can call base libraries as long as 
advantageous.  But under normal circumstances, base libraries would never 
call domain libraries.  Domain libraries would only call other domain 
libraries when the resulting dependency was clearly documented and had a 
compelling rationale.

 >Should the build/install infrastructure enforce this, and protect the 
user
 >from themselves?  i.e., if you did:
 >
 >	make install-libname1
 >
 >should it fail with an error indicating that libname1 also depends on
 >libname2, libname3, etc.?  Or automagically install libname2, libname3,
 >etc.?
 >
 >Installing piecemeal can be problematic; there should either be some
 >protection for naieve users or it should not be allowed, IMHO.

If installation of one of the boost libraries automatically installed 
dependent libraries, I'm sure users would appreciate it.

--Beman