April 1, 2011, 3:19 p.m.
posted by jackexe
The organization of source code in header files and dot-C files is a practical consequence of various incarnations of the so-called one-definition rule or ODR. An extensive discussion of this rule is presented in Appendix A.
The inclusion versus separation model debate has been a controversial one. The inclusion model is a pragmatic answer dictated largely by existing practice in C++ compiler implementations. However, the first C++ implementation was different: The inclusion of template definitions was implicit, which created a certain illusion of separation (see Chapter 10 for details on this original model).
[StroustrupDnE] contains a good presentation of Stroustrup's vision for template code organization and the associated implementation challenges. It clearly wasn't the inclusion model. Yet, at some point in the standardization process, it seemed as if the inclusion model was the only viable approach after all. After some intense debates, however, those envisioning a more decoupled model garnered sufficient support for what eventually became the separation model. Unlike the inclusion model, this was a theoretical model not based on any existing implementation. It took more than five years to see its first implementation published (May 2002).
It is sometimes tempting to imagine ways of extending the concept of precompiled headers so that more than one header could be loaded for a single compilation. This would in principle allow for a finer grained approach to precompilation. The obstacle here is mainly the preprocessor: Macros in one header file can entirely change the meaning of subsequent header files. However, once a file has been precompiled, macro processing is completed, and it is hardly practical to attempt to patch a precompiled header for the preprocessor effects induced by other headers.
A fairly systematic attempt to improve C++ compiler diagnostics by adding dummy code in high-level templates can be found in Jeremy Siek's Concept Check Library (see [BCCL]). It is part of the Boost library (see [Boost]).