Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!elroy.jpl.nasa.gov!decwrl!infopiz!lupine!rfg
From: rfg@NCD.COM (Ron Guilmette)
Newsgroups: comp.lang.c++
Subject: Re: Maintaining header files **and** inline management
Message-ID: <4189@lupine.NCD.COM>
Date: 2 Mar 91 19:28:54 GMT
References: <3699@lupine.NCD.COM> <1991Feb5.180503.24515@mathcs.sjsu.edu> <1746@tcs.tcs.com>
Organization: Network Computing Devices, Inc., Mt. View, CA
Lines: 31

In article <1746@tcs.tcs.com> gwu@nujoizey.tcs.com (George Wu) writes:
+In article <1991Feb5.180503.24515@mathcs.sjsu.edu>,
+horstman@mathcs.sjsu.edu (Cay Horstmann) writes:
+|> In fact, the AWK script is very fast (although maybe Ron's method of
+|> running diff on a header portion is faster) and I just rerun it with
+|> each recompile. 
+|> 
+|> It doesn't matter much what method you or your shop uses for building .h
+|> files, as long as it is not the manual "gosh, I guess it is time to look
+|> at that .h file again" method. 
+
+     Won't generating a new .h file with each re-compile of the .C file
+cause *all* other .C files dependent upon the interface to be recompiled?
+Ron preserves the original file, and hence the last modification date make
+uses to determine if a recompile is necessary.  This would save a lot of
+time in complex applications.

George has hit upon the real essence of my "tricky Makefile" scheme.  Using
this scheme allows you to avoid unnecessary recompilations in much the
same way as an Ada "library manager" tool would allow you to avoid
unnecessary recompilations of Ada compilation units.

(Please excuse my use of the `A-word' in this newsgroup.  I know that the
mere mention of it makes certain people violently ill, and I'll try to
avoid using it in future. :-)

-- 

// Ron Guilmette  -  C++ Entomologist
// Internet: rfg@ncd.com      uucp: ...uunet!lupine!rfg
// New motto:  If it ain't broke, try using a bigger hammer.