I like the idea of non-recursive make. Unfortunately, it is not always as easy to create non-recursive makefiles in practice as Miller's paper makes it seem. There are often special situations in various makefiles which are very difficult to handle without using recursion; remember make has no real "scoping" capabilities for variable values, etc. If your project is very simple: just building .c's in .o's into .a's and executables, you can do it without too much hassle. If, as has often been the case in projects I'm involved in, you're building lots of different kinds of source into everything from databases, to compiling internal languages into .c/.h files and building those, etc. it can get more complex.
Non-recursive makes took a very large step closer to reality with the introduction of target-specific variables into GNU make, and I expect that the new $(eval ...) function in GNU make 3.80 will help even more, as will the order-only prerequisite feature. Even so, there will likely be situations where recursion is the only way out, when the rubber really hits the road in a complex environment. -- ------------------------------------------------------------------------------- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org http://make.paulandlesley.org "Please remain calm...I may be mad, but I am a professional." --Mad Scientist _______________________________________________ Help-make mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/help-make
