Hi, Like the poster of
http://lists.gnu.org/archive/html/bug-make/2000-10/msg00011.html I've noticed that gnu make (I'm currently using 3.80) has a little undocumented behavior regarding the use of 'override'. Without 'override', an assignment in a Makefile to a make variable will not take effect if the make variable has been specified on the Make command line. Using 'override' allows assignments in a Makefile to modify variables that have been specified on the command line. That's basically what the documentation says. However, using 'override' also seems to prevent subsequent assignments to the variable, done within the Makefile without 'override', from taking effect, as if using 'override' made the variable look as if it had been defined on the command line. This does not appear to be documented. Is the behavior that 'override' seems to endow a variable with 'command line nature', even if it was not given on the command line, so that subsequent assignments without 'override' have no effect, intended or not? If it is intended, it should probably be documented. If it is not intended, and is viewed as a bug, and backwards compatibility issues permit, it should probably be fixed. Otherwise it should be documented. - Dan 121> make --version GNU Make 3.80 Copyright (C) 2002 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. 122> cat mk1 override FLEA = FLY FLEA += FLUE it : echo $(FLEA) 123> make -f mk1 echo FLY FLY 124> cat mk2 FLEA = FLUE override FLEA += FLY it : echo $(FLEA) 125> make -f mk2 echo FLUE FLY FLUE FLY 126> _______________________________________________ Bug-make mailing list Bug-make@gnu.org http://lists.gnu.org/mailman/listinfo/bug-make
