%% Krishnakumar B <[EMAIL PROTECTED]> writes:
kb> Specifically is there any way in GNU Make to specify that if a
kb> pattern substitute fails to replace anything, return null.
Not using variable substitution references.
kb> If I understand what Make does now, it tries to match the longest
kb> entry for % and ends up matching the whole name but still doesn't
kb> find .cc and hence doesn't perform the replace. But the problem
kb> arises because Make returns the whole % at this point even if no
kb> match requested was found.
Certainly. The semantics of pattern variable substitutions are that
they substitute for words that the pattern matches, and don't touch
words that the pattern does not match.
kb> If it's not possible to change the semantics (legacy reasons et
kb> al, though I don't know if anybody will really want this
kb> behaviour), will it be possible to issue a warning option if a
kb> variable pattern substitution results in a value which is the same
kb> as the original variable ? Even something that detects such a
kb> thing in only this specified syntactic usage is also very welcome.
No, I'm sorry, but that is not possible. We cannot issue warnings for
behavior which is well-defined, useful, behaves exactly as described in
the manual, and is used that way on purpose. Consider something like:
foo: $(OBJS) $(LIBS)
$(CC) ... $(^:lib%.a=-l%)
Here LIBS could be empty, or it could have all the libraries already in
the form -lbar. I'm not exactly sure what you mean by "only on this
specified syntactic usage", but I'm not thrilled about adding special
cases to the code which say things like "if we're patsubst'ing for .cc
or .cpp or .c (or whatever) then give a warning".
To get exactly what you're asking for, try this:
OBJS = $(patsubst %.cc,%.o,$(filter %.cc,$(SRC)))
But, you probably want this instead:
OBJS = $(addsuffix .o,$(basename $(SRC)))
HTH.
--
-------------------------------------------------------------------------------
Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at:
http://www.gnu.org http://www.paulandlesley.org/gmake/
"Please remain calm...I may be mad, but I am a professional." --Mad Scientist
_______________________________________________
Bug-make mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-make
