LISP_FLOAT_TYPE

Can I get rid of LISP_FLOAT_TYPE? Here are the arguments for doing so: 1) FSF Emacs got rid of it a few years ago (the last traces disappeared in February 2000, according to the Changelog). 2) It is unconditionally on anyway, so removing it will have no effect. 3) I'm working on enhanced number support (bignums, ratios, etc.). a) Some parts have to be done differently depending on whether LISP_FLOAT_TYPE is defined or not. Since it is always on, the choice is between writing code that will never be used for the "off" condition, or making the "off" code bitrot deliberately, in which case we might as well get rid of it. b) LISP_FLOAT_TYPE presents a view of the floating point support that is a gross oversimplification of what the state will be after my work bears fruit. The only argument I can come up with for not nuking LISP_FLOAT_TYPE is that somebody might someday port XEmacs to a platform with no floating point support whatever. That seems to be a pretty weak argument when stacked up against the ones above. -- Jerry James http://www.ittc.ku.edu/~james/