On Tue, 07 Jan 2003, larsi@gnus.org wrote: > Ted Zlatanov writes: > >>> While compiling the end of the data in file >>> /csa/forum/emerick/gnus/lisp/spam.el: ** The following functions >>> are not known to be defined: spam-BBDB-register-routine, >>> bbdb-create-internal, spam-enter-ham-BBDB >> >> I think this is because of the mess with the BBDB macros defined in >> bbdb-com.el. It shouldn't be a problem because of the >> condition-case, I thought. Can anyone suggest a better way to deal >> with those macros, or point me to the Emacs Lisp features I'm not >> aware of to hush up the warnings? > > Just put the functions/macros in the maybe-bind lists in lpath.el to > make the warning messages go away. Thanks for the advice. I used lpath.el and the warning went away. Now, whenever I start Gnus and try to use the BBDB ham processor spam-enter-ham-BBDB, I get this error: Adding address onedollar@brown.bidz.com to BBDB Followed link to /home/tzz/autodist/dotfiles/bbdb Parsing BBDB... (frobnicating...done) spam-enter-ham-BBDB: Invalid function: (macro . #[(records &optional name company net notes phone) " ; 9! ; 9! ;, 9,! ;: 9:! ;H 9H! n BBBBBE BBEFBBBB D BB DBB DBBEBBBBBE DBBEFBBB D DEBBEFFB BB BBBBBEBBBD\"  BBBBBE BBEFBBBB  BBB @@DABB\"Bс@B)BCBBDBBBBEBBB" [clauses name company net notes phone nil error "name must be atomic" "company must be atomic" ...] 20]) The curious thing is, when I do (eval-buffer) in spam.el, I don't get the errors anymore. So there is some sort of discrepancy, it seems, between spam.el and spam.elc as far as the BBDB goes. It may have been there before, I always use eval-buffer while testing. I think I'm doing things correctly, though. All the BBDB functionality is loaded inside a condition-case, and the functions plus the bbdb-search macro are aliased to 'ignore by lpath.el. Ted