Am Wednesday, 17. August 2011, 13:42:50 schrieb Carl Sorensen: > On 8/17/11 5:35 AM, "David Kastrup" <d...@gnu.org> wrote: > > So it would seem like another worthwhile frog task to get rid of > > SCM_CDRLOC in the source tree when feasible (when the list is being > > consulted front-to-back while it is in the process of being created, > > this would not be possible. However, since tail pointers in this manner > > can't be kept around beyond the function call, chances are that we have > > very few such cases if at all). > > > > It will simplify the code, make it more readable and faster, and > > decrease the amount of stack scanning that can go wrong. > > And there are only 29 occurrences, so it's a smaller task the the task of > rationalizing the guile interface.
It's smaller, but it requires a deeper analysis of the code (where the list might be used meanwhile during the loop, whether the list might already contain elements,e tc.). Most of the guile interface issues are simply copy-and-paste of very few templates,e.g. -) "something == SCM_EOL" => "scm_is_null (something)" -) "something == SCM_BOOL_T" => "scm_is_true (something)" -) "scm_equal_p (a, b_) == SCM_BOOL_F" => "!ly_is_equal (a, b)" Cheers, Reinhold -- ------------------------------------------------------------------ Reinhold Kainhofer, reinh...@kainhofer.com, http://reinhold.kainhofer.com/ * Financial & Actuarial Math., Vienna Univ. of Technology, Austria * http://www.fam.tuwien.ac.at/, DVR: 0005886 * LilyPond, Music typesetting, http://www.lilypond.org _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
