Serious bugfix in the op macro, and derived macro, in

the handling of missing meta-numbers (for
instance (op ... @1 .. @3): 1 and 3 occur, but not 2.

* eval.c (supplement_op_syms): Bugfix: loop was testing
and updating the wrong variable, causing it to be looking at the first
element of the syms list in every iteration.
Not a regression: this was implemented wrong when the function
was introduced on October 6, 2013.

2 files changed, 13 insertions, 3 deletions

diff --git a/ChangeLog b/ChangeLog
index 474d65eb..4243ab4c 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,17 @@
 2015-02-06  Kaz Kylheku  <kaz@kylheku.com>
 
+	Serious bugfix in the op macro, and derived macro, in
+	the handling of missing meta-numbers (for
+	instance (op ... @1 .. @3): 1 and 3 occur, but not 2.
+
+	* eval.c (supplement_op_syms): Bugfix: loop was testing
+	and updating the wrong variable, causing it to be looking at the first
+	element of the syms list in every iteration.
+	Not a regression: this was implemented wrong when the function
+	was introduced on October 6, 2013.
+
+2015-02-06  Kaz Kylheku  <kaz@kylheku.com>
+
 	* eval.c (subst_vars): Bugfix. When an ordinary variable
 	occurs in the quasi syntax (which is the result of substitution
 	of metanums like @1 by the op macro), if that variable
diff --git a/eval.c b/eval.c
index 87616a08..7e27c962 100644
--- a/eval.c
+++ b/eval.c
@@ -2542,9 +2542,7 @@ static val supplement_op_syms(val ssyms, val max)
   list_collect_decl (outsyms, tl);
   val si, ni;
 
-  for (si = ssyms, ni = one;
-       ssyms;
-       ni = plus(ni, one), ssyms = cdr(ssyms))
+  for (si = ssyms, ni = one; si; ni = plus(ni, one), si = cdr(si))
   {
     val entry = car(si);
     val num = car(entry);