This patches the info pages:
1) FS="\n" is not regexp, just one-character string;
2) the documentation of split() should also mention that split("",array)
   clears the array.

--- gawk-3.1.1/doc/gawk.texi.orig	Mon Apr 22 13:26:20 2002
+++ gawk-3.1.1/doc/gawk.texi	Mon May 20 12:07:18 2002
@@ -5042,8 +5042,7 @@
 
 Another way to separate fields is to
 put each field on a separate line: to do this, just set the
-variable @code{FS} to the string @code{"\n"}.  (This simple regular
-expression matches a single newline.)
+variable @code{FS} to the string @code{"\n"}.
 A practical example of a @value{DF} organized this way might be a mailing
 list, where each entry is separated by blank lines.  Consider a mailing
 list in a file named @file{addresses}, which looks like this:
@@ -12246,8 +12245,13 @@
 
 Before splitting the string, @code{split} deletes any previously existing
 elements in the array @var{array}.
-If @var{string} does not match @var{fieldsep} at all, @var{array} has
-one element only. The value of that element is the original @var{string}.
+
+If @var{string} is null, the array has no element. (So this is a portable
+way to delete an array.  @xref{Delete, ,The @code{delete} Statement}.)
+
+If @var{string} does not match @var{fieldsep} at all (but is not null),
address@hidden has one element only. The value of that element is the original
address@hidden
 
 @item sprintf(@var{format}, @var{expression1}, @dots{})
 @cindex @code{sprintf} function