-/* Flags for node->flags of ANYOF */
-
-#define ANYOF_CLASS 0x08 /* has [[:blah:]] classes */
-#define ANYOF_INVERT 0x04
-#define ANYOF_FOLD 0x02
-#define ANYOF_LOCALE 0x01
-
-/* Used for regstclass only */
-#define ANYOF_EOS 0x10 /* Can match an empty string too */
-
-/* There is a character or a range past 0xff */
-#define ANYOF_UNICODE 0x20
-#define ANYOF_UNICODE_ALL 0x40 /* Can match any char past 0xff */
-
-/* size of node is large (includes class pointer) */
-#define ANYOF_LARGE 0x80
-
-/* Are there any runtime flags on in this node? */
-#define ANYOF_RUNTIME(s) (ANYOF_FLAGS(s) & 0x0f)
+/* If the bitmap doesn't fully represent what this ANYOF node can match, the
+ * ARG is set to this special value (since 0, 1, ... are legal, but will never
+ * reach this high). */
+#define ANYOF_NONBITMAP_EMPTY ((U32) -1)
+
+/* The information used to be stored as as combination of the ANYOF_UTF8 and
+ * ANYOF_NONBITMAP_NON_UTF8 bits in the flags field, but was moved out of there
+ * to free up a bit for other uses. This tries to hide the change from
+ * existing code as much as possible. Now, the data structure that goes in ARG
+ * is not allocated unless it is needed, and that is what is used to determine
+ * if there is something outside the bitmap. The code now assumes that if
+ * that structure exists, that any UTF-8 encoded string should be tried against
+ * it, but a non-UTF8-encoded string will be tried only if the
+ * ANYOF_NONBITMAP_NON_UTF8 bit is also set. */
+#define ANYOF_NONBITMAP(node) (ARG(node) != ANYOF_NONBITMAP_EMPTY)
+
+/* Flags for node->flags of ANYOF. These are in short supply, so some games
+ * are done to share them, as described below. If necessary, the ANYOF_LOCALE
+ * and ANYOF_CLASS bits could be shared with a space penalty for locale nodes
+ * (and the code at the time this comment was written, is written so that all
+ * that is necessary to make the change would be to redefine the ANYOF_CLASS
+ * define). Once the planned change to compile all the above-latin1 code points
+ * is done, then the UNICODE_ALL bit can be freed up. If flags need to be
+ * added that are applicable to the synthetic start class only, with some work,
+ * they could be put in the next-node field, or in an unused bit of the
+ * classflags field. */
+
+#define ANYOF_LOCALE 0x01
+
+/* The fold is calculated and stored in the bitmap where possible at compile
+ * time. However there are two cases where it isn't possible. These share
+ * this bit: 1) under locale, where the actual folding varies depending on
+ * what the locale is at the time of execution; and 2) where the folding is
+ * specified in a swash, not the bitmap, such as characters which aren't
+ * specified in the bitmap, or properties that aren't looked at at compile time
+ */
+#define ANYOF_LOC_NONBITMAP_FOLD 0x02
+
+#define ANYOF_INVERT 0x04
+
+/* EOS, meaning that it can match an empty string too, is used for the
+ * synthetic start class (ssc) only. It looks like it could share the INVERT
+ * bit, as the ssc is never inverted. But doing that caused this reges to
+ * not match:
+ * 'foo/file.fob' =~ m,^(?=[^\.])[^/]* /(?=[^\.])[^/]*\.fo[^/]$,;
+ * (except the space between the * and the / above shouldn't be there; it was
+ * inserted to make this comment continue on.)
+ * Rather than try to figure out what was going on in the optimizer, I (khw)
+ * found a way to save a different bit. But my original line of reasoning was
+ * "The bit just needs to be turned off before regexec.c gets a hold of it so
+ * that regexec.c doesn't think it's inverted, but this happens automatically,
+ * as if the ssc can match an EOS, the ssc is discarded, and never passed to
+ * regexec.c" */
+#define ANYOF_EOS 0x10
+
+/* CLASS is never set unless LOCALE is too: has runtime \d, \w, [:posix:], ...
+ * The non-locale ones are resolved at compile-time */
+#define ANYOF_CLASS 0x08
+#define ANYOF_LARGE ANYOF_CLASS /* Same; name retained for back compat */
+
+/* Can match something outside the bitmap that isn't in utf8 */
+#define ANYOF_NONBITMAP_NON_UTF8 0x20
+
+/* Matches every code point 0x100 and above*/
+#define ANYOF_UNICODE_ALL 0x40
+
+/* Match all Latin1 characters that aren't ASCII when the target string is not
+ * in utf8. */
+#define ANYOF_NON_UTF8_LATIN1_ALL 0x80