Verify character class still non-empty after converting to byte class

For `[^\x00-\xff]`, while it is still treated as a full Unicode
character class, it is not empty. For instance `≥` would still be
matched.

However, when `CharClass::to_byte_class` is called on it (as is done
when using `regex::bytes::Regex::new` rather than `regex::Regex::new`),
it _is_ now empty, since it excludes all possible bytes.

This commit adds a test asserting that `regex::bytes::Regex::new` for
this case (in accordance with
#106) and adds an
`is_empty` check to the result of calling `CharClass::to_byte_class`,
which allows the test to pass.

Author: scooter-dangle

regex-syntax/src/parser.rs

@@ -596,7 +596,17 @@ impl Parser {
         Ok(Build::Expr(if self.flags.unicode {
             Expr::Class(class)
         } else {
-            Expr::ClassBytes(class.to_byte_class())
+            let byte_class = class.to_byte_class();
+
+            // If `class` was only non-empty due to multibyte characters, the 
+            // corresponding byte class will now be empty.
+            //
+            // See https://github.com/rust-lang-nursery/regex/issues/303
+            if byte_class.is_empty() {
+                return Err(self.err(ErrorKind::EmptyClass));
+            }
+
+            Expr::ClassBytes(byte_class)
         }))
     }
 

tests/bytes.rs

@@ -53,3 +53,9 @@ matiter!(invalidutf8_anchor3,
          r"^|ddp\xff\xffdddddlQd@\x80",
          R(b"\x8d#;\x1a\xa4s3\x05foobarX\\\x0f0t\xe4\x9b\xa4"),
          (0, 0));
+
+// See https://github.com/rust-lang-nursery/regex/issues/303
+#[test]
+fn negated_full_byte_range() {
+     assert!(::regex::bytes::Regex::new(r#"[^\x00-\xff]"#).is_err());
+}