work

1 files changed, 54 insertions, 0 deletions

diff --git a/tests/test/cpp/lex2.scm b/tests/test/cpp/lex2.scm
index 47bb4a16..b7087c3b 100644
--- a/tests/test/cpp/lex2.scm
+++ b/tests/test/cpp/lex2.scm
@@ -92,3 +92,57 @@
         (lexeme type: 'preprocessing-token body: '(punctuator "."))
         (lexeme type: 'preprocessing-token body: '(identifier "dir")))
   (lex "..\\listing.dir"))
+
+
+(test-equal "Propper H-string"
+  (list (lexeme type: 'preprocessing-token body: '(header-name (h-string "a"))))
+  (lex "<a>"))
+
+(test-equal "Unexpected h-string"
+  (list (lexeme type: 'preprocessing-token body: '(pp-number "1"))
+        (lexeme type: 'whitespace body: " ")
+        (lexeme type: 'preprocessing-token body: '(header-name (h-string " 2 ")))
+        (lexeme type: 'whitespace body: " ")
+        (lexeme type: 'preprocessing-token body: '(pp-number "3")))
+  (lex "1 < 2 > 3"))
+
+(test-equal "Quotation mark inside h-string"
+  (list (lexeme type: 'preprocessing-token body: '(header-name (h-string "a\"b"))))
+  (lex "<a\"b>"))
+
+(test-equal "Interaction of h-strings and regular strings"
+ (test-equal "Less than string, not h-string"
+   (list (lexeme type: 'preprocessing-token body: '(pp-number "1"))
+         (lexeme type: 'preprocessing-token body: '(string-literal "<"))
+         (lexeme type: 'preprocessing-token body: '(punctuator ">")))
+   (lex "1\"<\">"))
+
+ (test-equal "H-string, not string"
+   (list (lexeme type: 'preprocessing-token body: '(pp-number "1"))
+         (lexeme type: 'preprocessing-token body: '(header-name (h-string "\"")))
+         (lexeme type: 'other body: "\""))
+   (lex "1<\">\"")))
+
+(test-equal "Q-strings are lexed as regular strings"
+  (list (lexeme type: 'preprocessing-token body: '(punctuator "#"))
+        (lexeme type: 'preprocessing-token body: '(identifier "include"))
+        (lexeme type: 'whitespace body: " ")
+        (lexeme type: 'preprocessing-token body: '(string-literal "test")))
+  ;; # include here, since generated tokens could possible depend on that context,
+  ;; and the reason regular strings are returned is since the lexer doesn't check
+  ;; that context
+  (lex "#include \"test\"")
+  )
+
+
+
+(test-group "Unicode"
+  (test-equal "In string literals"
+    (list (lexeme type: 'preprocessing-token body: '(string-literal "åäö")))
+    (lex "\"åäö\""))
+
+  (test-equal "Outside string literals"
+    (list (lexeme type: 'other body: "å")
+          (lexeme type: 'other body: "ä")
+          (lexeme type: 'other body: "ö"))
+    (lex "åäö")))