Pass negative eval tests (reject invalid escaped URI characters).

git-svn-id: http://svn.drobilla.net/serd/trunk@430 490d8e77-9747-427b-9fa3-0b8f29cee8a0

src/reader.c

@@ -280,7 +280,7 @@ read_HEX(SerdReader* reader)
 }
 
 // Read UCHAR escape, initial \ is already eaten by caller
-static inline bool
+static inline uint32_t
 read_UCHAR(SerdReader* reader, Ref dest)
 {
 	const uint8_t b      = peek_byte(reader);
@@ -293,38 +293,39 @@ read_UCHAR(SerdReader* reader, Ref dest)
 		length = 4;
 		break;
 	default:
-		return false;
+		return 0;
 	}
 	eat_byte_safe(reader, b);
 
 	uint8_t buf[9] = { 0, 0, 0, 0, 0, 0, 0, 0, 0 };
 	for (unsigned i = 0; i < length; ++i) {
 		if (!(buf[i] = read_HEX(reader))) {
-			return false;
+			return 0;
 		}
 	}
 
-	uint32_t c;
-	sscanf((const char*)buf, "%X", &c);
+	uint32_t code;
+	sscanf((const char*)buf, "%X", &code);
 
 	unsigned size = 0;
-	if (c < 0x00000080) {
+	if (code < 0x00000080) {
 		size = 1;
-	} else if (c < 0x00000800) {
+	} else if (code < 0x00000800) {
 		size = 2;
-	} else if (c < 0x00010000) {
+	} else if (code < 0x00010000) {
 		size = 3;
-	} else if (c < 0x00110000) {
+	} else if (code < 0x00110000) {
 		size = 4;
 	} else {
 		r_err(reader, SERD_ERR_BAD_SYNTAX,
-		      "unicode character 0x%X out of range\n", c);
+		      "unicode character 0x%X out of range\n", code);
 		push_replacement(reader, dest);
-		return true;
+		return 0xFFFD;
 	}
 
 	// Build output in buf
 	// (Note # of bytes = # of leading 1 bits in first byte)
+	uint32_t c = code;
 	switch (size) {
 	case 4:
 		buf[3] = 0x80 | (uint8_t)(c & 0x3F);
@@ -345,7 +346,7 @@ read_UCHAR(SerdReader* reader, Ref dest)
 	for (unsigned i = 0; i < size; ++i) {
 		push_byte(reader, dest, buf[i]);
 	}
-	return true;
+	return code;
 }
 
 // Read ECHAR escape, initial \ is already eaten by caller
@@ -741,7 +742,8 @@ static Ref
 read_IRIREF(SerdReader* reader)
 {
 	TRY_RET(eat_byte_check(reader, '<'));
-	Ref ref = push_node(reader, SERD_URI, "", 0);
+	Ref      ref = push_node(reader, SERD_URI, "", 0);
+	uint32_t code;
 	while (true) {
 		const uint8_t c = peek_byte(reader);
 		switch (c) {
@@ -754,9 +756,10 @@ read_IRIREF(SerdReader* reader)
 			return ref;
 		case '\\':
 			eat_byte_safe(reader, c);
-			if (!read_UCHAR(reader, ref)) {
+			switch (code = read_UCHAR(reader, ref)) {
+			case 0: case ' ': case '<': case '>':
 				r_err(reader, SERD_ERR_BAD_SYNTAX,
-				      "invalid IRI character `%c'\n", c);
+				      "invalid escaped IRI character %X %c\n", code, code);
 				return pop_node(reader, ref);
 			}
 			break;

tests/good/test-escapes.nt

@@ -1,2 +1,2 @@
 <http://example.org/thing> <http://example.org/comment> "\\\r\n\t" .
-<http://example.org/thing> <http://example.org/comment> <http://example.org/\u003E> .
+<http://example.org/thing> <http://example.org/comment> <http://example.org/\u005C> .

tests/good/test-escapes.ttl

@@ -1,2 +1,2 @@
 <http://example.org/thing> <http://example.org/comment> "\\\r\n\t" .
-<http://example.org/thing> <http://example.org/comment> <http://example.org/\u003E> .
+<http://example.org/thing> <http://example.org/comment> <http://example.org/\u005C> .

tests/good/test-uri-escape.nt

@@ -1 +1 @@
-<http://example.org/node> <http://example.org/prop> <scheme:\u0001\u0002\u0003\u0004\u0005\u0006\u0007\u0008\u000B\u000C\u000E\u000F\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017\u0018\u0019\u001A\u001B\u001C\u001D\u001E\u001F\u0020!\u0022#$%&'()*+,-./0123456789:/\u003C=\u003E?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\u005C]\u005E_\u0060abcdefghijklmnopqrstuvwxyz\u007B\u007C\u007D~\u007F> .
+<http://example.org/node> <http://example.org/prop> <scheme:\u0001\u0002\u0003\u0004\u0005\u0006\u0007\u0008\u000B\u000C\u000E\u000F\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017\u0018\u0019\u001A\u001B\u001C\u001D\u001E\u001F%20!\u0022#$%&'()*+,-./0123456789:/%3C=%3E?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\u005C]\u005E_\u0060abcdefghijklmnopqrstuvwxyz\u007B\u007C\u007D~\u007F> .

tests/good/test-uri-escape.ttl

@@ -1 +1 @@
-<http://example.org/node> <http://example.org/prop> <scheme:\u0001\u0002\u0003\u0004\u0005\u0006\u0007\u0008\u000B\u000C\u000E\u000F\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017\u0018\u0019\u001A\u001B\u001C\u001D\u001E\u001F\u0020!\u0022#$%&'()*+,-./0123456789:/\u003C=\u003E?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\u005C]\u005E_\u0060abcdefghijklmnopqrstuvwxyz\u007B\u007C\u007D~\u007F> .
+<http://example.org/node> <http://example.org/prop> <scheme:\u0001\u0002\u0003\u0004\u0005\u0006\u0007\u0008\u000B\u000C\u000E\u000F\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017\u0018\u0019\u001A\u001B\u001C\u001D\u001E\u001F%20!\u0022#$%&'()*+,-./0123456789:/%3C=%3E?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\u005C]\u005E_\u0060abcdefghijklmnopqrstuvwxyz\u007B\u007C\u007D~\u007F> .

wscript

@@ -348,6 +348,14 @@ def test_manifest(ctx, srcdir, testdir, report, test_base, parse_base):
         passed = run_test(action_node, 1)
         report.write(earl_assertion(test, passed, asserter))
 
+    for i in sorted(model.triples([None, rdf.type, rdft.TestTurtleNegativeEval])):
+        test        = i[0]
+        name        = model.value(test, mf.name, None)
+        action_node = model.value(test, mf.action, None)[len(test_base):]
+
+        passed = run_test(action_node, 1)
+        report.write(earl_assertion(test, passed, asserter))
+
     for i in sorted(model.triples([None, rdf.type, rdft.TestTurtleEval])):
         test        = i[0]
         name        = model.value(test, mf.name, None)