Avoid writing illegal Turtle names as a result of URI qualifying (fix #734).

git-svn-id: http://svn.drobilla.net/serd/trunk@210 490d8e77-9747-427b-9fa3-0b8f29cee8a0

3 files changed, 41 insertions, 3 deletions

diff --git a/ChangeLog b/ChangeLog
index 5337a87a..2bc39b71 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,6 +1,7 @@
 serd (UNRELEASED) unstable; urgency=low
 
   * Fix pretty printing of successive blank descriptions, i.e. "] , ["
+  * Avoid writing illegal Turtle names as a result of URI qualifying
 
  -- David Robillard <d@drobilla.net>  UNRELEASED
 
diff --git a/src/env.c b/src/env.c
index 4ce28329..f7204121 100644
--- a/src/env.c
+++ b/src/env.c
@@ -152,6 +152,41 @@ serd_env_set_prefix(SerdEnv*        env,
 	return SERD_SUCCESS;
 }
 
+static inline bool
+is_nameStartChar(const uint8_t c)
+{
+	// TODO: not strictly correct (see reader.c read_nameStartChar)
+	return (c >= 'A' && c <= 'Z') || c == '_' || (c >= 'a' && c <= 'z');
+}
+
+static inline bool
+is_nameChar(const uint8_t c)
+{
+	// TODO: 0x300-0x036F | 0x203F-0x2040 (see reader.c read_nameChar)
+	return is_nameStartChar(c) || c == '-'
+		|| (c >= '0' && c <= '9') || c == 0xB7;
+}
+
+/**
+   Return true iff @c buf is a valid Turtle name.
+*/
+static inline bool
+is_name(const uint8_t* buf,
+        size_t         len)
+{
+	if (!is_nameStartChar(buf[0])) {
+		return false;
+	}
+
+	for (size_t i = 1; i < len; ++i) {
+		if (!is_nameChar(buf[i])) {
+			return false;
+		}
+	}
+
+	return true;
+}
+
 SERD_API
 bool
 serd_env_qualify(const SerdEnv*  env,
@@ -168,6 +203,9 @@ serd_env_qualify(const SerdEnv*  env,
 				*prefix_name = env->prefixes[i].name;
 				suffix->buf = uri->buf + prefix_uri->n_bytes;
 				suffix->len = uri->n_bytes - prefix_uri->n_bytes;
+				if (!is_name(suffix->buf, suffix->len)) {
+					continue;
+				}
 				return true;
 			}
 		}
diff --git a/src/reader.c b/src/reader.c
index e08956ef..24802164 100644
--- a/src/reader.c
+++ b/src/reader.c
@@ -711,7 +711,7 @@ static inline uchar
 read_nameStartChar(SerdReader* reader, bool required)
 {
 	const uint8_t c = peek_byte(reader);
-	if (c == '_' || is_alpha(c)) {
+	if (c == '_' || is_alpha(c)) {  // TODO: not strictly correct
 		return eat_byte(reader, c);
 	} else {
 		if (required) {
@@ -734,8 +734,7 @@ read_nameChar(SerdReader* reader)
 	case '-': case 0xB7: case '0': case '1': case '2': case '3': case '4':
 	case '5': case '6': case '7': case '8': case '9':
 		return eat_byte(reader, c);
-	default:
-		// TODO: 0x300-0x036F | 0x203F-0x2040
+	default:  // TODO: 0x300-0x036F | 0x203F-0x2040
 		return 0;
 	}
 	return 0;