parser: assert symbol names are valid

Introduce symbol_valid(), which returns true only if a given symbol
name is valid.  A valid symbol begins with an alphabetic character
and consists otherwise of alphanumerics or '_'.  It also must be
representable in the token buffer.

Signed-off-by: Alex Elder <elder@linaro.org>
Message-Id: <20211001232338.769309-29-elder@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>

1 files changed, 29 insertions, 0 deletions

diff --git a/parser.c b/parser.c
index 2498b79..a684a29 100644
--- a/parser.c
+++ b/parser.c
@@ -119,11 +119,40 @@ struct symbol {
 
 static struct list_head symbols = LIST_INIT(symbols);
 
+static bool symbol_valid(const char *name)
+{
+	const char *p = name;
+	struct symbol *sym;
+	char ch;
+
+	/* Symbol name must start with an alphabetic character */
+	if (!p || !isalpha(*p++))
+		return false;
+
+	/* Remainder of the name is alphanumeric or underscore */
+	while ((ch = *p++))
+		if (!(isalnum(ch) || ch == '_'))
+			return false;
+
+	/* Symbol name must fit in the token buffer */
+	if (p - name > TOKEN_BUF_SIZE)
+		return 0;
+
+	/* Finally, symbol names must be unique */
+	list_for_each_entry(sym, &symbols, node)
+		if (!strcmp(name, sym->name))
+			return false;
+
+	return true;
+}
+
 static void symbol_add(const char *name, enum token_id token_id, ...)
 {
 	struct symbol *sym;
 	va_list ap;
 
+	assert(symbol_valid(name));
+
 	va_start(ap, token_id);
 
 	sym = memalloc(sizeof(struct symbol));