lexer: skip interpreter line in any source buffer

Skip interpreter lines in any source buffer and handle the skipping in the
lexer itself, to avoid reporting wrongly shifted token offsets to the
compiler, resulting in wrong error locations and source contexts.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

1 files changed, 36 insertions, 0 deletions

diff --git a/lexer.c b/lexer.c
index d9d657e..25a5cf4 100644
--- a/lexer.c
+++ b/lexer.c
@@ -1141,6 +1141,38 @@ lex_step(uc_lexer *lex, FILE *fp)
 	return NULL;
 }
 
+static void
+uc_lexer_skip_shebang(uc_lexer *lex)
+{
+	uc_source *source = lex->source;
+	FILE *fp = source->fp;
+	int c1, c2;
+
+	c1 = fgetc(fp);
+	c2 = fgetc(fp);
+
+	if (c1 == '#' && c2 == '!') {
+		next_lineinfo(lex);
+
+		source->off += 2;
+
+		while ((c1 = fgetc(fp)) != EOF) {
+			source->off++;
+
+			if (c1 == '\n') {
+				update_lineinfo(lex, source->off);
+				next_lineinfo(lex);
+
+				break;
+			}
+		}
+	}
+	else {
+		ungetc(c2, fp);
+		ungetc(c1, fp);
+	}
+}
+
 void
 uc_lexer_init(uc_lexer *lex, uc_parse_config *config, uc_source *source)
 {
@@ -1171,6 +1203,10 @@ uc_lexer_init(uc_lexer *lex, uc_parse_config *config, uc_source *source)
 	lex->lead_surrogate = 0;
 
 	lex->lastoff = 0;
+
+	/* Skip any potential interpreter line */
+	if (lex->source->off == 0)
+		uc_lexer_skip_shebang(lex);
 }
 
 void