Skip shebang

1 files changed, 12 insertions, 7 deletions

diff --git a/src/tokenizer.c b/src/tokenizer.c
index de467c3..9ce1bba 100644
--- a/src/tokenizer.c
+++ b/src/tokenizer.c
@@ -29,11 +29,6 @@ static int tokenizer_get_end_of_multiline_comment(Tokenizer *self, int index);
 int tokenizer_init(Tokenizer *self, ArenaAllocator *allocator, BufferView code, BufferView code_name, const amal_compiler_options *compiler_options) {
     assert(code.size <= INT_MAX);
     assert(compiler_options);
-    /* Skip UTF-8 BOM */
-    if(code.size >= 3 && (u8)code.data[0] == 0xEF && (u8)code.data[1] == 0xBB && (u8)code.data[2] == 0xBF) {
-        code.data += 3;
-        code.size -= 3;
-    }
     self->code = code;
     self->index = 0;
     self->prev_index = 0;
@@ -43,6 +38,17 @@ int tokenizer_init(Tokenizer *self, ArenaAllocator *allocator, BufferView code,
     self->number_is_integer = bool_false;
     self->allocator = allocator;
     self->compiler_options = compiler_options;
+
+    /* Skip UTF-8 BOM */
+    if(self->code.size >= 3 && am_memcmp(self->code.data, "\xEF\xBB\xBF", 3) == 0) {
+        self->code.data += 3;
+        self->code.size -= 3;
+    }
+    /* Skip shebang */
+    if(self->code.size >= 2 && am_memcmp(self->code.data, "#!", 2) == 0) {
+        self->index = tokenizer_get_end_of_line_from_index(self, self->index + 2);
+    }
+
     return 0;
 }
 
@@ -608,9 +614,8 @@ int tokenizer_get_start_of_line_from_index(Tokenizer *self, int index) {
 }
 
 int tokenizer_get_end_of_line_from_index(Tokenizer *self, int index) {
-    char c;
     while(index < (int)self->code.size) {
-        c = self->code.data[index];
+        char c = self->code.data[index];
         if(c == '\n' || c == '\r')
             break;
         ++index;