[tokenization] Exposing some of the scanner's methods in header for use in the Python scanner so it can avoid the additional allocation

This commit is contained in:
Al
2015-03-17 18:38:30 -04:00
parent daf3f8706b
commit f794ef7222
3 changed files with 29 additions and 11 deletions

View File

@@ -6,10 +6,6 @@
#include "scanner.h"
typedef struct scanner {
unsigned char *src, *cursor, *start, *end;
} scanner_t;
int scan_token(scanner_t *s)
{
s->start = s->cursor;
@@ -264787,10 +264783,8 @@ yy17828:
}
tokenized_string_t *tokenize(const char *input) {
size_t token_start, token_length;
int token_type;
scanner_t scanner_from_string(const char *input) {
unsigned char *s = (unsigned char *)input;
scanner_t scanner;
@@ -264799,6 +264793,15 @@ tokenized_string_t *tokenize(const char *input) {
scanner.start = s;
scanner.end = s + strlen(input);
return scanner;
}
tokenized_string_t *tokenize(const char *input) {
size_t token_start, token_length;
int token_type;
scanner_t scanner = scanner_from_string(input);
tokenized_string_t *response = tokenized_string_new();
while ( ( token_type = scan_token(&scanner)) != END ) {
@@ -264814,3 +264817,4 @@ tokenized_string_t *tokenize(const char *input) {
return response;
}