From bbd9bab7ad73b24cb97944649859054411463d81 Mon Sep 17 00:00:00 2001
From: Leo Tenenbaum <pommicket@gmail.com>
Date: Sun, 19 Jan 2020 12:50:58 -0500
Subject: more work on #foreign

---
 tokenizer.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 50 insertions(+), 10 deletions(-)

(limited to 'tokenizer.c')

diff --git a/tokenizer.c b/tokenizer.c
index 1ca3b3b..8c26769 100644
--- a/tokenizer.c
+++ b/tokenizer.c
@@ -130,7 +130,19 @@ static inline void tokr_nextchar(Tokenizer *t) {
 	++t->s;
 }
 
-static char tokr_esc_seq(Tokenizer *t) {
+/* returns -1 if not a hex digit, otherwise 0-15 */
+static inline int char_as_hex_digit(char c) {
+	if (c >= '0' && c <= '9')
+		return c - '0';
+	if (c >= 'a' && c <= 'f')
+		return 10 + c - 'a';
+	if (c >= 'A' && c <= 'F')
+		return 10 + c - 'A';
+	return -1;
+}
+
+/* returns -1 if escape sequence is invalid */
+static int tokr_esc_seq(Tokenizer *t) {
 	/* TODO: add more of these incl. \x41, \100 */
 	switch (*t->s) {
 	case '\'':
@@ -145,8 +157,21 @@ static char tokr_esc_seq(Tokenizer *t) {
 	case 'n':
 		tokr_nextchar(t);
 		return '\n';
+	case '0':
+		tokr_nextchar(t);
+		return '\0';
+	case 'x': {
+		int c1 = char_as_hex_digit(t->s[1]);
+		if (c1 == -1) return 0;
+		int c2 = char_as_hex_digit(t->s[2]);
+		if (c2 == -1) return 0;
+		tokr_nextchar(t);
+		tokr_nextchar(t);
+		tokr_nextchar(t);
+		return (char)(c1 * 16 + c2);
+	}	
 	default:
-		return 0;
+		return -1;
 	}
 
 }
@@ -159,8 +184,16 @@ static Location token_location(Token *t) {
 }
 
 /* for use during tokenization */
-static void tokenization_err(Tokenizer *t, const char *fmt, ...) {
+static void tokenization_err_(
+#if ERR_SHOW_SOURCE_LOCATION
+							  const char *src_file, int src_line,
+#endif
+							  Tokenizer *t, const char *fmt, ...) {
 	va_list args;
+	if (!t->err_ctx->enabled) return;
+#if ERR_SHOW_SOURCE_LOCATION
+	err_fprint("Generated by line %d of %s:\n", src_line, src_file);
+#endif
 	va_start(args, fmt);
 	err_text_err(t->err_ctx, "error");
 	err_fprint(" at line %lu of %s:\n", (unsigned long)t->line, t->err_ctx->filename);
@@ -178,15 +211,21 @@ static void tokenization_err(Tokenizer *t, const char *fmt, ...) {
 	}
 }
 
+#if ERR_SHOW_SOURCE_LOCATION
+#define tokenization_err(...) tokenization_err_(__FILE__, __LINE__, __VA_ARGS__)
+#else
+#define tokenization_err tokenization_err_
+#endif
+
 /* for use after tokenization */
 static void tokr_err_(
 #if ERR_SHOW_SOURCE_LOCATION
 					  const char *src_file, int src_line,
 #endif
 					  Tokenizer *t, const char *fmt, ...) {
+	if (!t->err_ctx->enabled) return;
 #if ERR_SHOW_SOURCE_LOCATION
-	if (!t->token->pos.ctx->enabled) return;
-	err_fprint("At line %d of %s:\n", src_line, src_file);
+	err_fprint("Generated by line %d of %s:\n", src_line, src_file);
 #endif
 	va_list args;
 	va_start(args, fmt);
@@ -454,11 +493,12 @@ static bool tokenize_string(Tokenizer *t, char *str) {
 			if (*t->s == '\\') {
 				/* escape sequence */
 				tokr_nextchar(t);
-				c = tokr_esc_seq(t);
-				if (c == 0) {
+				int e = tokr_esc_seq(t);
+				if (e == -1) {
 					tokenization_err(t, "Unrecognized escape character: '\\%c'.", *t->s);
 					goto err;
 				}
+				c = (char)e;
 			} else {
 				c = *t->s;
 				tokr_nextchar(t);
@@ -502,12 +542,12 @@ static bool tokenize_string(Tokenizer *t, char *str) {
 				assert(*t->s);
 				if (*t->s == '\\') {
 					tokr_nextchar(t);
-					char c = tokr_esc_seq(t);
-					if (c == 0) {
+					int c = tokr_esc_seq(t);
+					if (c == -1) {
 						tokenization_err(t, "Unrecognized escape character: '\\%c'.", *t->s);
 						goto err;
 					}
-					*strptr++ = c;
+					*strptr++ = (char)c;
 				} else {
 					*strptr++ = *t->s;
 					tokr_nextchar(t);
-- 
cgit v1.2.3