shithub: sl

Info  •  Files  •  Log  •  Branches

string-encode: fix to work on string containing \0

--- a/src/string.c

+++ b/src/string.c

@@ -87,7 +87,7 @@

 BUILTIN("string-decode", string_decode)

 {

-	int term = 0;

+	bool term = false;

 	if(nargs == 2)

 		term = args[1] != FL_f;

 	else

@@ -97,7 +97,7 @@

 	cvalue_t *cv = ptr(args[0]);

 	char *ptr = (char*)cv_data(cv);

 	size_t nb = cv_len(cv);

-	size_t nc = utfnlen(ptr, nb);

+	size_t nc = u8_runelen(ptr, nb);

 	size_t newsz = nc*sizeof(Rune);

 	if(term)

 		newsz += sizeof(Rune);

--- a/src/utf8.c

+++ b/src/utf8.c

@@ -38,6 +38,15 @@

 	return trailingBytesForUTF8[(uint8_t)s[0]] + 1;

 }

+size_t

+u8_runelen(const char *s, size_t nb)

+{

+	size_t nr, i;

+	for(i = nr = 0; i < nb; nr++)

+		i += trailingBytesForUTF8[(uint8_t)s[i]] + 1;

+	return nr;

+}

+

 /* byte offset => charnum */

 size_t

 u8_charnum(const char *s, size_t offset)

--- a/src/utf8.h

+++ b/src/utf8.h

@@ -12,6 +12,9 @@

 /* returns length of next utf-8 sequence */

 size_t u8_seqlen(const char *s) fl_purefn;

+/* length of a utf-8 string in runes */

+size_t u8_runelen(const char *s, size_t nb) fl_purefn;

+

 char read_escape_control_char(char c) fl_constfn;

 /* given a wide character, convert it to an ASCII escape sequence stored in

--- a/test/unittest.lsp

+++ b/test/unittest.lsp

@@ -492,6 +492,9 @@

   (io-close b))

 (define s "привет\0пока")

+

+(assert (equal? s (string-encode (string-decode s))))

+

 (assert (eq? 21 (sizeof s)))

 (assert (eq? 21 (length s)))

 (assert (eq? 11 (string-length s)))