Fix mbstring support for CP1252 encoding

It's a bit surprising how much was broken here.

- Identify filter was utterly and completely wrong.
- Instead of handling invalid CP1252 bytes as specified by
  `mb_substitute_character`, it would convert them to Unicode 0xFFFD
  (generic replacement character).
- When converting ISO-8859-1 to CP1252, invalid ISO-8859-1 bytes would
  be passed through silently.
- Unicode codepoints from 0x80-0x9F were converted to CP1252 bytes 0x80-0x9F,
  which is wrong.
- Unicode codepoint 0xFFFD was converted to CP1252 0x9F, which is very wrong.

Also clean up some unneeded code, and make the conversion table consistent with
others by using zero as a 'invalid' marker, rather than 0xFFFD.

Author: alexdowad

ext/mbstring/libmbfl/filters/mbfilter_cp1252.c

@@ -74,32 +74,22 @@ const struct mbfl_convert_vtbl vtbl_wchar_cp1252 = {
 
 #define CK(statement)	do { if ((statement) < 0) return (-1); } while (0)
 
-/*
- * wchar => cp1252
- */
 int mbfl_filt_conv_wchar_cp1252(int c, mbfl_convert_filter *filter)
 {
-	int s=-1, n;
+	int s = -1;
 
 	if (c >= 0x100)	{
-		/* look it up from the cp1252 table */
-		s = -1;
-		n = 31;
-		while (n >= 0) {
-			if (c == cp1252_ucs_table[n] && c != 0xfffe) {
+		/* Look it up from the CP1252 table */
+		for (int n = 31; n >= 0; n--) {
+			if (c == cp1252_ucs_table[n]) {
 				s = 0x80 + n;
 				break;
 			}
-			n--;
-		}
-		if (s <= 0 && (c & ~MBFL_WCSPLANE_MASK) == MBFL_WCSPLANE_8859_1)
-		{
-			s = c & MBFL_WCSPLANE_MASK;
 		}
-	}
-	else if (c >= 0 && c < 0x100) {
+	} else if (c <= 0x7F || c >= 0xA0) {
 		s = c;
 	}
+
 	if (s >= 0) {
 		CK((*filter->output_function)(s, filter->data));
 	} else {
@@ -108,15 +98,15 @@ int mbfl_filt_conv_wchar_cp1252(int c, mbfl_convert_filter *filter)
 	return c;
 }
 
-/*
- * cp1252 => wchar
- */
 int mbfl_filt_conv_cp1252_wchar(int c, mbfl_convert_filter *filter)
 {
 	int s;
 
-	if (c >= 0x80 && c < 0xa0) {
+	if (c >= 0x80 && c < 0xA0) {
 		s = cp1252_ucs_table[c - 0x80];
+		if (!s) {
+			s = c | MBFL_WCSGROUP_THROUGH;
+		}
 	} else {
 		s = c;
 	}
@@ -126,17 +116,10 @@ int mbfl_filt_conv_cp1252_wchar(int c, mbfl_convert_filter *filter)
 	return c;
 }
 
-/* We only distinguish the MS extensions to ISO-8859-1.
- * Actually, this is pretty much a NO-OP, since the identification
- * system doesn't allow us to discriminate between a positive match,
- * a possible match and a definite non-match.
- * The problem here is that cp1252 looks like SJIS for certain chars.
- * */
 static int mbfl_filt_ident_cp1252(int c, mbfl_identify_filter *filter)
 {
-	if (c >= 0x80 && c < 0xa0)
-		filter->flag = 0;
-	else
-		filter->flag = 1; /* not it */
+	if (c >= 0x80 && c < 0xA0 && !cp1252_ucs_table[c - 0x80]) {
+		filter->flag = 1;
+	}
 	return c;
 }

ext/mbstring/libmbfl/filters/unicode_table_cp1252.h

@@ -32,9 +32,9 @@
  * as it only covers this range, while the rest cover 0xa0 onwards */
 
 static const unsigned short cp1252_ucs_table[] = {
- 0x20ac,0xfffd,0x201a,0x0192,0x201e,0x2026,0x2020,0x2021,
- 0x02c6,0x2030,0x0160,0x2039,0x0152,0xfffd,0x017d,0xfffd,
- 0xfffd,0x2018,0x2019,0x201c,0x201d,0x2022,0x2013,0x2014,
- 0x02dc,0x2122,0x0161,0x203a,0x0153,0xfffd,0x017e,0x0178
+ 0x20ac,0x0000,0x201a,0x0192,0x201e,0x2026,0x2020,0x2021,
+ 0x02c6,0x2030,0x0160,0x2039,0x0152,0x0000,0x017d,0x0000,
+ 0x0000,0x2018,0x2019,0x201c,0x201d,0x2022,0x2013,0x2014,
+ 0x02dc,0x2122,0x0161,0x203a,0x0153,0x0000,0x017e,0x0178
 };
 #endif /* UNICODE_TABLE_CP1252_H */