Fix #62545: wrong unicode mapping in some charsets

Undefined characters are best mapped to Unicode REPLACEMENT characters.
php · Mar 11, 2018 · 01ea314 · 01ea314
1 parent 76fc73c
commit 01ea314
Show file tree

Hide file tree

Showing 4 changed files with 26 additions and 5 deletions.
diff --git a/NEWS b/NEWS
@@ -100,6 +100,9 @@ PHP                                                                        NEWS
 - IMAP:
   . Fixed bug #75774 (imap_append HeapCorruction). (Anatol)
 
+- Mbstring:
+  . Fixed bug #62545 (wrong unicode mapping in some charsets). (cmb)
+
 - Opcache:
   . Fixed bug #75720 (File cache not populated after SHM runs full). (Dmitry)
   . Fixed bug #75579 (Interned strings buffer overflow may cause crash).

diff --git a/ext/mbstring/libmbfl/filters/unicode_table_cp1251.h b/ext/mbstring/libmbfl/filters/unicode_table_cp1251.h
@@ -30,7 +30,7 @@ static const unsigned short cp1251_ucs_table[] = {
  0x0402, 0x0403, 0x201a, 0x0453, 0x201e, 0x2026, 0x2020, 0x2021,
  0x20ac, 0x2030, 0x0409, 0x2039, 0x040a, 0x040c, 0x040b, 0x040f,
  0x0452, 0x2018, 0x2019, 0x201c, 0x201d, 0x2022, 0x2013, 0x2014,
- 0x003f, 0x2122, 0x0459, 0x203a, 0x045a, 0x045c, 0x045b, 0x045f,
+ 0xfffd, 0x2122, 0x0459, 0x203a, 0x045a, 0x045c, 0x045b, 0x045f,
  0x00a0, 0x040e, 0x045e, 0x0408, 0x00a4, 0x0490, 0x00a6, 0x00a7,
  0x0401, 0x00a9, 0x0404, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x0407,
  0x00b0, 0x00b1, 0x0406, 0x0456, 0x0491, 0x00b5, 0x00b6, 0x00b7,

diff --git a/ext/mbstring/libmbfl/filters/unicode_table_cp1252.h b/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,0xfffe,0x201a,0x0192,0x201e,0x2026,0x2020,0x2021,
- 0x02c6,0x2030,0x0160,0x2039,0x0152,0xfffe,0x017d,0xfffe,
- 0xfffe,0x2018,0x2019,0x201c,0x201d,0x2022,0x2013,0x2014,
- 0x02dc,0x2122,0x0161,0x203a,0x0153,0xfffe,0x017e,0x0178
+ 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
 };
 #endif /* UNICODE_TABLE_CP1252_H */
diff --git a/ext/mbstring/tests/bug62545.phpt b/ext/mbstring/tests/bug62545.phpt
@@ -0,0 +1,18 @@
+--TEST--
+Bug #62545 (wrong unicode mapping in some charsets)
+--SKIPIF--
+<?php
+if (!extension_loaded('mbstring')) die('skip mbstring extension not available');
+?>
+--FILE--
+<?php
+var_dump(
+    bin2hex(mb_convert_encoding("\x98", 'UTF-8', 'Windows-1251')),
+    bin2hex(mb_convert_encoding("\x81\x8d\x8f\x90\x9d", 'UTF-8', 'Windows-1252'))
+);
+?>
+===DONE===
+--EXPECT--
+string(6) "efbfbd"
+string(30) "efbfbdefbfbdefbfbdefbfbdefbfbd"
+===DONE===