Fix #63208: BSTR to PHP string conversion not binary safe

A `BSTR` is similar to a `zend_string`; it stores the length of the
string just before the actual string, and thus the string may contain
NUL bytes.  However, `php_com_olestring_to_string()` is supposed to
deal with arbitrary `OLECHAR*`s which may not be `BSTR`s, so we
introduce `php_com_bstr_to_string()` and use it for the only case where
we actually have to deal with `BSTR`s which may contain NUL bytes.

Contrary to `php_com_olestring_to_string()` we return a `zend_string`,
so we can save the re-allocation when converting to a `zval`.

We also cater to `php_com_string_to_olestring()` not being binary safe,
with basically the same fix we did for `php_com_olestring_to_string()`.

NEWS

@@ -6,6 +6,9 @@ PHP                                                                        NEWS
   . Fixed bug #79030 (Upgrade apache2handler's php_apache_sapi_get_request_time
     to return usec). (Herbert256)
 
+- COM:
+  . Fixed bug #63208 (BSTR to PHP string conversion not binary safe). (cmb)
+  
 - Curl:
   . Fixed bug #79741 (curl_setopt CURLOPT_POSTFIELDS asserts on object with
     declared properties). (Nikita)

ext/com_dotnet/com_olechar.c

@@ -103,3 +103,58 @@ PHP_COM_DOTNET_API char *php_com_olestring_to_string(OLECHAR *olestring, size_t
 
 	return string;
 }
+
+BSTR php_com_string_to_bstr(zend_string *string, int codepage)
+{
+	BSTR bstr = NULL;
+	DWORD flags = codepage == CP_UTF8 ? 0 : MB_PRECOMPOSED | MB_ERR_INVALID_CHARS;
+	size_t mb_len = ZSTR_LEN(string);
+	int wc_len;
+
+	if ((wc_len = MultiByteToWideChar(codepage, flags, ZSTR_VAL(string), (int)mb_len + 1, NULL, 0)) <= 0) {
+		goto fail;
+	}
+	if ((bstr = SysAllocStringLen(NULL, (UINT)(wc_len - 1))) == NULL) {
+		goto fail;
+	}
+	if ((wc_len = MultiByteToWideChar(codepage, flags, ZSTR_VAL(string), (int)mb_len + 1, bstr, wc_len)) <= 0) {
+		goto fail;
+	}
+	return bstr;
+
+fail:
+	char *msg = php_win32_error_to_msg(GetLastError());
+	php_error_docref(NULL, E_WARNING,
+		"Could not convert string to unicode: `%s'", msg);
+	LocalFree(msg);
+	SysFreeString(bstr);
+	return SysAllocString(L"");
+}
+
+zend_string *php_com_bstr_to_string(BSTR bstr, int codepage)
+{
+	zend_string *string = NULL;
+	UINT wc_len = SysStringLen(bstr);
+	int mb_len;
+
+	mb_len = WideCharToMultiByte(codepage, 0, bstr, wc_len + 1, NULL, 0, NULL, NULL);
+	if (mb_len > 0) {
+		string = zend_string_alloc(mb_len - 1, 0);
+		mb_len = WideCharToMultiByte(codepage, 0, bstr, wc_len + 1, ZSTR_VAL(string), mb_len, NULL, NULL);
+	}
+
+	if (mb_len <= 0) {
+		char *msg = php_win32_error_to_msg(GetLastError());
+
+		php_error_docref(NULL, E_WARNING,
+			"Could not convert string from unicode: `%s'", msg);
+		LocalFree(msg);
+
+		if (string != NULL) {
+			zend_string_release(string);
+		}
+		string = ZSTR_EMPTY_ALLOC();
+	}
+
+	return string;
+}

ext/com_dotnet/com_variant.c

@@ -96,7 +96,6 @@ bogus:
 
 PHP_COM_DOTNET_API void php_com_variant_from_zval(VARIANT *v, zval *z, int codepage)
 {
-	OLECHAR *olestring;
 	php_com_dotnet_object *obj;
 	zend_uchar ztype = IS_NULL;
 
@@ -164,13 +163,7 @@ PHP_COM_DOTNET_API void php_com_variant_from_zval(VARIANT *v, zval *z, int codep
 
 		case IS_STRING:
 			V_VT(v) = VT_BSTR;
-			olestring = php_com_string_to_olestring(Z_STRVAL_P(z), Z_STRLEN_P(z), codepage);
-			if (CP_UTF8 == codepage) {
-				V_BSTR(v) = SysAllocStringByteLen((char*)olestring, (UINT)(wcslen(olestring) * sizeof(OLECHAR)));
-			} else {
-				V_BSTR(v) = SysAllocStringByteLen((char*)olestring, (UINT)(Z_STRLEN_P(z) * sizeof(OLECHAR)));
-			}
-			efree(olestring);
+			V_BSTR(v) = php_com_string_to_bstr(Z_STR_P(z), codepage);
 			break;
 
 		case IS_RESOURCE:
@@ -236,12 +229,8 @@ PHP_COM_DOTNET_API int php_com_zval_from_variant(zval *z, VARIANT *v, int codepa
 		case VT_BSTR:
 			olestring = V_BSTR(v);
 			if (olestring) {
-				size_t len;
-				char *str = php_com_olestring_to_string(olestring,
-					&len, codepage);
-				ZVAL_STRINGL(z, str, len);
-				// TODO: avoid reallocation???
-				efree(str);
+				zend_string *str = php_com_bstr_to_string(olestring, codepage);
+				ZVAL_STR(z, str);
 				olestring = NULL;
 			}
 			break;

ext/com_dotnet/php_com_dotnet_internal.h

@@ -89,6 +89,8 @@ PHP_COM_DOTNET_API char *php_com_olestring_to_string(OLECHAR *olestring,
 		size_t *string_len, int codepage);
 PHP_COM_DOTNET_API OLECHAR *php_com_string_to_olestring(char *string,
 		size_t string_len, int codepage);
+BSTR php_com_string_to_bstr(zend_string *string, int codepage);
+zend_string *php_com_bstr_to_string(BSTR bstr, int codepage);
 
 
 /* com_com.c */

ext/com_dotnet/tests/bug63208.phpt

@@ -0,0 +1,17 @@
+--TEST--
+Bug #63208 (BSTR to PHP string conversion not binary safe)
+--SKIPIF--
+<?php
+if (!extension_loaded('com_dotnet')) die('skip com_dotnet extension not available');
+?>
+--FILE--
+<?php
+$string = "\u{0905}b\0cd";
+$variant = new VARIANT($string, VT_ARRAY | VT_UI1, CP_UTF8); // Array of bytes
+$converted = (string) $variant;
+var_dump(bin2hex($string));
+var_dump(bin2hex($converted));
+?>
+--EXPECT--
+string(14) "e0a48562006364"
+string(14) "e0a48562006364"