From: Dmitry Stogov <dmitry@php.net>
Date: Wed, 5 Sep 2007 10:18:23 +0000 (+0000)
Subject: Fixed bug #42488 (SoapServer reports an encoding error and the error itself breaks).
X-Git-Tag: php-5.2.5RC1~217
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=ee944bd1b5fa22f003805f760e63fa94fec0c487;p=php

Fixed bug #42488 (SoapServer reports an encoding error and the error itself breaks).
---

diff --git a/NEWS b/NEWS
index 77dd9b23d3..9c24c6204a 100644
--- a/NEWS
+++ b/NEWS
@@ -22,6 +22,8 @@ PHP                                                                        NEWS
   (Hannes)
 - Fixed bug #42468 (Write lock on file_get_contents fails when using a 
   compression stream). (Ilia)
+- Fixed bug #42488 (SoapServer reports an encoding error and the error itself
+  breaks). (Dmitry)
 - Fixed bug #42359 (xsd:list type not parsed). (Dmitry)
 - Fixed bug #42326 (SoapServer crash). (Dmitry)
 - Fixed bug #42086 (SoapServer return Procedure '' not present for WSIBasic
diff --git a/ext/soap/php_encoding.c b/ext/soap/php_encoding.c
index 65e703d5d2..8354f87ace 100644
--- a/ext/soap/php_encoding.c
+++ b/ext/soap/php_encoding.c
@@ -864,13 +864,50 @@ static xmlNodePtr to_xml_string(encodeTypePtr type, zval *data, int style, xmlNo
 			efree(str);
 			str = estrdup((char*)xmlBufferContent(out));
 			new_len = n;
-		} else if (!php_libxml_xmlCheckUTF8(BAD_CAST(str))) {
-			soap_error1(E_ERROR,  "Encoding: string '%s' is not a valid utf-8 string", str);
 		}
 		xmlBufferFree(out);
 		xmlBufferFree(in);
-	} else if (!php_libxml_xmlCheckUTF8(BAD_CAST(str))) {
-		soap_error1(E_ERROR,  "Encoding: string '%s' is not a valid utf-8 string", str);
+	}
+
+	if (!php_libxml_xmlCheckUTF8(BAD_CAST(str))) {
+		char *err = emalloc(new_len + 8);
+		char c;
+		int i;
+		
+		memcpy(err, str, new_len+1);
+		for (i = 0; (c = err[i++]);) {
+			if ((c & 0x80) == 0) {
+			} else if ((c & 0xe0) == 0xc0) {
+				if ((err[i] & 0xc0) != 0x80) {
+					break;
+				}
+				i++;
+			} else if ((c & 0xf0) == 0xe0) {
+				if ((err[i] & 0xc0) != 0x80 || (err[i+1] & 0xc0) != 0x80) {
+					break;
+				}
+				i += 2;
+			} else if ((c & 0xf8) == 0xf0) {
+				if ((err[i] & 0xc0) != 0x80 || (err[i+1] & 0xc0) != 0x80 || (err[i+2] & 0xc0) != 0x80) {
+					break;
+				}
+				i += 3;
+			} else {
+				break;
+			}
+		}
+		if (c) {
+			err[i-1] = '\\';
+			err[i++] = 'x';
+			err[i++] = ((unsigned char)c >> 4) + ((((unsigned char)c >> 4) > 9) ? ('a' - 10) : '0');
+			err[i++] = (c & 15) + (((c & 15) > 9) ? ('a' - 10) : '0');
+			err[i++] = '.';
+			err[i++] = '.';
+			err[i++] = '.';
+			err[i++] = 0;
+		}
+
+		soap_error1(E_ERROR,  "Encoding: string '%s' is not a valid utf-8 string", err);
 	}
 
 	text = xmlNewTextLen(BAD_CAST(str), new_len);
diff --git a/ext/soap/tests/bugs/bug42488.phpt b/ext/soap/tests/bugs/bug42488.phpt
new file mode 100755
index 0000000000..4069c9f6be
--- /dev/null
+++ b/ext/soap/tests/bugs/bug42488.phpt
@@ -0,0 +1,22 @@
+--TEST--
+Bug #42488 (SoapServer reports an encoding error and the error itself breaks)
+--SKIPIF--
+<?php require_once('skipif.inc'); ?>
+--INI--
+soap.wsdl_cache_enabled=0
+--FILE--
+<?php
+$request = <<<EOF
+<?xml version="1.0" encoding="UTF-8"?>
+<SOAP-ENV:Envelope xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/" xmlns:ns1="test:\" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/" SOAP-ENV:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/"><SOAP-ENV:Body><ns1:getBadUTF/></SOAP-ENV:Body></SOAP-ENV:Envelope>
+EOF;
+$soap = new SoapServer(NULL, array('uri'=>'test://'));
+function getBadUTF(){
+    return "stuff\x93thing";
+}
+$soap->addFunction('getBadUTF');
+$soap->handle($request);
+?>
+--EXPECT--
+<?xml version="1.0" encoding="UTF-8"?>
+<SOAP-ENV:Envelope xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"><SOAP-ENV:Body><SOAP-ENV:Fault><faultcode>SOAP-ENV:Server</faultcode><faultstring>SOAP-ERROR: Encoding: string 'stuff\x93...' is not a valid utf-8 string</faultstring></SOAP-ENV:Fault></SOAP-ENV:Body></SOAP-ENV:Envelope>
\ No newline at end of file