Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:32817
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain oracle.com from 148.87.113.118 cause and error)
Message-ID: <47152024.4010904@oracle.com>
Date: Tue, 16 Oct 2007 13:33:40 -0700
User-Agent: Thunderbird 1.5.0.13 (X11/20070824)
MIME-Version: 1.0
To: internals@lists.php.net
Content-Type: multipart/mixed;
 boundary="------------010402090801080002040602"
Subject: Using OnUpdateUTF8String in PHP 6
From: christopher.jones@oracle.com (Christopher Jones)

--------------010402090801080002040602
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit


With thanks to Sara we looked at OnUpdateUTF8String to access a
php.ini value in OCI8 in PHP 6.

One of our engineers sent me a proposed patch for zend_ini.c in PHP6
to allow OnUpdateUTF8String to work as he thought it should.  Any
comments?

Chris

------------

 > The problems I faced when unicode.semantics=On were:
 >
 > 1) Any OnUpdateUTF8String php.ini variable is seen as NULL in
 > oci_globals.
 >
 > 2) If the above is resolved, the memory of the variable is wiped out,
 > by the time we come to execution of the script, after module inits.
 >
 > 3) We still convert to UTF-16 when unicode.semantics=Off.

PS. Patch is attached - if it gets through.

-- 
Christopher Jones, Oracle
Email: christopher.jones@oracle.com    Tel:  +1 650 506 8630
Blog:  http://blogs.oracle.com/opal/   Free PHP Book: http://tinyurl.com/f8jad

--------------010402090801080002040602
Content-Type: text/plain;
 name="zend_ini.c.diff.txt"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="zend_ini.c.diff.txt"

--- zend_ini.c	2007-10-02 11:07:32.000000000 -0700
+++ zend_ini.c.new	2007-10-16 13:20:03.000000000 -0700
@@ -640,7 +640,8 @@
 
 ZEND_API ZEND_INI_MH(OnUpdateUTF8String) /* {{{ */
 {
-	UChar **p;
+	UChar **up;
+	char **p;
 	UChar *ustr = NULL;
 	int32_t ustr_len, capacity;
 	UErrorCode status = U_ZERO_ERROR;
@@ -651,30 +652,37 @@
 
 	base = (char *) ts_resource(*((int *) mh_arg2));
 #endif
+	/* Convert only if unicode semantics is on. Otherwise, same as OnUpdateString */
+	if (UG(unicode)){
+		/* estimate capacity */
+		capacity = (new_value_length > 2) ? ((new_value_length >> 1) + (new_value_length >> 3) + 2) : new_value_length;
+
+		while (1) {
+			ustr = peurealloc(ustr, capacity+1, 1);
+			u_strFromUTF8(ustr, capacity+1, &ustr_len, new_value, new_value_length, &status);
+			if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
+				capacity = ustr_len;
+				status = U_ZERO_ERROR;
+			} else {
+				break;
+			}
+		}
 
-	/* estimate capacity */
-	capacity = (new_value_length > 2) ? ((new_value_length >> 1) + (new_value_length >> 3) + 2) : new_value_length;
-
-	while (1) {
-		ustr = eurealloc(ustr, capacity+1);
-		u_strFromUTF8(ustr, capacity, &ustr_len, new_value, new_value_length, &status);
-		if (status == U_BUFFER_OVERFLOW_ERROR) {
-			capacity = ustr_len;
-			status = U_ZERO_ERROR;
-		} else {
-			break;
+		if (U_FAILURE(status)) {
+			zend_error(E_WARNING, "Could not convert UTF-8 INI value to Unicode");
+			efree(ustr);
+			return FAILURE;
 		}
-	}
 
-	if (U_FAILURE(status)) {
-		zend_error(E_WARNING, "Could not convert UTF-8 INI value to Unicode");
-		efree(ustr);
-		return FAILURE;
-	}
+		up = (UChar **) (base+(size_t) mh_arg1);
 
-	p = (UChar **) (base+(size_t) mh_arg1);
+		*up = ustr;
+	}
+	else {		/* Same as OnUpdateString */
+		p = (char **) (base+(size_t) mh_arg1);
 
-	*p = ustr;
+		*p = new_value;
+	}
 	return SUCCESS;
 }
 /* }}} */

--------------010402090801080002040602--