Re: Crash in var_dump and print_r

At 11:45 AM 8/26/2001 -0400, George Schlossnagle wrote:

Ahh... that makes much more sense. Should there be any protection though from populating a hash with NULL values, or is this solely the responsibility of the extension writer? While there is the zend_has_add_empty_element(), it appears as if there is no protection from adding a NULL pData pointer if you just call zend_hash_add_or_update().

If you're talking about HashTable's which are used in zval's for array values or object properties then there should be no protection. Extension writers need to make sure they don't write broken code the same way they shouldn't put junk pointers in those hash tables. Andi

On Sunday, August 26, 2001, at 11:32 AM, Andi Gutmans wrote:

What Zeev meant is that it should be a valid zval * but it isn't. Andi At 11:31 AM 8/26/2001 -0400, George Schlossnagle wrote:

If NULL's a valid zval*, then both of thise routines need to be fixed to handle null pointers. Here's some patches: --- zend.c.orig Sun Aug 26 11:14:28 2001

+++ zend.c      Sun Aug 26 11:20:22 2001

@@ -199,6 +199,10 @@ ZEND_API void zend_print_zval_r_ex(zend_write_func_t write_func, zval *expr, int indent) {

+       if(expr == NULL) {
+               ZEND_PUTS(" (NULL)");
+               return;
+       }
        switch(expr->type) {
                case IS_ARRAY:
                        ZEND_PUTS("Array\n");

--- var.c.orig Thu Aug 16 11:16:52 2001

+++ var.c       Sun Aug 26 11:24:54 2001

@@ -63,7 +63,10 @@

        if (level>1) {
                php_printf("%*c", level-1, ' ');
        }

-

+       if ((*struc) == NULL) {
+               php_printf(" (NULL)");
+               return;
+       }
        switch ((*struc)->type) {
                case IS_BOOL:
                        php_printf("%sbool(%s)\n", COMMON, Z_LVAL_PP(struc)?"true":"false");

On Sunday, August 26, 2001, at 06:15 AM, Zeev Suraski wrote:

It looks like domxml is responsible here. The hash table that's sent to it contains a NULL value, which should be a valid zval *... Zeev At 13:04 26-08-01, Joey Smith wrote:

This may be due to the way domxml is doing things, but the test script in bug #10936 still creates a crash, and the backtrace points to zend_print_zval_r_ex...see attachment 1. Attachment 2 show that var_dump() gives more or less the same result. -- PHP Development Mailing List <http://www.php.net/> To unsubscribe, e-mail: php-dev-unsubscribe@lists.php.net For additional commands, e-mail: php-dev-help@lists.php.net To contact the list administrators, e-mail: php-list- admin@lists.php.net

-- Zeev Suraski <zeev@zend.com> CTO & co-founder, Zend Technologies Ltd. http://www.zend.com/ -- PHP Development Mailing List <http://www.php.net/> To unsubscribe, e-mail: php-dev-unsubscribe@lists.php.net For additional commands, e-mail: php-dev-help@lists.php.net To contact the list administrators, e-mail: php-list- admin@lists.php.net

-- PHP Development Mailing List <http://www.php.net/> To unsubscribe, e-mail: php-dev-unsubscribe@lists.php.net For additional commands, e-mail: php-dev-help@lists.php.net To contact the list administrators, e-mail: php-list- admin@lists.php.net