axis-c-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Leland" <st...@pouchapond.com>
Subject memory #2 in generated code
Date Sun, 27 Jun 2010 18:01:32 GMT
Hi,

I am running on Windows Server 2008 R2, and using a build I made yesterday
(svn 957643) I generated code from our wsdl and ran it - it has a leak that
was also present in the 1.6.0 release.

Given this in the wsdl file:

      <s:complexType name="ResponseMessageType">
        <s:sequence>
          <s:element minOccurs="0" maxOccurs="1" name="Attendee" 
type="tns:EmailAddress" />
          <s:element minOccurs="0" maxOccurs="1" name="MessageText" 
type="s:string" />
          <s:element minOccurs="0" maxOccurs="1" name="ResponseCode" 
type="tns:ResponseCodeType" />
          <s:element minOccurs="0" maxOccurs="1" name="DescriptiveLinkKey" 
type="s:int" />
          <s:element minOccurs="0" maxOccurs="1" name="MessageXml">
            <s:complexType>
              <s:sequence>
                <s:any minOccurs="0" maxOccurs="unbounded" />
              </s:sequence>
            </s:complexType>
          </s:element>
        </s:sequence>
        <s:attribute name="ResponseClass" type="tns:ResponseClassType" 
use="required" />
      </s:complexType>

...

      <s:simpleType name="ResponseClassType">
        <s:restriction base="s:string">
          <s:enumeration value="Success" />
          <s:enumeration value="Warning" />
          <s:enumeration value="Error" />
        </s:restriction>
      </s:simpleType>



 the following code is generated twice - after the comments say it is 
building MessageText, the value it is comparing to 'key' is 
"nil|http://www.w3.org/2001/XMLSchema-instance" and  after the comments say 
it is building MessageXml, the value it is comparing to 'key' is 
"ResponseClass", like this:

                  attrib_text = NULL;
                  if(attribute_hash)
                  {
                       axutil_hash_index_t *hi;
                       void *val;
                       const void *key;

                       for (hi = axutil_hash_first(attribute_hash, env); hi; 
hi = axutil_hash_next(env, hi))
                       {
                           axutil_hash_this(hi, &key, NULL, &val);


                               if(!strcmp((axis2_char_t*)key, 
"ResponseClass"))

                               {
                                   parent_attri = (axiom_attribute_t*)val;
                                   break;
                               }
                       }
                  }


The leak can be fixed by making the body of the for loop look like this:

                  attrib_text = NULL;
                  if(attribute_hash)
                  {
                       axutil_hash_index_t *hi;
                       void *val;
                       const void *key;

                       for (hi = axutil_hash_first(attribute_hash, env); hi; 
hi = axutil_hash_next(env, hi))
                       {
                           axutil_hash_this(hi, &key, NULL, &val);


                               if(!strcmp((axis2_char_t*)key, 
"ResponseClass"))

                               {
                                   parent_attri = (axiom_attribute_t*)val;
                                   AXIS2_FREE(env->allocator, hi); 
// hi normally deleted when axutil_hash_next comes to end
                                   break;
                               }

                               if (val)
                              {
                                   AXIS2_FREE(env->allocator, val);
                              }
                       }
                  }

Regards,

Steve



---------------------------------------------------------------------
To unsubscribe, e-mail: c-dev-unsubscribe@axis.apache.org
For additional commands, e-mail: c-dev-help@axis.apache.org


Mime
View raw message