You are viewing a plain text version of this content. The canonical link for it is here.
Posted to slide-dev@jakarta.apache.org by Martin Holz <ho...@fiz-chemie.de> on 2003/11/26 15:57:12 UTC

Testsuite problems with Content-Type

Hello,

several Testcases report a wrong Content-Type header,
because the charset is not handled. One example 
would be xmltestcases/functional/XML/multiplePutGets.xml.

<exceuteStep>
   <method>GET</method>
   <url>/slide/files/lock.xml</url>
   <headerErrors>
	<nonMatchingHeader name="Content-Type">
	   <expectedValue>text/xml</expectedValue>
	   <receivedValue>text/xml;charset=ISO-8859-1</receivedValue>
	</nonMatchingHeader>
   </headerErrors>
   <result>ERROR</result>
   <time>111</time>
</exceuteStep>

The response seems to be legal. Is it possible to change the test case,
so it accepts this response with and without charset?

Martin


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


Re: Testsuite problems with Content-Type

Posted by Michael Smith <ms...@speedlegal.com>.
Martin Holz wrote:
> Michael Smith <ms...@speedlegal.com> writes:
> 
> 
>>Though the charset is legal (the protocol permits it), it's incorrect
>>in this case, I think - so the test shouldn't be changed.
>>
>>If you're using a recent (latest?) version of tomcat 4.1.x, I think
>>it's a newly introduced bug (judging from various reports on
>>tomcat-dev). I believe it's been fixed in cvs, but not yet in a
>>release. If you're NOT using tomcat, then I don't know what the cause
>>of this is.
> 
>  
> Yes, I am using Tomcat 4.1.29. Reading the reports from tomcat-dev, 
> i am even more confused. The charset parameter is unnecessary, because
> ISO-8859-1 is the default. It may also confuse some clients.
> But why is it illegal?
> 

In this case (because it's a text/* mime-type) it's not illegal. 
However, it IS incorrect - it's stating that the content is encoded 
using ISO-8859-1. It's not - I'm pretty sure it's UTF-8.

The best solution would be for slide to set the content type to include 
the _correct_ charset parameter. Leaving it out entirely is also 
perfectly acceptable.

Mike


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


Re: Testsuite problems with Content-Type

Posted by Martin Holz <ho...@fiz-chemie.de>.
Michael Smith <ms...@speedlegal.com> writes:

> Though the charset is legal (the protocol permits it), it's incorrect
> in this case, I think - so the test shouldn't be changed.
> 
> If you're using a recent (latest?) version of tomcat 4.1.x, I think
> it's a newly introduced bug (judging from various reports on
> tomcat-dev). I believe it's been fixed in cvs, but not yet in a
> release. If you're NOT using tomcat, then I don't know what the cause
> of this is.
 
Yes, I am using Tomcat 4.1.29. Reading the reports from tomcat-dev, 
i am even more confused. The charset parameter is unnecessary, because
ISO-8859-1 is the default. It may also confuse some clients.
But why is it illegal?

Martin


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


Re: Testsuite problems with Content-Type

Posted by Michael Smith <ms...@speedlegal.com>.
Martin Holz wrote:
> Hello,
> 
> several Testcases report a wrong Content-Type header,
> because the charset is not handled. One example 
> would be xmltestcases/functional/XML/multiplePutGets.xml.
> 
> <exceuteStep>
>    <method>GET</method>
>    <url>/slide/files/lock.xml</url>
>    <headerErrors>
> 	<nonMatchingHeader name="Content-Type">
> 	   <expectedValue>text/xml</expectedValue>
> 	   <receivedValue>text/xml;charset=ISO-8859-1</receivedValue>
> 	</nonMatchingHeader>
>    </headerErrors>
>    <result>ERROR</result>
>    <time>111</time>
> </exceuteStep>
> 
> The response seems to be legal. Is it possible to change the test case,
> so it accepts this response with and without charset?
> 
> Martin
> 

Martin,

Though the charset is legal (the protocol permits it), it's incorrect in 
this case, I think - so the test shouldn't be changed.

If you're using a recent (latest?) version of tomcat 4.1.x, I think it's 
a newly introduced bug (judging from various reports on tomcat-dev). I 
believe it's been fixed in cvs, but not yet in a release. If you're NOT 
using tomcat, then I don't know what the cause of this is.

Mike


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