Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cu-dev] leading slash in @WebServlet in TCK tests
Hi,

Thanks, good to hear I all got cleared up :)

Kind regards,
Arjan

On Wed, Mar 23, 2022 at 4:23 PM Kyle Aure <kylejaure@xxxxxxxxx> wrote:
Hello everyone,

Just FYI I did open an issue against the Servlet spec to ask for clarification on whether or not the leading slash is required or options.

The response is, YES!  The reason is that there are different meanings for the following paths "", "/", and ".*"

Thank you,
Kyle Jon Aure

On Tue, Mar 22, 2022 at 11:25 AM Petr Aubrecht <aubrecht@xxxxxxxxxxxx> wrote:

Ready for review:

https://github.com/eclipse-ee4j/concurrency-api/pull/188

Thanks everybody!

Petr


On 3/22/22 16:57, Petr Aubrecht wrote:

Thanks, I will do PR, because I already have this commit :-)

Petr



On 3/22/22 15:30, Kyle Aure wrote:
Hello Arjan,

Thanks for pointing this out.  As I was porting these tests from the Platform TCK to a Standalone TCK many of the @WebServlet URL Paths were declared as constants in separate Constant.java files.
In order to make the TCK more readable, I wanted to remove those constants and just use simple strings for the servlet URL.
I think sometime during that process the leading slash was removed and I didn't notice it because Liberty handles that case (the platform I was using to test).
I can go in and add the leading slash back in if it is necessary per spec, or if it is helpful for other implementors.


Thank you,
Kyle Jon Aure


On Tue, Mar 22, 2022 at 8:30 AM arjan tijms <arjan.tijms@xxxxxxxxx> wrote:
Hi,

Thanks, so I guess Kyle would be the first one to ask about this. I'm cc'ing him here.

I wonder if it might be related to the injected Arquillian URL ending or not ending on a /. I've seen issues with that before. Maybe Kyle made this change to fix the Arquillian connector on Liberty.

Kind regards,
Arjan Tijms


On Tue, Mar 22, 2022 at 1:22 PM Petr Aubrecht <aubrecht@xxxxxxxxxxxx> wrote:

https://github.com/eclipse-ee4j/concurrency-api/commit/11cb74c

package rename and final changes 
update copyright

rename packages again

Use central ContextServiceDefinitionServlet

Some other random example:

https://github.com/eclipse-ee4j/concurrency-api/commit/b75689b

Enable arquillian on transaction tests 
Update spec tests to use Arquillian

Update api tests to use Arquillian

Petr


On 3/22/22 13:12, arjan tijms wrote:
Hi,

Just wondering, what did the commit comment said that removed that leading slash?

On Tuesday, March 22, 2022, Petr Aubrecht <aubrecht@xxxxxxxxxxxx> wrote:

Hi,

I'm looking at the TCK tests for Concurrent and one of the interesting things is removing leading slash in the @WebServlet annotation. Even the existing ones were removed.

E.g.

@WebServlet("/ManagedThreadFactoryDefinitionServlet")

was changed to

@WebServlet("ManagedThreadFactoryDefinitionServlet")


I looked at the spec, but it tells nothing. All examples contain the leading slash.

https://github.com/jakartaee/servlet/blob/master/spec/src/main/asciidoc/servlet-spec-body.adoc#811-webservlet

Can you point me to some place, which explains, why it should work without it? I know, that it is not required for @Path, but it is a different spec.

Thank you

Petr



_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev
_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev

_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev


_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev
_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev
_______________________________________________
cu-dev mailing list
cu-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cu-dev

Back to the top