You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cocoon.apache.org by Grzegorz Kossakowski <gr...@tuffmail.com> on 2007/08/28 20:53:42 UTC
True OO for servlet service
Hello,
I don't have that much time to spend these days on Cocoon but I would really like to see true Object
Oriented approach implemented in upcoming[1] release of cocoon-servlet-service. The main task is
tracked by COCOON-2038[2] where Rice Yeh has uploaded his patch implementing true OO behaviour in
servlet-service-fw. According to Rice, this patch contains a bug-fix as well (details in the comments).
After reviewing Rice's patch I decided that we really need test coverage for this quite complicated
functionality for both ServletServiceContext class usage clarity and casual hunting for bugs. Daniel
pointed me to another bug report (COCOON-1939[3]) that involves OO aspects of servlet-service-fw.
That makes point about test coverage even more strong.
What I would like to ask you for in this e-mail is to add additional tests to
ServletServiceContextTestCase class that exhibit problems mentioned in both issues. As soon as I
have working test cases I guess that fixing problems would be quite easy. Since it was you who
discovered problems it should be quite easy to write tests for them, too.
Alexander and Rice, can I count on your help with this before next release in late September is cut?
[1] http://article.gmane.org/gmane.text.xml.cocoon.devel/74999
[2] https://issues.apache.org/jira/browse/COCOON-2038
[3] https://issues.apache.org/jira/browse/COCOON-1939
--
Grzegorz Kossakowski
http://reflectingonthevicissitudes.wordpress.com/
Re: True OO for servlet service
Posted by Grzegorz Kossakowski <gk...@apache.org>.
Rice Yeh pisze:
> Hi Grek,
Hi Rice :)
> Attached is the patch that is based on the present
> cocoon-servlet-service-impl implementation. cocoon-servlet-service-impl
> has some change since I posted my patch on COCOON-2038, which has
> already too many files attached to it. In order not to confuse you, I
> have the the patch attached to this mail. So you do not need to patch
> the files in COCOON-2038.
There is a legal problem with such approach. When attaching patch to a JIRA issue you check the
option that you intend this file for inclusion to the source codes licensed by Apache License. This
agreement is a must in order to commit your patch.
I already cleaned up COCOON-2038 so you can safely attach your patch as
cocoon-servlet-service-impl-with-tests.patch.
> I find you already have written a test case
> class ServletServiceContextTestCase for ServletServiceContext.
Yes and I wanted to know if testContextInServletCalledFromExplicitSuperCall covers the bug we were
talking about in COCOON-2038 as I'm not sure.
> I add one
> more test case in it to test COCOON-1939. All the test cases in
> ServletServiceContextTestCase are passed.
> So after applying this patch, you might close COCOON-2038 by referring
> back to this mail if there are no further problems.
Great, thanks Rice for your work! As long as all doubts are dispelled I'll commit your patch.
--
Grzegorz Kossakowski
http://reflectingonthevicissitudes.wordpress.com/
Re: True OO for servlet service
Posted by Grzegorz Kossakowski <gk...@apache.org>.
Rice Yeh pisze:
> Hi Grek,
> Attached is the patch that is based on the present
> cocoon-servlet-service-impl implementation. cocoon-servlet-service-impl
> has some change since I posted my patch on COCOON-2038, which has
> already too many files attached to it. In order not to confuse you, I
> have the the patch attached to this mail. So you do not need to patch
> the files in COCOON-2038. I find you already have written a test case
> class ServletServiceContextTestCase for ServletServiceContext. I add one
> more test case in it to test COCOON-1939. All the test cases in
> ServletServiceContextTestCase are passed.
> So after applying this patch, you might close COCOON-2038 by referring
> back to this mail if there are no further problems.
I closed both issues and applied your patch.
Many thanks for your invaluable work!
--
Grzegorz Kossakowski
http://reflectingonthevicissitudes.wordpress.com/
Re: True OO for servlet service
Posted by Rice Yeh <ri...@gmail.com>.
Hi Grek,
Attached is the patch that is based on the present
cocoon-servlet-service-impl implementation. cocoon-servlet-service-impl has
some change since I posted my patch on COCOON-2038, which has already too
many files attached to it. In order not to confuse you, I have the the patch
attached to this mail. So you do not need to patch the files in COCOON-2038.
I find you already have written a test case class
ServletServiceContextTestCase for ServletServiceContext. I add one more test
case in it to test COCOON-1939. All the test cases in
ServletServiceContextTestCase are passed.
So after applying this patch, you might close COCOON-2038 by referring back
to this mail if there are no further problems.
Regards,
Rice
On 8/29/07, Grzegorz Kossakowski <gr...@tuffmail.com> wrote:
>
> Hello,
>
> I don't have that much time to spend these days on Cocoon but I would
> really like to see true Object
> Oriented approach implemented in upcoming[1] release of
> cocoon-servlet-service. The main task is
> tracked by COCOON-2038[2] where Rice Yeh has uploaded his patch
> implementing true OO behaviour in
> servlet-service-fw. According to Rice, this patch contains a bug-fix as
> well (details in the comments).
>
> After reviewing Rice's patch I decided that we really need test coverage
> for this quite complicated
> functionality for both ServletServiceContext class usage clarity and
> casual hunting for bugs. Daniel
> pointed me to another bug report (COCOON-1939[3]) that involves OO aspects
> of servlet-service-fw.
> That makes point about test coverage even more strong.
>
> What I would like to ask you for in this e-mail is to add additional tests
> to
> ServletServiceContextTestCase class that exhibit problems mentioned in
> both issues. As soon as I
> have working test cases I guess that fixing problems would be quite easy.
> Since it was you who
> discovered problems it should be quite easy to write tests for them, too.
>
> Alexander and Rice, can I count on your help with this before next release
> in late September is cut?
>
> [1] http://article.gmane.org/gmane.text.xml.cocoon.devel/74999
> [2] https://issues.apache.org/jira/browse/COCOON-2038
> [3] https://issues.apache.org/jira/browse/COCOON-1939
>
> --
> Grzegorz Kossakowski
> http://reflectingonthevicissitudes.wordpress.com/
>
Re: True OO for servlet service
Posted by Grzegorz Kossakowski <gk...@apache.org>.
Rice Yeh pisze:
> Hi Greg,
Hi Rice, to not confuse names too much it's better Grzegorz or Grek even Greg is English equivalent
for polish Grzegorz :-)
> First off, I have to say sorry for not fulfill my promise in
> cocoon-2038 in time just because I just changed my work. I will try to
> use the 2 days of this coming weekend to help write test cases for my
> patch. But I cannot promise I will complete it in the 2 days.
>
> I just have a svn update. It seems there are a lot of change on cocoon.
> I have to solve the conflicts first.
Thanks Rice, I'm really looking forward your tests!
--
Grzegorz Kossakowski
http://reflectingonthevicissitudes.wordpress.com/
Re: True OO for servlet service
Posted by Rice Yeh <ri...@gmail.com>.
Hi Greg,
First off, I have to say sorry for not fulfill my promise in cocoon-2038
in time just because I just changed my work. I will try to use the 2 days of
this coming weekend to help write test cases for my patch. But I cannot
promise I will complete it in the 2 days.
I just have a svn update. It seems there are a lot of change on cocoon. I
have to solve the conflicts first.
Rice
Rice
On 8/29/07, Grzegorz Kossakowski <gr...@tuffmail.com> wrote:
>
> Hello,
>
> I don't have that much time to spend these days on Cocoon but I would
> really like to see true Object
> Oriented approach implemented in upcoming[1] release of
> cocoon-servlet-service. The main task is
> tracked by COCOON-2038[2] where Rice Yeh has uploaded his patch
> implementing true OO behaviour in
> servlet-service-fw. According to Rice, this patch contains a bug-fix as
> well (details in the comments).
>
> After reviewing Rice's patch I decided that we really need test coverage
> for this quite complicated
> functionality for both ServletServiceContext class usage clarity and
> casual hunting for bugs. Daniel
> pointed me to another bug report (COCOON-1939[3]) that involves OO aspects
> of servlet-service-fw.
> That makes point about test coverage even more strong.
>
> What I would like to ask you for in this e-mail is to add additional tests
> to
> ServletServiceContextTestCase class that exhibit problems mentioned in
> both issues. As soon as I
> have working test cases I guess that fixing problems would be quite easy.
> Since it was you who
> discovered problems it should be quite easy to write tests for them, too.
>
> Alexander and Rice, can I count on your help with this before next release
> in late September is cut?
>
> [1] http://article.gmane.org/gmane.text.xml.cocoon.devel/74999
> [2] https://issues.apache.org/jira/browse/COCOON-2038
> [3] https://issues.apache.org/jira/browse/COCOON-1939
>
> --
> Grzegorz Kossakowski
> http://reflectingonthevicissitudes.wordpress.com/
>