You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cocoon.apache.org by Jon Evans <jo...@misgl.com> on 2003/11/21 11:48:59 UTC
Possible BUG? in raw-request-param, used with cinclude
Hi,
My colleague Simon Hutchinson posted a problem to the list earlier this
week (Subject: URL Encoding (Coccon 2.1 cvs as of about 2 weeks ago)).
Tony Collen pointed him to the raw-request-param input module.
We have found what we believe to be a bug, when using raw-request-param
to get the value of a param passed via a cinclude.
Basically, the raw-request-param input module seems to work OK, unless
the param got passed via a cinclude, in which case it returns null.
We have built a test case for it, which you can download from
http://www.merlindev.co.uk/test.zip (1595 bytes).
Archive: test.zip
Length Date Time Name
-------- ---- ---- ----
0 11-21-03 10:30 test/
46 11-21-03 10:11 test/testrequest.xml
1290 11-21-03 10:22 test/sitemap.xmap
617 11-21-03 10:23 test/testrequest.xsl
156 11-21-03 10:19 test/testcinclude.xml
-------- -------
2109 5 files
Instructions:
unzip this to a sitemap which mounts sub-sitemaps (e.g. the top
directory of a deployed cocoon)
browse to http://localhost:8080/cocoon/test/testrequest?test=foo+bar
you should get some output showing the value of "test", from both
request-param and raw-request-param.
Now browse to http://localhost:8080/cocoon/test/testrequestviacinclude
This matcher starts with an xml file containing a cinclude request for
cocoon:/testrequest?test=foo+bar+baz . This time, the output page will
indicate:
Value of param "test" from request input module: "foo bar baz"
Value of param "test" from raw-request input module: ""
Hopefully there is enough information there for someone to either
confirm that it is a bug, or point out my mistake. :-)
We're using 2.1 CVS HEAD as of yesterday.
On another note, apart from this and a few other minor issues, the
website we are moving to Cocoon is really coming along well. Hopefully
I'll be able to publish a URL to the test site (I'll have to check first).
Thanks,
Jon
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
Re: Possible BUG? in raw-request-param, used with cinclude
Posted by Geoff Howard <co...@leverageweb.com>.
Jon Evans wrote:
> Geoff Howard wrote:
>
>> Post this summary to bugzilla, and upload the test case to the bug as an
>> attachment.
>
> Good plan. :-) Just done it.
Thanks. By the way, a good analysis and a test case as you have done
are _really_ appreciated.
Geoff
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
Re: Possible BUG? in raw-request-param, used with cinclude
Posted by Jon Evans <jo...@misgl.com>.
Geoff Howard wrote:
> Post this summary to bugzilla, and upload the test case to the bug as an
> attachment.
Good plan. :-) Just done it.
Cheers,
Jon
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
Re: Possible BUG? in raw-request-param, used with cinclude
Posted by Geoff Howard <co...@leverageweb.com>.
Post this summary to bugzilla, and upload the test case to the bug as an
attachment.
Geoff
Jon Evans wrote:
> Hi,
>
> My colleague Simon Hutchinson posted a problem to the list earlier this
> week (Subject: URL Encoding (Coccon 2.1 cvs as of about 2 weeks ago)).
>
> Tony Collen pointed him to the raw-request-param input module.
>
> We have found what we believe to be a bug, when using raw-request-param
> to get the value of a param passed via a cinclude.
> Basically, the raw-request-param input module seems to work OK, unless
> the param got passed via a cinclude, in which case it returns null.
>
> We have built a test case for it, which you can download from
> http://www.merlindev.co.uk/test.zip (1595 bytes).
>
> Archive: test.zip
> Length Date Time Name
> -------- ---- ---- ----
> 0 11-21-03 10:30 test/
> 46 11-21-03 10:11 test/testrequest.xml
> 1290 11-21-03 10:22 test/sitemap.xmap
> 617 11-21-03 10:23 test/testrequest.xsl
> 156 11-21-03 10:19 test/testcinclude.xml
> -------- -------
> 2109 5 files
>
> Instructions:
> unzip this to a sitemap which mounts sub-sitemaps (e.g. the top
> directory of a deployed cocoon)
> browse to http://localhost:8080/cocoon/test/testrequest?test=foo+bar
>
> you should get some output showing the value of "test", from both
> request-param and raw-request-param.
>
> Now browse to http://localhost:8080/cocoon/test/testrequestviacinclude
> This matcher starts with an xml file containing a cinclude request for
> cocoon:/testrequest?test=foo+bar+baz . This time, the output page will
> indicate:
>
> Value of param "test" from request input module: "foo bar baz"
>
> Value of param "test" from raw-request input module: ""
>
> Hopefully there is enough information there for someone to either
> confirm that it is a bug, or point out my mistake. :-)
>
> We're using 2.1 CVS HEAD as of yesterday.
>
> On another note, apart from this and a few other minor issues, the
> website we are moving to Cocoon is really coming along well. Hopefully
> I'll be able to publish a URL to the test site (I'll have to check first).
>
> Thanks,
>
> Jon
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
> For additional commands, e-mail: users-help@cocoon.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org