Additionally ISO8601 date are not supported

When sending:

   "timestamp":"2012-09-06T13:40:34.874+02:00"

I get an "Unparseable date" exception. The following code was used to serialize the date:

    DateTime dateTime = new DateTime( new Date() );
    DateTimeFormatter dateTimeFormatter = ISODateTimeFormat.dateTime();
    String dateTimeAsString = dateTimeFormatter.print( dateTime );

What string based date format should be used?

Regards,

Minto


Quoting [email protected]:

Result of my debug session:

1) Missing output details

Currently the on the client side debug log shows the following server error:

    "HTTP/1.1 500 Internal Server Error[\r][\n]"
    "HTTP/1.1 500 Internal Server Error[\r][\n]"
"Content-Type: application/json;profile=urn:org.restfulobjects/error[\r][\n]"
    "Content-Length: 8824[\r][\n]"
    "Server: Jetty(6.1.26)[\r][\n]"

Details are missing here. Server side I have an exception saying 'Unparseable date: "2012-09-05 09:30"'. The exception message is swallowed since it does not appear in either client or server logging.

The same happened with other messages

2) Long values < maxint not detected as longs

In my domain model a field is defined as long. When entering this field as:

    arguments.mapPut("melderVolgnummer", 1);

It is detected server side as Integer instead of Long. Forcefully adding 1L or JsonRepresentation( new LongNode(1)) makes no difference. Server side it appeared Jackson sees this field as IntNode. As a result isLong() (on both JsonRepresentation and IntNode) returns false. This results in the following exception.

   IllegalArgumentException(formatExMsg(path, "is not a long"))

Attached patch solves this issue.

Regards,

Minto


Quoting [email protected]:

Hi Folks,

I managed to get something (using RO-applib) partly working. My current implementation is based on test code from the TCK.

Retrieving some information works fine:

 1) "GET /services/ HTTP/1.1[\r][\n]" (works OK)
 2) "GET /services/servicename? ...." (works OK)
 3) "POST /services/servicename/...." (Fails :( )

In step 3 I wanted to create a new domain object. Unfortunately the server responded with a 500 Internal Server Error. Initially I thought this was due to ISIS-265. But after applying the patch the htmlviewer works fine but the RO-viewer still gives the same response.

Looking in the server log I can't find any information about why POST request actually failed. Next step for me is to start debug the RO-viewer.

If anyone has a clue what is going on here, I am all ears.

Regards,

Minto


Quoting [email protected]:

Oh boy, this look complex.

The code in RO-tck looks like some sort of meta programming. I am looking for a piece of code to create a new domain object using RO-applib. But I have a hard time digesting what's there.

Can someone give me some more hints (hopefully a snippet)

Regards,

Minto


Quoting Dan Haywood <[email protected]>:

On 1 September 2012 10:29, Minto van der sluis <[email protected]> wrote:


Should stick with the current release
0.2.0 or move ahead. How stable is the current trunk version?


It's pretty stable; starting to move towards getting a release out in the
next month or two.

And, what's new in trunk is the JDO object store, which may (perhaps)
replace the SQL objectstore, at least in the short-term.

Also, the v0.2.0 release holds an incomplete cut of the RO viewer (whatever
was implemented back in Feb).

~~~
If you do decide to track trunk, you might find it easiest to fork my
github copy of Isis [1]




I will have a closer look at the latest/trunk ro-applib. I guess
RestfulRestfulClient is entry point to start from.


Yes, org.apache.isis.viewer.restfulobjects.applib.RestfulClient.

This lets you use either a HATEOAS or a templated URL approach.




But to be able to use
it I probably have to switch to using the trunk version.


Not sure; I think you'll find that 0.2.0 does have something.  Even so, I
do recommend that you switch to trunk



Speaking slightly selfishly, I'd love you to have a go with using the RO
applib; it'll help us determine where the gaps and annoyances are for
"real-life" use.

I could give it a try If I knew were to start. Is the 0.2.0 json viewer
documentation still a good starting point? Since I am lazy ;-) I wonder
if their exists some sample application already. This could give me a
headstart.


The tests aren't a bad place to look; you'll find these in
restfulobjects-tck.

Admittedly, these have been knocked-about a bit and so quite a few are
currently are @Ignore'd.  However,
the org.apache.isis.viewer.restfulobjects.tck.resources.home.HomePageResourceTest_accept
runs and passes.  (It also uses the isisWebServerRule which bootstraps the
web.xml within an integration test).

Hope that helps some

Dan

[1] https://github.com/danhaywood/apache-isis









Reply via email to