[ECODP-dev] Fwd: ckan issue when reusing uri's

John Glover john.glover at okfn.org
Tue Nov 12 11:08:27 UTC 2013


Hi Bert,

I do not think that this is the case. You should be able to update them by
making package_update API calls, is this not working?

Regards,
John


On 12 November 2013 11:59, Bert Van Nuffelen <bert.van.nuffelen at tenforce.com
> wrote:

> Hi John,
>
> The problem is that there seems to be a different in behaviour which
> ultimately leads to an unable to edit situation.
>
> package 1: for publisher P1, dataset name N1 and url *URL1*
> package 2: for publisher P2, dataset name N2 and url *URL1*
>
> you can create package 1 successfully, then you can create package 2
> successfully, but then you *cannot* update any of them because of the
> validation rule.
>
> So 2 'valid' posts leads to a 'kind of deadlock' situation.
>
> kind regards,
>
> Bert
>
>
>
> 2013/11/12 John Glover <john.glover at okfn.org>
>
>> Hi Bert,
>>
>> I'm not sure that I understand: why would you expect the package_create
>> validation to fail the first time, assuming that the two names do not
>> already exist?
>>
>> If you instead want to change (edit) an existing dataset, you must make a
>> package_update call (which should not fail if the name exists, the opposite
>> in fact).
>>
>> If you want to rerun a package_create call, you must delete and purge the
>> datasets first.
>>
>> Regards,
>> John
>>
>>
>> On 12 November 2013 10:56, Bert Van Nuffelen <
>> bert.van.nuffelen at tenforce.com> wrote:
>>
>>> Hi John and Darwin,
>>>
>>> We run our rdf2ckan tests again and we spotted the next situation. Can
>>> you verify if our analysis is correct and if the validation rule in the
>>> create is not active as we suspect.
>>>
>>> Hereunder is the problem description:
>>>
>>> When running rdf2ckan tests, I found an issue in ckan: I have two
>>> packages, try-hijack1.zip and try-hijack2.zip that are uploaded to ckan.
>>> Both packages have the same uri field (called url in the ckan json). They
>>> are different packages (with a different name in JSON) and they also have
>>> different publishers.
>>>
>>> I run the tests two times. The first time, the packages are created. The
>>> second time, the packages both return a CKAN validation error, saying that
>>> the URL is already in use.
>>>
>>> This means I can no longer edit the packages now. In fact, the error in
>>> the validation is expected, but it should already fire the first time I run
>>> the packages. As you said, the problem is probably that the ckan validation
>>> only occurs during the update procedure and not during the create procedure.
>>>
>>> I have included the reports from the two runs and the jsons for the
>>> packages with this mail.
>>>
>>> kind regards,
>>>
>>> Bert
>>>
>>>
>>> --
>>> Bert Van Nuffelen
>>>
>>> Semantic Technologies Software Architect at TenForce
>>> www.tenforce.be
>>>
>>> Bert.Van.Nuffelen at tenforce.com
>>> Office: +32 (0)16 31 48 60
>>> Mobile:+32 479 06 24 26
>>> skype: bert.van.nuffelen
>>>
>>> _______________________________________________
>>> Ecodp-dev mailing list
>>> Ecodp-dev at lists.okfn.org
>>> http://lists.okfn.org/mailman/listinfo/ecodp-dev
>>>
>>>
>>
>> _______________________________________________
>> Ecodp-dev mailing list
>> Ecodp-dev at lists.okfn.org
>> http://lists.okfn.org/mailman/listinfo/ecodp-dev
>>
>>
>
>
> --
> Bert Van Nuffelen
>
> Semantic Technologies Software Architect at TenForce
> www.tenforce.be
>
> Bert.Van.Nuffelen at tenforce.com
> Office: +32 (0)16 31 48 60
> Mobile:+32 479 06 24 26
> skype: bert.van.nuffelen
>
> _______________________________________________
> Ecodp-dev mailing list
> Ecodp-dev at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/ecodp-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.okfn.org/mailman/private/ecodp-dev/attachments/20131112/1b79f2ad/attachment.html>


More information about the ecodp-dev mailing list