[PATCH saem_ref] [pkg] use new import path for cubicweb-seda

Denis Laxalde denis.laxalde at logilab.fr
Fri Feb 24 11:54:37 CET 2017


Philippe Pepiot a écrit :
> On 02/24/2017 11:44 AM, Philippe Pepiot wrote:
>> On 02/24/2017 11:19 AM, Denis Laxalde wrote:
>>> Philippe Pepiot a écrit :
>>>> On 02/24/2017 11:00 AM, Philippe Pepiot wrote:
>>>>> # HG changeset patch
>>>>> # User Philippe Pepiot <philippe.pepiot at logilab.fr>
>>>>> # Date 1487930142 -3600
>>>>> #      Fri Feb 24 10:55:42 2017 +0100
>>>>> # Node ID 7e9aca4d08ef63873707041c5db2af823ebcceda
>>>>> # Parent  c3194697a71904e757bc8153302457325e25c199
>>>>> # Available At https://hg.logilab.org/review/cubes/saem_ref
>>>>> #              hg pull https://hg.logilab.org/review/cubes/saem_ref -r
>>>>> 7e9aca4d08ef
>>>>> # Tested at: https://jenkins.logilab.org/job/cubicweb-saem_ref/100/
>>>>> [pkg] use new import path for cubicweb-seda
>>>>
>>>> One test doesn't pass but I don't understand how it is related to this
>>>> patch, any hints ?
>>>>
>>>> https://jenkins.logilab.org/job/cubicweb-saem_ref/100/TOXENV=py27,distributions=debian_jessie/console
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>>               raise SelectAmbiguity(msg % (winners, args,
>>>> kwargs.keys()))
>>> E               SelectAmbiguity: select ambiguity: [<class
>>> 'cubicweb_seda.views.sedalib.SEDAComponentsBreadcrumbsAdapter'>, <class
>>> 'cubicweb_saem_ref.views.seda.SEDAComponentsBreadcrumbsAdapter'>]
>>> E               (args: (<cubicweb.devtools.fake.FakeRequest object at
>>> 0x7f559c4bbd90>,), kwargs: ['entity'])
>>>
>>> same component registered twice?
>>>
>>> Not related to your changes in this repo, rather due to pulling new
>>> things from cubicweb-seda.
>>>
>>
>> https://hg.logilab.org/master/cubes/seda/file/tip/cubicweb_seda/views/sedalib.py#l33
>>
>>
>>
>> https://hg.logilab.org/review/cubes/saem_ref/file/7e9aca4d08ef/cubicweb_saem_ref/views/seda.py#l55
>>
>>
>>
>> These two adapters have exactly the same __select__ and adapter of seda
>> cube is not unregistered by saem_ref cube. Should top-level cube win in
>> this case (so it maybe an issue in cube ordering/dependency) ?
>
> Wait, this code is new in seda:
> https://hg.logilab.org/review/cubes/seda/rev/36388c9744c8
>
> I guess saem_ref adapter should be dropped or explicitly unregister seda
> one.

yes.



More information about the saem-devel mailing list