[Yanel-dev] import site and introspection

[Josias Thöny]

Michael Wechner wrote:
> Andreas Wuest wrote:
> 
>> Hi Josias
>>
>> On 6.3.2007 1:10 Uhr, Josias Thöny wrote:
>>
>>> Hi all,
>>>
>>> While working on the import site feature I noticed a problem:
>>> After a site has been crawled and the dump has been imported into 
>>> yanel, the pages are not editable because the introspection files 
>>> don't exist.
>>> Is it really necessary to create an introspection file for each 
>>> imported file? Or is there a better solution?
>>
>>
>> Personally, I would all be for dynamically generated introspection 
>> objects, based on some meta-data (e.g. isModifiable, etc., paired with 
>> some sane fall-back values if a specific property is not present).
>>
> me too, but we need to discuss how it's best to accomplish this. For 
> instance there should always be the check if an introspection exists in 
> case to overwrite the generic case (this could be done throught the 
> chain of responsibility).

What about this:
The page /en/about.html contains a link to its introspection file.
If this link is "/en/about.html?yanel.usecase=introspection", a
dynamically generated introspection file will be returned.
If the link is something like "/en/introspection-about.xml", it points
to a static introspection file in the repository.

Could there be any problem with static export because of the
request-parameter?

The other option would be to use a naming convention for the
introspection file, and if the introspection file does not exist, a 
default one will be found by pattern matching.
The drawback of this approach is that it does not allow to use different 
  default introspection files for each resource type. I mean, with url 
pattern matching it's not possible to say "all file-resources should use 
introspection-file.xml and all xml-resources should use 
introspection-xml.xml".

WDYT?

josias

> 
> I think it's best if we add a new item to the roadmap and specify it 
> first ;-)
> 
> Cheers
> 
> Michi
> 
>