J. Wolfgang Kaltz wrote:

Andreas Hartmann schrieb:

Michael Wechner wrote:

[...]

I think you can still store the documents in a structured way, but each document has a unique ID which is mapped onto the path within the hierarchy/structure, e.g. by hashmap. This unique ID (uid) is then being used within the Scheduler or for references
for instance and not the path itself.


IMO this is mostly a usability problem at document creation time. Right now when the user creates a new document in the default publication, she has to guess a document id which does not exist in this branch. This is not ideal but still acceptable, as typically there will only be a few documents within that branch. But if we want the user to provide a publication-wide unique id at creation, Lenya will need to propose such an id in the creation mask (a sort of auto-increment).


The UID is given automatically by the repository layer, whereas the user
can still choose the name or path arbitrarily.


So IMO we need something like Publication.getNewUniqueDocumentId() before changing the current id structure, otherwise we will have a big usability issue.

But I am not sure where the last used id would be stored ...


the repository layer does all this handling and Lenya would have nothing to do with it.
The user normally won't see any UIDs

it would clearly be too inefficient to look at the current documents and their ids. Maybe we need to store some new structure, like a "publication state node" ? The sitetree doesn't feel right for this.


the sitetree is just a "view" of the various documents.

Michi



--
Wolfgang

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




--
Michael Wechner
Wyona      -   Open Source Content Management   -    Apache Lenya
http://www.wyona.com                      http://lenya.apache.org
[EMAIL PROTECTED]                        [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to