Hi,
I'm sorry to post two pain in the neck items the same day but I want to raise some concerns about the wisdom of moving JTableContent to the legacy folder as was done recently.
I know from the UCM thread that eventually the idea is that JDatabaseObject will replace the JTable package, but UCM is not even a pull request yet and as far as I know if the pull request were sent today it would not include removal of the JTable package or JTableContent. So as of now I would think that many platform applications for web applications that have content ... would probably be using the content table. Further, if and when applications are encouraged to use JContent (assuming it becomes a pull request and that it is accepted) it would seem to me to be highly preferable that those applications be using the old JTableContent which, according to everything I have heard, was the model for setting up the core content table in JContent. Yes any random table can be declared a content type in JContent and just structurally have many blank fields, but the truth is it is strongly preferable for them to be tracking assets, to have created_by and date and checked out information and so on and to be actually using these fields (because like it or not JContent is going to use them).
So, to make it short and sweet: A web application pretty often will want to render data from a database and it seems to me that a platform for web applications would want to provide an api for that.
That said I would like to see JTableContent remove its dependencies on certain CMS concepts such as the existance of something called com_content, something called article, and of a table called #__categories. We could even make the name of the table override-able, which would be forward looking anyway. I think just as has been done elsewhere there are ways to do so without breaking existing applications.
Thanks for your patience,
Elin