Hi Jim

Yes, I think there is alot of missing information about advantages/disadvantages etc. I have spent quite a bit of time just building classes to test the cursoradpater to find out how it works because I cannot find the information.

Regarding the idea of why you might want to add cursoradpters to containers I have the following scenario. I created all my interface classes based on containers. I can then reuse them or combine them on a form at run time. I built into them a render() method that renders the interface in DHTML so I can use these container based interface objects with Web-Connect which means I can run my application from a browser or as a regular Windows app. I do not have to design web interfaces because I simply create the interface object and ask it to render itself in DHTML and send the results back to the browser. These interface objects look exactly the same in the browser or on the desktop. To do this I need to add the cursoradapter class to the container based interface object not to the dataenvironment of a form because I could avoid using forms completely in the web version.

Simon


>I have found an extraordinary amount of new documentation in the VFP8 Help concerning the CursorAdapter Class, and my first reading suggests that it is very thorough. And that's a good thing!
>
>However, I have yet to locate anything that describes the basic objectives of the feature or provides a general overview of it's design and operation.
>The Help item titled "Remote Data Connectivity Using CursorAdapter Class" lists a bunch of capabilities of the new class, but leaves me wondering on many factors. For instance:
>"Use either the data source of the CursorAdapter object or the data environment.":
>1. What is the difference between the two options? (advantages/problems of either)
>2. Would I ever 'want' to mix the two?... If so, under what circumstances?
>
>"Use the CursorAdapter class as a standalone class without an associated data environment.":
>1. What is the advantage of using it this way?
>2. What do I lose by doing this?
>3. Does a single CursorAdapter class object handle ALL tables I may need or is it one per or...?
>
>"Optionally define the structure of a cursor associated with a CursorAdapter object.":
>1. In what kinds of circumstances would I find this beneficial?
>
>"Control loading of data into the cursor associated with a CursorAdapter object.":
>1. what does this mean and why is it a good thing?
>
>"Control how data is added, updated, and deleted using CursorAdapter properties and methods.":
>1. what does this mean and why is it a good thing?
>2. How does this differ from the way we did things until now?
>
>"Add CursorAdapter objects to containers other than the data environment, such as forms, form sets, and other containers.":
>1. It would be nice to have an example or two of when one might want to do this.
>2. Are there any constraints if one elects to do this?
>
>Then, in a bulletted list of VFP enhancements to support CursorAdapter it says:
>"Updates the TABLEUPDATE() function so that it recognizes and handles CursorAdapter objects.":
>This is evident in the write-up for TABLEUPDATE(). But in my reading of all of the CursorAdapter Class stuff, I never saw exactly when or how I was supposed to actually issue TABLEUPDATE(), and I did get the impression that this is done "under the covers" so to speak.
>
>All in all, I am more confused than enlightened by the documentation. As I said, I find it thorough. But it's like we went right to the microscope without showing 'what' we are looking at first. I'd find it handy to have an nice overview, and in fact I suspect that there's one in there somewhere.
>Does anyone know if there is, and where if so?
>
>Thanks in advance