InfoPath Dev

Hello all,
I have some non-Infopath reports that were previously filled out by off-site users and uploaded via a webservice. The data is stored in a database. Now I'm converting to InfoPath reports and don't want to make users re-enter much of this data when they create new reports (much of the data will be based off prior reports).
I'm not sure how to proceed.
Ideally, I would like to have the user's InfoPath form call a retrieval webservice, which would return data (probably XML) and InfoPath would populate its fields. Then the user could make any changes and submit. But I can't find any articles where this is done.
I also considered having a webservice accessible from our website which retrieves the info and basically creates the InfoPath-ready XML file, to let them save on their machine. The challenge here is the XML must have various tags in it reflecting stuff on the user's machine, which I won't know (ie, xmlns for my:, the path where the user has his master XSN file, version numbers, etc).
Can anyone point me in the right direction?

Hi James and welcome to our forum!
The two methods you list are both valid - a) use a secondary data source to query info for auto-population purposes, and b) create a Web service layer on top of your database that generates an XML that is InfoPath compliant.

Obviously, the first approach is faster for the reasons you mention, but it lacks support for permissions. Do you care about what your users can see? Will you just assume everyone can read everything or do you want to authenticate users? To implement permissions will take some work. Assuming everyone can read everything, it will be easier, but is often not what people need. Finally, if the users have to submit the resulting forms to the DB, you need to worry about write permissions as well and locking if they are editing existing forms.

The second approach is more time consuming but ultimately more flexible as it provides a layer of abstraction. This is the approach we take with our Database Accelerator Web service product. We solve the format issues but we don't read directly from the relational schema - instead, we store both files and data in the DB and read from the files for easier roundtripping and better data fidelity. We also have a Web service that supports easy enumeration of data for pre-population purposes that we often use on top of the Database Accelerator.

If you'd like to try our Version 2 beta, let me know and I can send you a copy.

Regards,

Thanks Patrick. I think the first option might work for my needs. Part of my problem was that I couldn't find a way to configure a data source to "receive." Finally, I stumbled upon that option, only visible if I go to Tools, Data Connections, and hit Add from there (I get the dialog for Submit/Receive). So, I may be able to use that to get what I need.

The next part of the problem is this: I have configured a data source to receive data from a webservice. But I can't find anything anywhere that explains what format Infopath expects the returned data to be in, or how to DO anything with that data. Currently my webservice sends back the entire XML from the previously-filled-in report (probably not correct). When I hit the "refresh" button I created in InfoPath, it calls the webservice, webservice returns the data, and InfoPath doesn't error. But how DO I do anything with what I've received (or how do I know I've received anything)?