Opening Documents

Besides crawling resources, we should also be able to open them in their default viewer.

At first this may sound like a job for the DataAccessor, which after all has knowledge about how to load the stream from the physical source.

On second thought, I believe that for the opening of files you need some other service, parallel to DataAccessor, that is also scheme-specific and that takes care of opening the files. Reasons:

• DataAccessors actually retrieve the files/webpages/..., which is often not necessary for some DataObject openers. For example, for opening a local file you can instruct Windows to do just that. Similarly, a web page can be retrieved and shown by a web browser, there is no need for our codebase to retrieve the contents and feed it to the browser.

• There may be several alternative ways to open some type a resource. For example, the java.net JDIC project contains functionality for opening files and webpages, we at Aduna have written our own classes to do that, there may be customer-specific implementations, etc.

• DataOpener may be a possible interface (already in the gnowsis architecture) that opens DataObjects. They could be related to the DataSource, as all they need to find the DataObject in question would be configured in the DataSource. A tricky DataOpener is (for example) the Mozilla Thunderbird email client: opening emails there is tricky to do but may be related to a conventional IMAP DataSource. Hence, If thunderbird is there, a special DataOpener could work, while when thudnerbird is away, a default text window would show the email.

Manipulating Documents - NO

A second, more deep wish, would be to manipulate data objects. A typical manipulation would be to "move" the file or email from one folder to another or to "send via email" a data object. We have to leave these features out, as they are too hard.