[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Other Ozone XML/OODB



Hello Falko,

The point is not too easy to explain in an e-mail extent. Trying not to
be too long (and maybe boring): our strategy is producing a translation
from the DTD to a Data-model (Objectual) (in fact a meta-translator).
Then, for each DTD we'll have a data-model-structure (or data-base).
Then you can use all the tools ODMG 2.0 provides.

I have never mean I'd prefer to parse the XML file every time to produce
a DOM structure. I am trying to avoid the DOM structure for working
in-house. Only when I need to talk to an "alien", I have to "speak
speranto", and then I have to produce XML again o DOM binaries
(depending on our agreement and needs).

The idea exposed in the first pragraph it is no so easy to put into
practice, but it have a lot of "good properties" (as long as "good
problems" to be solved). We have already do some prototypes and it works
fine.

Pedro.


> Another point is that if you are using XML only to publish data then a stream
> processing can be used and the DOM need only to be stored in memory. I think
> COCOON is an example for that. In other cases the XML is not only used for
> publishing but it represents the actual content of the application. I think of
> documentation and information systems. So far it seems that an query language is
> needed for the second case only. (?)
> 
> In the near future we are going to start such a web based information system. It
> will contain a lot of "text" documents - project descriptions, press releases
> whatever. We need not only to publish them but also to search, modify, build new
> documents of parts of other documents... Why not storing the XML as DOM instead
> of parsing the XML every time I need the DOM?
> 
> Falko
> --
> ______________________________________________________________________
> Falko Braeutigam                         mailto:falko@softwarebuero.de
> softwarebuero m&b (SMB)                    http://www.softwarebuero.de