Re: GUADEC

[Rodrigo Moya <rmoya tsai es>]

Chris Wiegand wrote:

> At 09:21 AM 3/21/00 +0300, Cédric PINEAU wrote:
> >
> >       Is your project GPLed ?
> >       I'm currently working on the gnome-db report myself, and well ...
> >       would be interrested in seeing what you've done so far !
> >       (especially the XML DTD)
> >
> It's OpenReports on sourceforge (no openreports.sourceforge.net yet)
>
> Sadly, not very far. I also work on another program, Geheimnis, which we're
> trying to get to 1.0, so I don't have alot of time to work on openreports.
> I've got an idea what I want to do with the XML, such as groupings, loops,
> and listing data sources, but I've been fighting with myself how I want to
> format the output. I can see two different ways:
>
> 1) Force output to be HTML. This would make the report XML similar to
> coldfusion in that you'd have HTML tags as desired (XHTML, actually) and
> invent a <report::xml> and <report::field> tag.

What I've thought myself (and Cédric as well if I remember well) about the
output is to use the gnome-print lib. Thus, you write all output to the
gnome-print supplied structure, and then once you've finished, you're able to
both have a print preview or to output it directly to the printer. Also, it's
possible with gnome-print to produce a postscript file.
Thus, you have a big part of the work done.

>
>
> 2) Use XML for a format-independant "result" file. The results of the
> various queries to the server(s)/datasource(s) would be stored as an
> temporary XML file, then an output engine (or display engine) would pick up
> the XML and format it according to the user's desires. The formatting could
> be CSV, HTML, piped back into a database, displayed on screen, printed,
> emailed in various formats, etc..., however, this method makes more work
> for me, and HTML can do most of those things anyways.

>
> Hmmm........I think tonight when I get home I will check out the anon
> gnome-db cvs once I find it, and poke around a little. Admittedly, getting
> to drop my current work on the input engines (I only have CSV done, and
> that barely) would free me up to work on recursion functions, and defining
> a good XML DTD...
>
>

It would be great if you joined gnome-db and work with Cédric. And about the
input engines, I must tell you that we plan to add as many data sources as
possible in gnome-db, including CSV, LDAP..., so concentrating on the
generation/output of the report won't drop your ideas about different input
sources. The only thing to do is to add all of these input engines to
gnome-db.

Cheers