Re: Antw: New gda-report's DTD

[Carlos Perelló Marín <carlos hispalinux es>]

Gerhard Dieringer escribió:

> Carlos Perelló Marín wrote (in several mails):
> > Ok
> >
> > as i have see it seems all right. I think that need another revision,
> > but it's much more better that my DTD, i will upload it into the gnome's
> > CVS. Well, if you don't have write access, of course.
>
> I don't have write access.

Well, this night i will upload it.


>
>
> > > <!ATTLIST pageheader  [or pagefooter]
> > >   height        NMTOKEN   #REQUIRED
> > >   positionfrequency  (%positionfrequency;)   "all"
> > >   pagenumfrequency (%numberfrequency;)    "all"
> > >  >
> > ....
> > But, then if you want print to headers in the same page?
>
> I a user defines one pageheader with  pagenumfrequency="all" and the second with pagenumfrequency="even" then of course the even pages have two page headers and the odd pages have one.
>

Ok.


>
> > Well, i think that a "repfield" should have a label.
>
> I don't see the need to distinguish repfield-labels and other labels, but maybe I'm wrong.
>

I don't understand well your comment in the Fri, 28 Jul 2000 13:06:45 +0200 email:

"I wouldn't consider a label, that is found in the metadata
dictionaries call a 'label', but a repfield of it's own. But
maybe I'm wrong."

Sorry, please forget that.


>
> > I think that yours opinios are equal or even better than my opinions.
>
> Nice to hear.
>

;-D


>
> > Well, as i tell you in another mail, i think that we must connect this
> > option with the options at /etc/paper.conf
>
> I don't have this file on my GNU/Linux-box (SuSE 6.3).
>

Forget this also, this file is a Gnome file, and this part should be gnome indepedient.


>
> > > 3.) They distinguish fields and calculated fields where the last actually are SQL aggregates and they allow them only in the report footer. I think both
> > > points don't make too much sense because:
> > > a) a field gets a calculated field, if the corresponding query delivers a aggrgate function
> > > b) calculated (aggregate) fields are sometime usefully in the detail section on a report.
> > >
> >
> > Well, this could change, couldn't it?
>
> I don't understand your question. what could change?

This:  "... I think both points don't make too much sense because: ..."

>
> > > A report designer - or another application that wants to output a report - generates a xml-report-template. This is send to a transformation engine that
> > > takes the contained queries, sends them to Vivien's gda-xml-query for processing and inserts the results into the xml-report. The results should be put
> > > into label elements, that replace the repfield elements.
>
> > I don't know if this should be correct. Why you want to save the report
> > query in a XML file? Why not save it only in RAM ?
>
> IMO, I should be possible to save both -- the report template with the queries inside and the resulting report with the data inside -- into a file.
> Then you can process the report again, with changed data in the DB or reprint the report with the old data, if needed.
>

Well, it is only an extra feature.


>
> > Yes, i want develop one report designer, but i think that we must
> > develop before a report viewer. What do you think?
>
> IMO the report designer should be developed first -- or parallel to the viewer, because
> - the viewer needs some input, that actually is the output of the designer
> - the report should be populated with some data from gda-xml-query, that is not ready now.
>

Well, i think that this isn't so urgent, because you always could create the report file with "vi" :-D, but the representation is not so trivial.


>
> > I'm trying to port a Crystal Report's application to Linux, then i'm
> > looking at Crystal reports structure, but it isn't the unique thing i
> > will see!!!
>
> I don't know Crystal Report
>
> I also attached a new enhanced version of the DTD with some new elements: dataheader, datafooter, picture. The dataheader and data footer are similar to the group header and footer, but for an imaginary group, that contains all the data. The purpose of the picture element should be obvious.

>
> Gerhard
>
> PS Please reply to GDieringer@compuserve.com because I will be on holiday for two weeks.
>
>   ------------------------------------------------------------------------
>
>    gda-xml-report.dtdName: gda-xml-report.dtd
>                      Type: Plain Text (text/plain)

I will read the DTD tonight and then i will upload it into Gnome's CVS, ok?