Hugo, the use of an extractor/generator seems to us a little bit convoluted... a Transformer will sit more confortably in the pipeline. However, in case of multiple images on the same page, your solution would probably be faster: we understand there are more ways to skin a cat :)
Another thing: the XML dictionary you use is based on the assumption that values are linked to categories and series based on their position, an assumption we'd like to dispense with. Thanks for your contribute ! P.S. Here's an example of our draft XML dictionary (as you may notice, values are linked to catrogies and series by ids) <?xml version="1.0" encoding="iso-8859-1"?> <chart:chart xmlns:chart="http://ipzs.it/chart";> <chart:definition> <chart:properties type="bar" title="Sales and Inventory 2001" width="300" height="300" xaxislabel="Months" yaxislabel="Million EUR"/> <chart:seriesdefinition> <chart:series id="sales" title="Sales" color="0 0 255"/> <chart:series id="inventory" title="Inventory" color="255 0 0"/> </chart:seriesdefinition> </chart:definition> <chart:dataset> <chart:categories> <chart:category id="1" label="January"/> <chart:category id="2" label="February"/> <chart:category id="3" label="March"/> </chart:categories> <chart:seriesdata> <chart:seriesgroup series="sales"> <chart:item category="1" value="20"/> <chart:item category="2" value="25"/> <chart:item category="3" value="32"/> </chart:seriesgroup> <chart:seriesgroup series="inventory"> <chart:item category="1" value="40"/> <chart:item category="3" value="47"/> </chart:seriesgroup> </chart:seriesdata> </chart:dataset> </chart:chart> Luca Morandini Istituto Poligrafico e Zecca dello Stato [EMAIL PROTECTED] [EMAIL PROTECTED] > -----Messaggio originale----- > Da: Hugo Burm [mailto:hugob@;xs4all.nl] > Inviato: mercoledì 23 ottobre 2002 9.51 > A: [EMAIL PROTECTED] > Cc: Giaimo Pino > Oggetto: RE: [SUMMARY] JFreeChart, anyone ? > > > Hello, > > I have an application working that uses an extractor and a reader > instead of > a transformer. > It works exactly like the SVG extractor example. > > In the xml source you can have something like: > (the example below defines a chart with two categories and series of four, > the data points are hard coded here. In the real app they are > retreived from > a database) > > <jfc width="400" height="320" xmlns="http://datagram.nl/survey";> > <title> This is a test title</title> > <type>1</type> > <category-title>Uw score</category-title> > <category-title>Gemiddeld</category-title> > <serie-title>Niet beantwoord</serie-title> > <serie-title>Geen risico</serie-title> > <serie-title>Aanvaardbaar risico</serie-title> > <serie-title>Te hoog risico</serie-title> > <category>73,20,50,10</category> > <category>23,70,30,20</category> > </jfc> > > The standard Cocoon extractor (the one that is used for SVG) extracts it > into a separate pipeline. > That pipeline consists of a reader that reads the jfc fragment > and creates a > jfree chart (at the moment png). > Just like in the SVG example, an xsl transformation is used to replace the > jfc fragment in the original xml document with a link. > The browser replaces the link with the rendered jfree chart. > Sounds complex. > But if you understand the SVG extractor example, it is straightforward. > > At the moment it is working for html and pdf output, but only for > (stacked) > bars. And no error handling yet etc. I need some more time to implement a > better jfc chart definition (see the jfc tag above) and a better > parser for > this. > > And I am now working on a reader that creates SVG instead of PNG (should > look better when incorporated in pdf output). This is straightforward > because I have some example code from jfree chart. > > And, as always, I am struggling with the cache monster... > > Hugo > [EMAIL PROTECTED] > > We are protected from the virus by Norton Antivirus Corporate Edition --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
