> From: Jay Chiu [mailto:[EMAIL PROTECTED] > > > But if you can even not solve even this simple problem, then > please do not try to solve the problem for me. Geln Mazza asked > me to file a bugzilla report and Thomas DeWeese thought batik > implementation may require some changes. You may check fop-dev > mail list for the same subject for details. >
Well, still haven't begun to make any changes yet, but after taking a look at the sources, here's what I feel should be modified in FOP (is why I've posted this one to dev as well: Dunno if this would help you guys, as you probably know the FOP api like the back of your hand...) : You were trying to write a FOP extension for the Batik extension-tags; this seems me to be leading to more difficulties, since Batik is already being used to draw an image from the embedded SVG's. So I guess, we would, instead favor the idea of allowing embedded 'Extensible' SVG... - first, as Thomas DeWeese suggested, you want FOP to use ExtensibleSVGDOMImplementation instead of SVGDOMImplementation ( My search indicates this replacement will have to be made in 4 classes or so ( SVGElement, SVGUtilities, SVGRenderer and perhaps PDFTranscoder - the latter not necessarily). - Since we would want the batik-extension handling only in the SVG context, I guess in SVGObj you would want to modify the functions getName() and getNameSpace() to conditionally return the correct namespace for the batik-extensions ( perhaps adding an extra function SVGObj.isDomExtension() to test for this ) - Of course, the necessary mappings would then have to be added to SVGElementMapping... - SVGUtilities should be modified to accommodate the possibility of drawing batik-extensions... - SVGRenderer would have to be modified to use the functionality added to SVGUtilities. ( Don't really know for sure, but this one would only be necessary if you would also want FOP to be able to render XSL-FO to Extensible SVG... ) For the rest, I'm still looking... No matter if it's not really necessary, I just like to learn :) Greetz, Andreas Delmelle --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
