Hi Cameron On 25.12.2009 04:12:16 Cameron McCormack wrote: > Hi Jeremias. > > Jeremias Maerki: > > FYI, I've just finished a first experiment using Cameron's foreignObject > > branch. I've written a plug-in that can render fo:block-containers > > inside svg:foreignObject (using Apache FOP). In the end, it's even > > XSL-FO in SVG in XSL-FO. ;-) I need this for a project I'm working on. > > As soon as things quiet down I'll try to make it available. > > Nice work!
Nah, it's your accomplishment mostly. I've just written a second plug-in. > > Here's a minimal example of what I tried: > > … > > I think it would be worthwhile to do at least a partial merge of the > > foreignObject branch into trunk as Cameron's stuff seems quite fine to > > me. > > I’m not sure how stable foreign object handler interface is. At least, > I never got around to handling updates to the XHTML subtree and updating > the GVT appropriately. The interface might need some more methods to > handle that efficiently. But for static content, the interface is > probably OK as is. Ah, I didn't think about the dynamic updates. I don't need these as I'm just generating PNGs and PDFs. The interface looks fine for my purposes. The only thing I noticed is that we probably need to find a way to make requiredExtensions work. At the moment the namespace for the ForeignObjectHandler doesn't get registered for the requiredExtensions check. > > The obvious problem is the XHTML implementation which depends in the > > Flying Saucer XHTML renderer. I'd suggest to move that into the contrib > > directory with a separate build. Since Flying Saucer is LGPL, we can't > > redistribute the JAR but if people download the JAR themselves we can > > host the source code for the plug-in as optional component for Batik. > > Sounds good to me. And that jar in contrib would have the handler class > and the SPI resource file to find it, I guess. Right. That's all that is necessary. Adding support for MathML (JEuclid) (or barcodes) will be nothing after that. ;-) Jeremias Maerki --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
