Hi Howard, You state that "the paths are now unpredictable", but it doesn't seem so to me from your example. You just have an extra couple of patterns
/root/page (as before) /root/test (new pattern) /root/test/page (new pattern) If this is the case, Digester should still work fine. The solution will be a little more complicated but not excessively so... Regards, Simon On Thu, 2003-01-23 at 23:28, Howard Miller wrote: > Hi, > > This might be a little off topic... sorry in advance > > I have an application that (using struts) displays a series of simple forms > in a web application. The forms are defined using an xml file parsed by Digester. > It looks like > > <root> > <page> > ... stuff for this page.. > </page> > <page> > .. stuff for the next page > </page> > .. and so on > > > this works fine. But things have got more complex and I now need to be able > to put tests around the pages and possibly within the pages. This will be arbitrary. > so for example.. > > <root> > <page>.....</page> > <test condition=".."> > <page>...</page> > <page>...</page > </test> > <page> > ...some fixed stuff... > <test condition="..."> > ...conditional part of page > </test> > </page> > > this seems a whole different ball game! I'm really asking if Digester is up > to the job as the "paths" are now very unpredictable. I have had a look at using > straight DOM, but this is going to be a nightmare to develop the tree-walking > algorithm. > > Any thoughts greatly appreciated. > > Howard Miller > http://uk2.net > > -- > To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> > > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
