Xalan/XLST builds an in memory representation of the entire document. So this will take lots of memory.
If this is a one time thing, using xlst in a JVM with enough heap allocation can work. If this is a server with lots of these processes running then you probably wany to look at using SAX instead. > -----Original Message----- > From: Clas Persson [mailto:[EMAIL PROTECTED] > Sent: Monday, March 04, 2002 10:19 AM > To: [EMAIL PROTECTED] > Subject: Transforming large documents > > > Hi! > I need to transform a large document (200Mb+). > The structure looks like this: > > <root> > <object> > <id>111111112222</id> > <sex>2</sex> > <diagLoc>018000</diagLoc> > <repHosp>10013 </repHosp> > <repClin>241</repClin> > ... to a total of 35 different elements > ... > </object> > ... > </root> > > There are almost 200 000 objects. > I need to change the name of the root element and select some > of the 35 > elements out of the document. Do I have to write a parser to > split it up or > is there a way to solve it just using XSLT? > What's the best approach? > /Clas >
