DOM or SAX in parsing.Please GUIDE

Question

If I have a huge document that I need to parse,
which is the recommended parser I shud use?
SAX or DOM.??
Isnt DOM impractical to use if the document fills
gigabytes?
Can some one please answer?

Answer 1

SAX reads through the document and sends you events of what is read.
SAX will never store any of the data which is read.
You can do whatever you like with these events.
DOM reads through the document and makes a memroy-model of the XML-file.
This will store the complete XML-document in memory.
Afterwards, you can browse this structered (in memory) data within your program.
There is a tradeof between DOM and SAX.
SAX has a smaller (constant) memory-usage (certainly with large XML-files),
and has a faster processing of the data.
But SAX delivers no real structured data and is only read in one direction.
SAX should be used when large files are read only once in a linear direction
and if the processing of the data does not rely to much on the structure of the data.
DOM has a larger memory-usage (certainly with large XML-files),
and has a slower processing of the data (due to the initial parsing of the file).
But DOM delivers real structured data which can be manipulated on the structure in any direction.
DOM should be used when smaller files are read, when many different processing-steps should
be performed on the data or when the processing of the data is heavily relying on the structure of the data.
remark that DOM is using SAX as the underlying mechanism for reading the data.
DOM will read the data using SAX and will build the in-memory structured model of the data
according to the SAX-events.
remark also that other solutions such as JDOM and minimal DOM's exist.
kind regards,