I am exploring options for using streaming in XSL. I know two obvious cases:
and. You need to convert a very large document that cannot be stored in memory. B. You only need a small part of the document, and often this “small part” is near the top. Then you can save time through an early exit.
I am writing to ask if there is a third real use case in practice:
C. You have a simple transformation and want to give up the processor time needed to build the XML tree. To give an example, imagine that warehouse deliveries are stored in an XML structure in the following format:
High = Year
2nd level = month
Level 3 = Shipment Day
Level 4 = Shipment Identifier
5th level = individual items in shipment
For example, consider a transformation whose purpose is to pull out information at the "month" level .... we only need data stored in the attributes of the elements of the month and not requiring information about the descendants of these nodes.
Is it possible that such a conversion could benefit from streaming, although the entire document needs to be read? I was hoping that some time could be achieved because there is no need to build trees, but in my limited testing it seems like it is not.
SAXON 9.5.1.3, 20% , .
, , , , , ? ( , SAXON, .)
, , , ?