doc.txt

Distributed Programming II
Assignment 1 - DTD and Schema Documentation
Student: Fabio Bulgarella (s237094)
Politecnico di Torino - A.Y. 2017/2018
-------------------------------------------

Both DTD and XML Schema was developed in order to validate with success the same xml file.
Obviously schema has a way more complex and fine validation subsystem.

The elements structure is pretty simple:
nfv -> the root element;
	catalog -> vnf
	nffgs -> nffg -> node -> link
	hosts -> host -> noderef (i.e. allocated nodes)
	connections -> connection

For optimization reasons I preferred to use almost everywhere attribute instead of element.
Nevertheless, this choice didn't compromise readability of the xml document, actually
a reader can evaluate much more information in a small space.

About DTD, functionality like ID, IDREF and #REQUIRED keyword were exploited where possible.

Some consideration about Schema:
- Venetian Blind Design was used, maximizing reuse of components. So all except root element is a complexType.
- Nffg, node, host and vnf elements are characterized by a name, which uniquely identifies them in the document.
- Link element is characterized by a name, which uniquely identifies it in the respective nffg.
- Keyref where used for those attribute that refers to unique elements.
- A simpleType "nameType" was declared to constraint "name" attribute format (with pattern [A-Za-z][A-Za-z0-9]*).
- Other simpleTypes "nonNegativeInt" and "nonNegativeFloat" was declared for attribute memory, storage, latency etc...