Piazza forum

Schedule

Staff

Prerequisites

Project

Additional Reading List

Database Systems: Advanced Topics and Implementation
CSE232B

Welcome!

For announcements regarding the course, please sign up to our Piazza forum.

Milestone 1 of the project will be due in week 4, Milestone 2 in week 7, milestone 3 in week 10. Stay tuned for exact dates.

Schedule

.	Mon	Tue	Wed	Thu	Fri
Lecture	9:00-9:50 pm Pepper Canyon 109		9:00-9:50 pm Pepper Canyon 106		9:00-9:50 pm Pepper Canyon 106
Instructor Office Hours					10:00 - 11:00 am  CSE 3238
TA Office Hours			4:00-5:00pm CSE B270A

Staff

• Instructor: Alin Deutsch, deutsch at cs dot ucsd dot edu, office CSE 3238
  
• TAs:
  Nikos Koulouris, nkoulour   at cs dot ucsd dot edu
  Ainur Smagulova, asmagulo   at cs dot ucsd dot edu
  

Prerequisites

• cse232A or instructor's permission, granted only if the following prerequisites are fulfilled
• Java (8B or 11 or equivalent)
• SQL (132A or equivalent)

Class Project

You may team up with (at most) one partner for the class projects. Use the Piazza forum to advertise if you are looking for a partner.

Our class project is the construction of an XQuery processor. We consider a subset/modification of XML’s data model, XQuery, and XQuery’s type system as described in this note. The processor receives an XQuery, parses it into an abstract tree representation, optimizes it and finally executes the optimized plan.

• Milestones 1 and 2 (Naïve Evaluation) [due in weeks 4 and 7, respectively]: A straightforward query execution engine receives the simplified XQuery and an input XML file and evaluates the query using a recursive evaluation routine which, given an XQuery expression (path, concatenation, element creation, etc) and a list of input nodes, produces a list of output nodes. For the XQuery parser, we recommend the ANTLR4. Provided with a grammar, ANTLR generates a compiler which automatically constructs abstract syntax trees of  its input expressions.
  Milestone 1 delivers a naive evaluator for XPath, while Milestone 2 extends it to XQuery.


• Milestone 3 (Efficient Evaluation): Implement a join operator as defined in this note. Implement an algorithm which detects the fact that the FOR and WHERE clause computation can be implemented using the join operator. You may assume that the input XQueries to be optimized are in the simplified syntax given in the note. No need to first normalize your queries to this form.

To access XML files you can use the standard DOM interface. There are a number of XML DOM parser implementations.
The Java distribution includes one (see documentation here). As an alternative, the Xerces-J project from Apache is quite mature and stable.

The W3C specification of DOM is here.

Test cases for project Phase I. For the data, download Shakespeare's play, Julius Caesar, in XML form (the associated DTD is here). Queries can be found here.

Additional Reading

• Minimization of XPath queries.
• Minimization of Tree Pattern Queries. S. Amer-Yahia et al.  []
• Efficient Algorithms for Tree Pattern Minimization P. Ramanan []
• More XPath Optimization
• XPath: Looking Forward D. Olteanu et al. []
• Processing of XML Streams.
• Processing XML Streams with Deterministic Automata T. Green et al.  []
• Stream Processing of XPath Queries with Predicates A. Gupta et al.  []
• XML Stream Processing Using Tree-Edit Distance Embeddings M. Garaofalakis  []
• Systems for publishing relational data as XML
• Agora: Living with XML and Relational I. Manolescu et al. (a very short overview paper that goes together with the next one)  []
• Answering XML Queries over Heterogeneous Data Sources I. Manolescu et al.  []
• XPERANTO: Publishing Object-Relational Data as XML M. Carey et al. []
• Querying XML Views of Relational Data J. Shanmugasundaram et al. []
• Storing XML data in relations
• Storing Semi-Structured Data With STORED M. Fernandez et al.  []
• Relational Databases for Querying XML Documents: Limitations and Opportunities J. Shanmugasundaram et al  []
• Storing and querying ordered XML using a relational database system. I. Tatarinov et al.  []
• Efficient XML Query Processing
• On supporting containment queries in relational database management systems. C. Zhang et al. []
• Structural joins: A primitive for efficient XML query pattern matching. S. Al-Khalifa et. al  []
• Holistic twig joins: optimal XML pattern matching N. Bruno et al. et. al  []
• XR-Tree: Indexing XML data for efficient structural joins H. Jiang et. al  []
• Accelerating XPath location steps T. Grust et. al  []
• Pushing XPath Accelerator to its Limits C. Gruen et. al  []
• An evaluation of XML indexes for structural join H. Li et al. et. al  []
• Staircase Join: Teach a Relational DBMS to Watch its (Axis) Steps. T. Grust et al.  []
• Architecture and Performance of the Hardware Accelerators in IBM’s PowerEN Processor T. Heil et. al  []
• Keyword Search in XML Documents
• GalaTex: A Conformant Implementation of the XQuery Full-Text Language S. Amer-Yahia et al. []
• Efficient keyword search for smallest LCAs in XML databases Y. Xu et al. []
• Interconnection Semantics for Keyword Search in XML S. Cohen et al. []
• XML Algebra
• TAX: A Tree algebra for XML H. V. Jagadish et al. []
• An Algebra for XML Query P. Wadler et al. []
• XML Extensions
• The Active-XML Project S. Abiteboul et al. []

Grading

The project constitutes 80% of the final grade; final constitutes 15%; class/Piazza participation constitute 5%.

Reading List

Formal XPath Semantics note

A brief informal XQuery tutorial (much briefer and more readable than the W3C standard below)

Textbook material (from the warmly recommended textbook "Web Data Management"):

• XML Data Model (slides, chapter)
  
• XPath (slides, chapter)
  
• XQuery (slides, chapter)
  

• XQuery Advanced Topics slides.

The complete XQuery and XML Schema documentation (the WWW Consortium standards):

• XQuery Data Model
• XQuery
• XQuery Formal Semantics (“XQuery core”)
• XML Schema