• DocumentCode
    3342495
  • Title

    High-Performance XML Parsing and Validation with Permutation Phrase Grammar Parsers

  • Author

    Zhang, Wei ; Van Engelen, Robert A.

  • Author_Institution
    Dept. of Comput. Sci., Florida State Univ., Tallahassee, FL
  • fYear
    2008
  • fDate
    23-26 Sept. 2008
  • Firstpage
    286
  • Lastpage
    294
  • Abstract
    The extensibility, flexibility, expressiveness, and platform-neutrality of XML delivers key advantages for interoperability. The interoperability of XML Web services often comes at the price of reduced efficiency of message composition, transfer, and parsing compared to simple binary protocols. This paper presents a high-performance XML parsing and validation technique that is time and space optimal. A schema-specific parsing method is developed that uses a two-stack push-down automaton (PDA) for single-pass parsing and validation without backtracking. The schema validity constraints are packed in a compact parsing table derived from a permutation phrase grammar. This approach reduces both the space and time requirements of XML parsing and validation. By contrast, other XML schema-specific parsing methods trade efficiency for space (larger code and/or data size) or trade space for efficiency (backtracking). Performance results show that the method is significantly faster than traditional validating and non-validating XML parsers.
  • Keywords
    Web services; XML; grammars; open systems; Web services; high-performance XML parsing; interoperability; permutation phrase grammar parsers; push-down automaton; schema-specific parsing; Automata; Computer science; Data processing; Doped fiber amplifiers; Engines; Personal digital assistants; Runtime; Simple object access protocol; Web services; XML; Parsing; Permutation Phrase Grammar; Validation; XML;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Web Services, 2008. ICWS '08. IEEE International Conference on
  • Conference_Location
    Beijing
  • Print_ISBN
    978-0-7695-3310-0
  • Electronic_ISBN
    978-0-7695-3310-0
  • Type

    conf

  • DOI
    10.1109/ICWS.2008.101
  • Filename
    4670187