Detecting Script-to-Script Interactions in Call Processing Language
Masahide NAKAMURA, Pattara LEELAPRUTE, Ken-ichi MATSUMOTO, and Tohru KIKUNO
Graduate School of Information Science, Nara Institute of Science and Technology,
Japan
{masa-n, matumoto}@is.aist-nara.ac.jp
Graduate School of Information Science and Technology, Osaka University,
Japan
{pattara, kikuno}@ist.osaka-u.ac.jp
Abstract. This paper addresses a problem to detect feature interactions
in a CPL (Call Processing Language) programmable service environment on Internet
telephony. In the CPL environment, the previous works cannot be directly applied,
because of new complications introduced:(a) features created by non-experts
and (b) distributed feature provision. To cope with the problem (a), we propose
eight types of semantic warnings which guarantee some aspects of semantic
correctness in each individual CPL script. Then, as for (b), we present an
alternative definition of feature interactions, and propose a method to implement
run-time feature interaction detection. The key idea is to define feature
interactions as the semantic warnings over multiple CPL scripts, each of
which is semantically safe. We also demonstrate tools, called CPL checker
and FI simulator, to help users to construct reliable CPL scripts.