Hostname: page-component-cd9895bd7-q99xh Total loading time: 0 Render date: 2024-12-23T05:01:39.678Z Has data issue: false hasContentIssue false

Correspondence assertions for process synchronization in concurrent communications

Published online by Cambridge University Press:  03 March 2005

EDUARDO BONELLI
Affiliation:
Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030, USA and LIFIA, Faculty of Informatics, University of La Plata, La Plata (CP 1900), Argentinahttp://guinness.cs.stevens.edu/~ebonelli (email: ebonelli@cs.stevens.edu)
ADRIANA COMPAGNONI
Affiliation:
Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030, USAhttp://www.cs.stevens.edu/~abc (email: abc@cs.stevens.edu)
ELSA GUNTER
Affiliation:
Department of Computer Science, University of Illinois at Urbana - Champaign, Thomas M. Siebel Center for Computer Science, 201 N. Goodwin, Urbana, IL 61801-2302 http://www.cs.uiuc.edu/~egunter (email: egunter@cs.uiuc.edu)
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

High-level specification of patterns of communications such as protocols can be modeled elegantly by means of session types (Honda et al., 1998). However, a number of examples suggest that session types fall short when finer precision on protocol specification is required. In order to increase the expressiveness of session types we appeal to the theory of correspondence assertions (Clarke & Marrero, 1998; Gordon & Jeffrey, 2003b). The resulting type discipline augments the types of long-term channels with effects and thus yields types which may depend on messages read or written earlier within the same session. This new type system can be used to check:

  1. source of information,

  2. whether data is propagated as specified across multiple parties,

  3. if there are unspecified communications between parties, and

  4. if the data being exchanged has been modified by the code in an unspecified way.

We prove that evaluation preserves typability and that well-typed processes are safe. Also, we illustrate how the resulting theory allows us to address shortcomings present in the pure theory of session types.

Type
Research Article
Copyright
2005 Cambridge University Press
Submit a response

Discussions

No Discussions have been published for this article.