Subject: MSC-News: Ambiguities in the Textual Grammar
From: Jochen Klose (Jochen.Klose#Informatik.Uni-Oldenburg.DE)
Date: Fri Nov 13 1998 - 13:35:14 GMT
The originator of this message is responsible for its content.
-----From "Jochen Klose" <Jochen.Klose#Informatik.Uni-Oldenburg.DE> to mscnews -----
I'm trying to write a parser for MSCs and have found two ambibuities in
the textual grammar that cannot be resolved automatically by the parser.
The first occurs in the definition for <substitution list> and was
introduced with the extensions for <substitution>:
A <substitution list> is a comma separated list of <substitution>s.
A <substitution> may substitute one or more <name>s by one or more
<name>s (e.g. via the rule for <replace timer>). The ambiguity arises,
because it is not clear whether a comma separates two <name>s or two
Consider for example the following substitution:
name1 by name2, name3, name4 by name5
There are two ways this construct could be parsed:
a) - name1 by name2
- name2, name3 by name5
b) - name1 by name2, name3
name4 by name5
The second ambiguity is that both the <msc statement> and the <instance
event> may start with keyword "instance". Since an <instance event>
(followed by an <end>) may be an <instance event list> the ambiguity
shows when an <instance event list> forms the end of a <msc statement>
as in the rule
<event defintion> ::= <instance name> : <instance event list>.
The keyword "instance" may now indicate either the continuation of the
<instance event list> or, since an <event defintion> is a <msc
statement>, the beginning of new <msc statement>.
My question now is, if anyone has an idea how to resolve these
ambiguities (apart from changing the language).
-- email :Jochen.Klose#Informatik.Uni-Oldenburg.de
-----End text from "Jochen Klose" <Jochen.Klose#Informatik.Uni-Oldenburg.DE> to mscnews ----- For help, email "majordomo#sdl-forum.org" with the body of your email as: help or (iff this does not answer your question) email: owner-mscnews#sdl-forum.org
This archive was generated by hypermail 2a23 : Wed Jun 19 2013 - 13:16:38 GMT