[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Off Topic? Guarded commands...

The reason may be simply esthetic:
I do a lot of unix shell scripting, but even
though I write case...esac and if...fi all
the time it irritates me.

Kevin Kelleher

} -----Original Message-----
} From: Peter H. Froehlich [mailto:pfroehli@ics.uci.edu]
} Sent: Tuesday, December 11, 2001 16:25
} To: ll1-discuss@ai.mit.edu
} Cc: Christian Stork
} Subject: Off Topic? Guarded commands...
} Hi there!
} Sorry, I am "abusing" this list because there are so many smart and 
} experienced PL people here. If you prefer to reply in private I can 
} post a summary later.
} I have always been intrigued by Dijkstra's guarded commands, i.e. 
} the if .. fi, do .. od things, and I keep wondering why they never 
} make it into any (reasonably popular) programming language.
} I come back to this question in cycles, and usually it happens when 
} I run across a paper or book that mentions them. Two recent 
} examples are when I browsed lots of compiler construction books for 
} a course I gave, and when I read up a little on the ESC project at 
} SRC. It seems that guarded commands are used only
} (a) to "abstractly" proof algorithms that are then implemented in 
} another notation/language
} (b) to "annoy" students in text books on programming languages and 
} compiler construction
} (c) to serve as an "intermediate representation" for more 
} "standard" language constructs (see SRC's ESC or the work on 
} "guarded single assignment form" done by Brandis at ETH, 
} Switzerland)
} So, to sum up what I am looking for: Does anyone know references on 
} "real" programming languages that use guarded commands, or can 
} anyone shed some (possibly subjective, possibly anecdotal) light on 
} why they don't seem to catch on?
} Peter
} --
} Peter H. Froehlich @ http://www.ics.uci.edu/~pfroehli/