Let’s go ahead and permit both an empty graph and a graph of a single node. I cannot think of a practical use for this. But in general, you don’t want to create a design situation when ‘empty’ conditions can’t be explicitly defined,
so would prefer not to do so here without a clear reason to forbid it.
Get Outlook for iOS
From: <> on behalf of Larry Golding (Comcast) <>
Sent: Sunday, April 1, 2018 9:23:02 AM
To:
Subject: RE: [sarif] Another graph design issue: minimum number of nodes and edges
Ping on this design question.
From: <>
On Behalf Of Larry Golding (Comcast)
Sent: Friday, March 30, 2018 3:10 PM
To:
Subject: [sarif] Another graph design issue: minimum number of nodes and edges
Does a graph have to have any nodes, or any edges? Is it sensible for an empty graph to describe a portion of the program that turns out to be empty? Like, “There
aren’t any functions that call function
foo” (it’s dead code).
Larry
