From: Avdi Grimm <avdi@...>
Date: 2007-02-22T02:28:11+09:00
Subject: Interval relationships (Was: Re: Range#overlap?)

------=_Part_8740_11495638.1172078889045
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

On 2/21/07, Pierre-Charles David <pcdavid@gmail.com> wrote:
>
> FWIW, people working with temporal logics and event algebras define 13
> possible relationships between two intervals (see Fig. 4 page 10 of
> [1]), with a precise (if not always intuitive) vocabulary.



Very interesting.  As it happens I was thinking about algorithms for sets of
intervals just this morning, for a personal project of mine.  Do you have
any more resources you could share on this topic?

In particular I'm looking for the most straightforward way to consolidate a
set of intervals such that the resulting set is composed of the smallest
number of non-contiguous, non-overlapping intervals that encompass all of
the starting intervals. E.g.:

|---|
   |---|
       |---|
              |---|
             |------|

would become:

|---------|
            |------|


Any insight you could provide would be much appreciated.

--
Avdi

------=_Part_8740_11495638.1172078889045--