substring of B and vice versa, case insensitive. For a one-off, the
trade between writing it and the time saved doing the matches by hand is
going to be a loser.
Trying to match partial and human-mangled data like this is a job for
algorithms more in the realms of fuzzy logic, which can do *irregular*
matches. Regular expressions, as the name implies, require a pattern to
fit perfectly.
--Ben
Kristen Winsor wrote:
> Yes, I am basically trying to get the closest match (or a list of
> closest matches)
>
> In the database in the company table I have entries such as ABC Company,
> Inc. On the network a user has created a folder called ABC Company.
> these are a "match"; however, I have no way to identify that this folder
> called ABC Company really "points" to ABC Company, Inc.
>
> We are going to allow access to folders from w/in our intranet; however,
> since most of the folder names do not have a match to an entry in the
> db, I wanted to rename the folders or similar.
>
> I used cfdirectory /list to get all of the folders names and put the
> names into a tmptable thinking I could then start the match process.
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings] [Donations and Support]
