Technical Questions
 Crystal Reports Forum : Crystal Reports 9 through 2022 : Technical Questions
Message Icon Topic: Exception Reporting ideas Post Reply Post New Topic
Author Message
jrhodes
Newbie
Newbie


Joined: 08 Jun 2010
Online Status: Offline
Posts: 9
Quote jrhodes Replybullet Topic: Exception Reporting ideas
     Posted: 14 Dec 2011 at 10:04am
Hi all,

We have a large database of books. What I'm noticing is a lot of duplicate entries (same title, different unique id) are occurring and generally the database needs a good 'clean'. Unfortunately our entry system doesn't have any validation so duplicates can be created easily.

I want to create a report that examines these records and identifes duplicate records (based on a book title) or 'similar' titles. I understand this is called exception reporting? The idea is on weekly basis we can run reports and clean up the data before it becomes a larger problem later.

At a basic level I can sort by title and get identical titles by comparing one record to the next (using NEXT and PREVIOUS functions). But what I want to do is identify records that might be 'similar' e.g. the same title but a slight variation such as a comma or hypen etc.

Does anyone have any experience with this? Or any idea how this might be done? I'm thinking regular expressions but not sure how these are done in Crystal? I'm looking for a good starting point. ---Thanks for any ideas.
IP IP Logged
DBlank
Moderator
Moderator


Joined: 19 Dec 2008
Online Status: Offline
Posts: 9053
Quote DBlank Replybullet Posted: 14 Dec 2011 at 11:42am
generally this is not unique to crystal and has more to do with what logic would you want to apply to flagging these records. Form there it comes to how would you do that in crystal. Also how do you stop flagging them once you have reviewed the infomation and determined it to be OK.
Do you knly have title or do you also have author?
do you have a set of data that is more reliable and you wnat to use to check against 'newer unvalidated' data and is there something your your data set that you can use to differentiate these rows?
 
Lockwelle would likely use a stored procedure to do this and I would lean toward that as well. It would be easier in that to add the table as many times as you need to using a master table of 'good or scrubbed data' to check agains t various version of the 'new unscrubbed data' joining on LIKE statements multiple joins on author last name and partial title match, titel matches when replacing all puncuation with a space, etc.
IP IP Logged
Printable version Printable version

Forum Jump
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot delete your posts in this forum
You cannot edit your posts in this forum
You cannot create polls in this forum
You cannot vote in polls in this forum