[Webinar] Streamline your web hosting managementRegister Today

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 584
  • Last Modified:

sql server 2008 NULLS in Data

HI all,
I'm working on a data migration project and have to transfer data to clients sql server. In my source I have lot of NULL values in some of the columns. what is the best practice. Where ever nulls are in source should I replace them with ''  and then send them the data OR leave nulls as it is as they can query nulls easily than empty fields?
What is the common and best practice. What if they export these nulls to excel. Will it show them nulls too.  Please suggest.
0
BI 0824
Asked:
BI 0824
3 Solutions
 
James ElliottManaging DirectorCommented:
NULL is the absence of data. I personally would not change that. SQL handles nulls very well.

Here are some functions that do it:

http://www.w3schools.com/sql/sql_null_values.asp
0
 
Jim HornMicrosoft SQL Server Developer, Architect, and AuthorCommented:
The topic of NULL vs. '' or 0 or some other character is always a big topic of debate.  Most developers would prefer the NULLs, as there is a big difference between the absense of data (posted above) and a zero, but in some cases based on business need another character is preferred.   Most business people have no frEEaking clue what 'NULL' means, and it only confuses them.

>What if they export these nulls to excel.
Excel often interprets a NULL as the string 'NULL', and write it out when a data set is exported to Excel.  In that case it may be advisable to write a query that replaces the NULL values with '', and export the query.

>What is the common and best practice
Use the ISNULL or COALESCE functions to convert NULLs into some other value.
0
 
Scott PletcherSenior DBACommented:
From a purely data perspective, NULLs are extremely valuable in differentiating between unknown/missing data and known data that is actually 0 or blank ('').  From the data side, NULLs are in invaluable advantage that should not be given up.

From the purely development and querying side, NULLs are a royal pita.  Writing queries to properly handle NULLs can be quite tricky, even for those who moderately understand NULLs, but are not really expert at it.  This is particularly true for "not equal" conditions.  For example, if you write " state <> 'FL' " in a query, you might expect to see NULL states, but you won't.

Then what to do?  I suggest that data accuracy/integrity is the most important part of a database.  Therefore, I'd opt to use NULLs and give clear guidelines to developers/power users that write queries on how to properly deal with NULLs.
0

Featured Post

[Webinar] Kill tickets & tabs using PowerShell

Are you tired of cycling through the same browser tabs everyday to close the same repetitive tickets? In this webinar JumpCloud will show how you can leverage RESTful APIs to build your own PowerShell modules to kill tickets & tabs using the PowerShell command Invoke-RestMethod.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now