Break Up a Text String in Microsoft Access
Using the InStr Function and String Manipulation
Q: I have an Excel spreadsheet where a single
column can sometimes have several different types of data in it. I need
to be able to separate those pieces of data out into different fields.
How can I do it?
A: Here is the sample data
that the customer sent me:
As you can see, the Price field contains all kinds of different
bits of data... sometimes a single price like $450,000, sometimes a
price per square foot, sometimes a range of prices, and sometimes just
"See Agent". We need to be able to break all of these down - and even
separate out the minimum and maximum values for those items that give
the range of prices so the data looks like this:
How do I do this? Well, the key is to know all of your String
Manipulation Functions and to use them properly. In this lesson I needed
to use all of these functions:
In addition to:
Here is a video tutorial that shows everything I did, step-by-step, and the
sample database that contains all of the query code:
Now, this video goes at a pretty fast clip. It's designed for people who have
already taken all of the classes listed above (at least through
Access 308).
If you have not, then I suggest you take those classes first, and then try to
tackle this one!
ADDENDUM (7/24/09): Travis asks: How do I convert this date from
1090621 into 6/21/2009?
Answer: You'll need to learn the string manipulation functions: left,
right, mid, from above.
Now, since the first three digits are 109, I'm assuming this was from a legacy
(pre-Y2K) database. This is going to make it even more fun. You have to change
the way the month and year are calculated based on the length of the whole date.
Dates from 1900 to 1999 are 6-digits. Days from 2000 to now are 7 digits.
I would start by breaking this down into different query fields:
MyDay: right(D,2)
MyMonth: IIF(Len(D)=6,mid(D,3,2),mid(D,4,2))
MyYear: IIF(Len(D)=6,left(D,2),left(D,3))
MyYear2: CInt(MyYear) + 1900
NewDate: DateSerial(MyYear2,MyMonth,MyDay)
That last one should be what you're looking for.
ADDENDUM (7/25/09): This seems to be a popular topic lately. Here's
another one. Scott says he has a table with values that look like this:
He needs to separate out that code - the [a] or [1] - from the rest of the
name. Again, our string functions come to the rescue. Here's another video:
By Richard Rost
Click here to sign up for more FREE tips
| 
