adoquery Datatype decision based on 51% of the records...

I really hope somebody can help me with the
following ;) :

I have a program reading the records from an excel file using the connection
like :

strConn:='Provider=Microsoft.Jet.OLEDB.4.0;' +
              'Data Source=' + Form1.ED_filename.Text + ';' +
              'Extended Properties="Excel 8.0;HDR=YES;MaxScanRows=0;MaxBufferSize=2048";';

and getting the fields type with:

var loopint:integer;
  for loopint:=0 to adoquery1.FieldCount-1 do

The Problem I have is that in a column called 'PurchaseOrder' I have 80% of the values that are numeric (like '0008332')
and 20% that are strings (like 'M002233')

When I use datatype, it returns FLOAT. I've tested it using the default 16 rows and inserting records like 9 strings and 7 values
(it returns STRING) and 9 values and 7 strings (it return FLOAT).

The problem is that if I have a field that has both Values and Text... it should return text, not values.

Thank you in advance!

Who is Participating?
aruanaConnect With a Mentor Commented:
it may not be a good idea to use FieldList[].datatype to check for data in Excel.
What you are getting may be based on the first row only.
Also, what if a user enters numberic value in text format ? In excel, the appearance is the same as numeric value format.
i dont think you can avoid reading all of them.
50K rows for excel is certainly not a good idea for storing data.
is there any reason why you cant transfer all of them into MySQL as strings and later process them and change the data type if neccessary? this would be faster.
BaTy_GiRlConnect With a Mentor Commented:

>The problem is that if I have a field that has both Values and Text... it should return text, not values

Why you dont manipulate the rows which has strings, cutting the strings, maybe reviewing letter by letter...
use MidStr to cut the strings, and the function IsNumeric to review letter by letter    

dcrudoAuthor Commented:

Because the Program I'm trying to write should be able to automatically transfer Whole Excel worksheets data, not
always the same format, directly into a MySQL database.

I will not be using always the same format (the one above is just one example to replicate the issue).

If possible, I would like to avoid checking record by record... because some worksheets are quite big
(20 columns by 50K rows).

dcrudoAuthor Commented:
The answer to my problem:

Add 'IMEX=1' in the connection string will force JET to read all fields to return a Mixed field type and not
on the higher ration (51% number = number, 51% string = string)... if the field contain mixed number or will always return string (mixed type).

Since no one of the answers helped me a lot... i will split the points for the effort!

Thank you!

Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.