Error in importing data

[Tony]

I am trying to import a file from excel into a already
existing table. But, it is saying that it cannot do it.
Does anyone have any suggestions on what I am doing
wrong? Thanks

[MaximoAssist]

The most common problem I see is usually in the first 2 rows in the sheet, so
here is kind of a standard approach to match things up.

Common problems: ultimate field/column names in row 1 do not match those in
the table EXACTLY or data in row 1 or 2 is interpretted in the wrong format
for at least one field.

Solution to finding the problem:

Import the spreadsheet into its own new table with the following
modifications made to the sheet:

Make sure row 1 has the field names and insert a new row 2 where each column
has 55 or more x's or any plain text character. This will force all fields to
be interpretted as text, and then the primary error that appears should be
based on the row 1 data that becomes the field names (remember to pick this
on the import wizard).

On the off chance you have any large data in the cells, then you should make
insert 260 x's (or more) to force access to interpret each field as memo. You
can play with this a little to just force memo where required, but it is
usually easier to inspect and track down bad data with text fields.

Once you identifiy the problem usually the rest is easy. Best of luck.

[onedaywhen]

MaximoAssist wrote:

> Make sure row 1 has the field names and insert a new row 2 where each
column
> has 55 or more x's or any plain text character. This will force all
fields to
> be interpretted as text, and then the primary error that appears
should be
> based on the row 1 data that becomes the field names (remember to
pick this
> on the import wizard).

It's not always possible/practical or desireable to physically change
the Excel data. For an explanation of the reasons and a way to change
the TypeGuessRows registry key to get mixed columns 'seen' as TEXT,
see:

http://www.dicks-blog.com/archives/2004/06/03/external-data-mixed-data-types/

> On the off chance you have any large data in the cells, then you
should make
> insert 260 x's (or more) to force access to interpret each field as
memo.

This is a good point: having a column 'seen' as TEXT will result in any
MEMO data to be curtailed at 255 characters. In this scenario, using a
dummy row 1 (after the header row) and using TypeGuessRows=1 may be the
best solution.

Jamie.

--