Dinbandhu wrote:
Drew,
I tried hard for almost two hours to import a table using the copy table
wizard, but could not get success. I went over your tips (below)
carefully, and incorporated all the points relevant to my particular
table.
My table has 23 columns, all of type 'text' except for four: the ID
which is AutoNumber (in MSA terminology), two of type "date/time", and
one of type "yes/no". And that table has 40 rows.
So I carefully went through the type fields when the copy table wizard
would get to that window.
For the ID field, according to your below recommendation I set it by
hand. It was already set to BIGINT, but I changed it to something else
and then brought it back to BIGINT. By so doing, the number of digits
allowed went up from 4, to 19. I do not know whether I was able to set
it the way you wanted, but this was the best way I could figure out to
increase the number of permitted digits. You had written to be sure to
set it by hand, and then added that "AutoIncrement in Access is always a
LONG INTEGER". There didn't seem to be any option in BASE for selecting
"LONG INTEGER", so I did what I have described above.
For the two "date/time" fields, BASE had selected something else other
than "date/time"-- it had selecting something obviously inappropriate.
So I changed both to the "date/time" type.
And the "yes/no" column was also incorrectly set. So I changed it to the
"yes/no" (boolean) option.
All the remaining were text fields, appropriately designated so by BASE,
with "var char". But I went through each and made the number of allowed
characters "255".
After making the above-noted changes, upon clicking the "finish" button,
the wizard still came back with the message that there was an error, and
would I like to continue writing the table anyway. I answered yes. This
time, unlike earlier, the table is not empty. It filled in five rows.
But the data is wrong. For example, in the "date" column, all five cells
have the same date in them (01/01/70 12:00 AM)-- which is wrong. I do
not have any such date in the table.
Also, the order of the columns is different from the actual MSA table in
my windows partition. But that error seems to stem from the way BASE's
"connected" table appears. In that table--the one BASE made in the
"connect" mode--the data appears correct, but the order of the columns
in the table is all wrong. And that incorrect order is reflected in the
order of columns in the table made by the copy wizard.
So if you have any tips/suggestions as to how to make this import
proper, please kindly let me know.
Otherwise I am thinking I may get more success by using the .csv method
for doing the import. I have already done the export from MSA into
a .csv format, and would just need instruction as to how to import
that .csv table into linux BASE.
Thanks for all your help,
Regards,
Swarup
<snip>
I don't know if this will help, but you could try opening your .csv file
in Calc and see what you get there. If it looks good, you should be able
to set the next column in Calc (X, I guess) for the ID field. Set the
first value as one, then set the formula in the next row as "=X1+1"
(without the quotes, of course), press Enter to set X2, then drag the
fill handle from X2 down across the rest of the rows; they should then
all fill in with sequential values, which will be your ID field. Now you
can define a new table the way you really want it (primary key last, as
auto-value BIGINT; I'd recommend using Design view so you can see
exactly what you're getting). In Calc, select the whole table by
clicking in the box at the far left of the column designator row, then
copy. Back in Base, this time when you do the paste, tell it to append
the data to your new table. Then keep your fingers crossed!
(Apologies if this is beating on lots of obvious stuff, but I wanted to
say too much rather than too little!)
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
