John Jason Jordan wrote:
Kexi can import from Access, but only tables, not queries, forms, reports, etc.
So essentially all I need is MainBank.
Same situation for Base and MS Access under Windows by the way,.
Your suggestions for table design won't work because I have not been clear about the workflow. I said previously that there were three exams, but that was misleading. In fact there will be potentially thousands of exams, all based on the 7,000+ questions in the database. Each time I make a new exam I enter some specific string not used for any previous exam into the field "xxx" for each question that I want in the exam. The string is followed by a three digit number for sorting.
I believe I do understand the general work flow, I've been down the road
on this type of application before. The table design was precisely to
address it in an efficient fashion.. But different folks solve the same
problem different ways.
The final step is to export as PDF with copy prohibition for delivery to the
customer.
OK that is different - I figured you where delivering the database as an
application - personally if I where building this for a service
provider, such as yourself, I would probably use PostgreSQL or MySQL but
that is neither here nor there really.
In the past (using Access) I had a dummy table called "worktable." Instead of
merging into Writer based on a query I merged into Writer everything that was in
worktable. Thus, the queries were Insert queries into worktable, not Select queries. I
also had a Delete query to empty worktable. I created the Insert queries on the fly
because all I had to do was copy an existing query and edit the select string. The sort
was still performed in the query, however - the records in worktable were in the proper
order for the merge into Writer.
Well - in the end it is all just a result set to the merge. When you
select a Table object in OpenOffice.org it simply creates the query
SELECT * FROM Tablename and runs it. ( caveat - any sort order and
filter settings made in a table's data view window will also be applied
to the created query for Table objects, so there is a subtle difference
between query and table objects )
Copy, or select into, queries are a missing piece in the Base GUI no
dounbt - of course you can do the same with a SQL statement. Under Base
using any of the relational database engines an SQL SELECT INTO
statement you would not necessarily have to worry about the delete query
as creating a session long temp table is an option. But for now you have
to create those style queries in the SQL window and not the query
designer. ( although you can use a query object as a holding cell for
the SQL text, so you don't have to retype it each time - just copy paste )
questioncode required, 12 uppercase chars + three digits, must be
unique, keyed field
stem must be unlimited amount of text (some stems are very
long)
romannumeral1-4 four fields for Roman numeral answers, must be unlimited text
answera-d four fields for answers, must be unlimited text
answer the letter answer (A-D), required, must be one
uppercase letter of A-B-C or D
explanation must be unlimited text (some explanations are half a
page long)
miscellaneous there are half a dozen miscellaneous short text fields
that have been used in the past for selecting. I need to preserve these because
they contain the record of where the questions been used previously.
The unlimited text fields above are the ones I can't figure out. Apparently there is a "memo" field type, but it appears to want to save the text for each field in each record as a separate text file. If that is correct I'd end up with about 70,000 files on my hard disk. That is unworkable, but I can't figure out any other field type to use where the contents can be any kind of text in any quantity. I could specify a very large number of characters, but I wouldn't want the database to use that size unless necessary.
Memo fields are correct for this and no they do not save into separate
files, they also allocate space only for the amount of data stored in
them ( plus a couple of bytes ). Not sure what you mean by "any kind of
text" exactly. Do you mean you actually want to store text and formatting?
One thing, don't involve the clipboard here - just open the two Base
files ( the one that connects via ODBC to Kexi and the new native file )
drag the table from the one to the other, be sure to drop it into the
table section of the Base window, or on the Table icon.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
