[PLUG] MS Access -> ? for OOo

[John Jason Jordan]

On Sun, 10 Feb 2008 18:01:38 -0800
"Rogan Creswick" <creswick at gmail.com> dijo:

> csv is well-defined for funny characters -- if you have problems with
> a file either the file is corrupt or the csv implementation is
> inferior (there are *lots* of these, unfortunately, so your argument
> may well be sound based on that alone).  Generally csv isn't sufficent
> for most people simply because it only stores values (duh) this is
> probably not an issue for a db dump, but if you have spreadsheets,
> then you should know (or realize?) that csv does not persist formulas.

Another problem with .csv is getting it merged into an OOo Writer
document without the quote marks and with the field delimiters - and
all merging as it should. OOo does not remember the settings from the
last time you merged data, so you have to go through and carefully tell
OOo what the delimiters are. And the default delimiters are
non-standard - e.g., it uses a semi-colon for the field delimiter, so
as not to conflict with the comma, which it thinks should be used for
the thousands separator. If you need to merge data into OOo, then .csv
will drive you nuts.

OOo Base would be perfect, but the Linux version cannot read MS
Access .mdb files. Kexi can do so, however. But then OOo can't read
Kexi data either - evidently Kexi uses "SQLite format 3." According to
the literature OOo Base is supposed to be able to read SQL and open
document formats, but SQLite doesn't seem to be on the list. At the
same time, the only format Kexi can export to is .csv. Around and
around I go.

The Access database file contains 43 tables, now happily residing in
a .kexi file. Originally the Access file had numerous queries as well,
but those didn't survive the import into Kexi. I had it set up in
Access so that one table ("Worktable") would be the table that all word
processor merges worked from. Then I had a query to delete all data
from Worktable, and numerous other queries to add data to it. Lots of
times I would quickly create an ad hoc query when I needed specific
records, always merging the records to an empty Worktable, and from
there into OOo Writer. I did all the querying and all the data editing
from within Access. I can probably replicate the workflow with Kexi,
but there seems to be no way to merge data out of Kexi into a word
processor without exporting as a .csv. I can't even get KWord to merge
data from Kexi. Kexi gets rave reviews, and I have to agree it is
really easy to use. But it's like the Roach Motel - data checks in, but
it never checks out.

Kexi is also very fast, while OOo Base is uniformly lambasted in
reviews as being pathetically slow. I did bring in the one .csv file
(5,000 records) and tried a simply select query. It took roughly four
minutes. I did the same query in Kexi and it finished in less time than
it takes you to read this sentence.

My problem is that the database contains test questions that must be
merged into a word processor so that styles can be applied to the
different parts of the question during the merge. For example, the
indents and tabs for the stem must be different from the indents and
tabs for the answers. The only way to accomplish this is a merge into a
word processor that supports styles. But i can't find a common format
that everything likes.

The nice thing about standards is that there are so many to choose from.