import/export xml?

101 views
Skip to first unread message

tim atherton

unread,
May 3, 2009, 1:53:40 PM5/3/09
to ICA-AtoM Users
Looking for a bit of help

I have basically no experience with xml so...

How do I actually go about exporting data from an old database system
and getting it into the correct xml format to be able to import it
into ICA-AtoM?

The old system is a filemaker pro database (which uses mainly RAD
based fields), and the data can be exported from there in a number of
different forms - csv, tab, xls, xml etc, so once exported I can work
on adjusting the filed names and so on to comply with the EAD xml
upload.

But I'm foggy on what I actually need to do with that exported data
and how I get it formatted into xml so it can then be all loaded into
AtoM.

I've got the stuff about the EAD xml templates for AtoM and so on, but
is there a sort of idiots guide to xml around somewhere? I basically
need an outline of the steps between exporting the data from filemaker
in the most useful form and then getting it all into an EAD xml
template format for importing to AtoM (especially that latter bit)

If that makes sense....


thanks

tim a

Richard Dancy

unread,
May 4, 2009, 5:10:20 PM5/4/09
to ICA-AtoM Users
Hi Tim,

We also use a FileMaker database here at the SFU Archives. We maintain
our descriptions in the FM database, then export as EAD to upload to
our local version of ICA-AtoM which will provide the web interface for
researchers (for the time being we're keeping the internal FileMaker
database because it is tied in with a bunch other databases for
accessioning, locations, reference, records management etc).

For export to EAD, we've written our own FileMaker scripts. Each
description record has its own "zzFindingAid__ead" field which is a
text field set by the script to take the data from various RAD fields
then wrap EAD tags around it. The script processes the lower levels
first and then inserts them into the "zzFindingAid__ead" field of its
parent and so on up the line. So when you get to the fonds record, the
"zzFindingAid__ead" field contains all the coding for the entire
finding aid, nested hierarchically (e.g. series 1 with all its sub-
series, files, items, followed by series 2 with all its sub-series,
files, items etc). Then what actually gets exported is just this one
field from the one fonds record, exported as a "tab separated text"
file but with the extension changed to ".xml".

There might be easier ways to do this. But if you just export a set of
records using eg. FileMaker's xml export, I think you'll be stuck
having to do a lot of nasty EAD coding manually.

I can send you copies of the scripts if you'd like -- though one of
the hassles with FM is that you can't actually export scripts, but I
can print pdfs and send you those. They're also totally customized to
our system, so may or may not be useful to you depending on which
version of FM you are using (we're on FM10) and how your data's
structured. Let me know and I'll send them to you off-list.

Cheers,
Richard

tim atherton

unread,
May 7, 2009, 3:11:11 PM5/7/09
to ICA-AtoM Users
Richard - if you could send me those, they may porver very useful -
thanks

tath...@st-albert.net

tim

Richard Dancy

unread,
May 7, 2009, 6:15:48 PM5/7/09
to ica-ato...@googlegroups.com
Hi Tim,

OK, here's the scripts as a PDF.

Cheers,
Richard
eadScripts.pdf

Rafael Antonio

unread,
May 8, 2009, 8:36:47 AM5/8/09
to ica-ato...@googlegroups.com
Dear users,

I still sugesting another export format - ISO2709 is a very useful a used for a larger number of former solutions.
Please have it in mind,

Rafael Antonio
Reply all
Reply to author
Forward
0 new messages