Faster Way to Import XML DATA to Catalog a Library?

[kn199024]

Hi everyone,

I'm working on a project, which involves cataloging about 2500-3000 books for the College of Writing at UC Berkeley. Of course I did not want to manually type in the tedious details of each book, so I learned the basics of FMP and XML importing. I can't use the pre built library catalog programs because they don't have a web publishing feature like FMP does.

So far I can use the XML import feature of FMP to import the necessary information of each book by querying http://isbndb.com/data-intro.html.

Although this makes it a lot easier to catalog all these books, I just know there's still a much faster way. I don't want to individually import each book's XML data by entering in their ISBN one at a time. Do you guys have any ideas? I was thinking of a script that can take as input and ISBN and automatically fills out the fields for me, or better yet the script can take in a list of ISBN and fill out all the books information. Is that possible in FMP? I have very limited programming experience but I definitely can learn quick.

To clarify how the XML import process works, I attached the XSLT file (you're welcome to use if you have similar project) that I made to import the XML data into FMP, and a screen shot.

Sample of what the http query returns:

http://isbndb.com/api/books.xml?access_key=2O6CQY46&index1=isbn&value1=0312250738

Thanks in advance!

Credits to this person for XSLT tutorial

http://edoshin.skeletonkey.com/2005/10/use_modular_xsl.html#more

LibCatalog.txt

Edited June 23, 2009 by Guest

[Fenton]

If you have the ISBN of a book in a FileMaker record, and are using that in your URL request for xml data import, then you could just do a FileMaker Loop, to go thru the found set and do each one.

Maybe you've never done a Loop? The basic script is like this:

Freeze Window

Go To Record/Request [First]

Loop

   Import XML [ your URL request, your xsl stylesheet ]

   Go To Record/Request [Next, [x] Exit after last] (don't forget to check this)

End Loop

Commit Records [no dialog]

Unless they provide a way to submit multiple ISBN's at once.

[kn199024]

Thanks for the quick reply, I just need some clarification.

Do I make a new value list of all the ISBN numbers, or individual urls to request if needed, then use the loop script to create a new record for each of the entries in that value list? If so how can I tell the script to use the entries in the value list as the url for importing the xml?