Hi there -

I've managed to get the XML (well, XMP) data from imported images into a field labeled "XMP", but now I would like to parse that data into different fields according to hierarchical subject heading.

The data in the "XMP" field looks like this:

Nation|Germany Image Type|Photo - WW1 Era Original Aircraft Manufacturer|Friedrichshafen Aircraft Type|Biplane Aircraft Type|Seaplane Aircraft Seats/Use|Bomber Setting|In Flight

My fields are named "Nation", "Image Type", "Aircraft Manufacturer", "Setting", etc. Obviously, the text string that I would like to parse immediately follows the subject heading and the "|", then ends right before the "".

Any help with this is much appreciated!

I've done a lot of XML scraping

This should work for each field you want to scrape by modifying just the first variable 'fieldNameStart'.

Here's the formula I like to use for each field (modified for your example - hopefully didn't create an error with modifying my code to match your needs):

Let (

[

fieldNameStart = "Nation|" ;

fieldNameClose = "" ;

fieldNameLength = Length ( fieldNameStart ) ;

dataBegin = Position ( YourXMPfield ; fieldNameStart ; 1 ; 1 ) ;

dataEnd = Position ( YourXMPfield ; fieldNameClose ; dataBegin ; 1 ) ;

dataLength = dataEnd - dataBegin - fieldNameLength

] ;

Middle ( YourXMPfield ; dataBegin + fieldNameLength ; dataLength )

)

Thanks! I'll give it a try now and report back.

It works great - you're an absolute legend.

Now - one final question:

How should I alter this script if there are multiple instances of each keyword, or if there are none?

For example, sometimes in the xml data there are multiple "Nation" keywords - like this:

Nation|Germany Nation|France Nation|Britain

I would like to have the script add each string to the "Nation" field, separated by a comma or dash.

Likewise, sometimes there is no "Nation" data for the record at all - and when this happens, your script put a ton of unrelated xml data into the field.

Thank you again for your help!!

Edited July 14, 200817 yr by Guest

To handle no-occurrences I modified the code to:

Let (

[

fieldNameStart = "Nation|" ;

fieldNameClose = "" ;

fieldNameLength = Length ( fieldNameStart ) ;

dataBegin = Position ( YourXMPfield ; fieldNameStart ; 1 ; 1 ) ;

dataEnd = Position ( YourXMPfield ; fieldNameClose ; dataBegin ; 1 ) ;

dataLength = dataEnd - dataBegin - fieldNameLength

] ;

Case (

dataBegin > 0 ;

Middle ( YourXMPfield ; dataBegin + fieldNameLength ; dataLength ) ;

"" )

)

To handle multiple occurrences of 'Nation' and/or others, you have a couple options.

1) if you have Advanced you could create a recursive function to handle all instances of 'Nation'

2) if no FMPAdvanced, you'll need a field for every possibility. ie for up to 5 possible Nations you'll need 5 fields. Use the above code for each field. Change the Position code in each field to match the occurrence.

* Nation3 would replace the position function parameters to 'Position(YourXMPfield;fieldNameStart;1;[color:red]3)'

This method for capturing multiple occurences is working perfectly. Thank you so much for your help!