Bulk Import Questions

Hi,
I very much appreciate the high-quality modules that are shared for Omeka-S.

I have a workflow that is going to require hundreds of batches of imports for items and media. The Bulk Import module may be useful for this. But there are a couple of things I am having trouble figuring out:

  1. Is it possible to map a metadata property as a numeric data type using the spreadsheet options? I see that the documentation says " The header can have a language (with @language ), a datatype (with ^^datatype )" but I don;t find any examples n the Test/_Data directory.

  2. Is it possible to change the data-type of a property to Numeric after the item is created by merely replacing the value and leaving the property name the same? I have a lot of items that need tohave a certain property switched to Numeric. How would you approach this?

  3. I’m having trouble figuring out how to save a mapping from a spreadsheet. I can create a new reader and adjust the mapping and import, but when I go back to it, the mapping is back to the defaults.

Thank you very much.
–pbc

I’ve answered part of my question: The bulk Edit tool will let me change a property data-type to numeric.

The mapping form of the module is simpler than the csvimport one, because it relies on the headers. You can use dcterms:title @fr ^^numeric:integer to import number.
For the 3, the mapping cannot be saved currently, but if you use the good headers, it’s not needed. But it’s a possible improvement.

Thank you very much Daniel. Also thanks for incorporating the identifier relation feature to Bulk Uploader.

Would the following be good examples of rigorously typed column headings:

I_Count ^^numeric:integer
I_Celsius ^^numeric:double // or would it be ^^numeric:decimal ?
I_Date ^^numeric:date/time

Also, does having the types of properties specified in the custom-ontology plugin create any actual constraints with regard to the upload tools?

Thanks again.

Custom ontologies properties are properties like the other ones, like dcterms:title, etc.

For the datatype, it should be one of the datatypes that are managed by Omeka, here by the module Numeric DataType (https://github.com/omeka-s-modules/NumericDataTypes/blob/7e6ace5df3cada61709113a9e5d7c84e69c44f15/config/module.config.php#L28-L31).

Or in module RdfDatatype: https://github.com/Daniel-KM/Omeka-S-module-RdfDatatype/blob/master/config/module.config.php#L7-L18

Very helpful! Thank you.