I am using the Preview of the Azure Search Blob Indexer. All of the data that is indexed is contained in the metadata of the blobs. I have therefore set AzureSearch_SkipContent to 'true' to prevent content indexing.
I am having trouble getting DateTimeOffset fields to work, as I cannot figure out the correct format to store the DateTimeOffset value as.
When I create the indexer, I get an error response (400 Bad Request) with the response body holding the following content:
{"error":{"code":"","message":"Column 'ProcessDT' is of type String that is not compatible with the field of type Edm.DateTimeOffset in the index"}}
I have tried various string formats (the value on the blob I tested is
2015-01-03T11:13:00.000+01:00
so it should correspond to a format that is compatible to OData 4.0 and therefore azure search as stated here: https://msdn.microsoft.com/en-us/library/azure/dn798938.aspx
converting strings into DateTimeOffset field values is not currently supported. We recognize this is a painful limitation, and will address this in the near future. As a partial workaround, you can index those values as strings (so you can format them as you wish and still use them for search and filtering). Thanks for bringing this up!