The Publish to Azure Search publisher will post documents to an Azure Search index through 

https://<server>/indexes/<index>/docs/index?api-version=<apiVersion>

as described by Azure Search in the Bulk API.

On this page

Features


Features of the Publish to Azure Search publisher include:

  • You can customize the Azure Search feed by editing the Groovy script.
  • It is connector independent.
  • It runs from any machine with access to the Azure Search Cloud Service.


Limitations 


The publisher is subject to the following Azure Search Service limitations.

  • Index schema - Certain properties of your index schema can be set only once.
    • They cannot be updated in the future.Therefore, any schema updates that require re-indexing (such as changing field types) are not currently possible after the initial configuration. 
      Important:  Make sure your index has all of the required fields with the correct configuration before saving it.

  • Document keys - can only contain letters, digits, underscore ( _ ), dash ( - ), or equal sign ( = ).
  • Document field names - must start with a letter and contain only letters, digits, or underscore.

  • Batch size - the maximum is 16 MB. You can pass a batch of multiple documents to the Index API all at once, so the size limit per document depends on how many documents are in the batch. 
    • If the batch size exceeds 16 MB, the publisher will attempt to split the batch and truncate the content field of the document (if exists) to reduce the size. 
    • The maximum document size is 16 MB of JSON for a batch containing a single document, 
      Note:  If the document size still exceeds 16 MB after truncating the content field, publishing will not be possible without modifying the Groovy Transformation file. 

  • Maximums
    • 1000 documents per batch of index uploads, merges, or deletes.
    • 32 fields in the $orderby clause in the schema.
    • 32,766 bytes (32 KB minus 2 bytes) of UTF-8 encoded text for the field size for the Filterable, Sortable, Facetable, and Searchable fields.

  • Is not possible to delete all documents of an index directly.
    • If necessary, delete and re-create the index. 
    • If multiple full crawls are executed, data of previous crawls will still exist inthe index.
      Note:  The publisher is unable to delete all documents of an index directly (since it does not "know" the how the schema is configured in order to recreate it).

More information is available at Service Limits in Azure Search and Create an Azure Search Index Using the Azure Portal.


  • No labels