This article is continuation of Azure Search – Setup Blog storage for azure search. I recommend you to go through my previous article before you continue this.
Login to https://portal.azure.com.
Choose Azure Search from Azure Market Place. Refer screenshot below
Click Create and name the azure search and select subscription, resource group, location and Pricing tier. I selected Free. You can select as per your requirement
After provisioning the Azure search, click import data
In Import Data, you can have multiple options to import data. You have to choose “Azure Blob Storage” for ours.
Enter name as “msaztipslblindex”, select “Data to extract” as Content and metadata. You can select only metadata or storage only. Connection string you take it from the Blog storage account. Remember we have used connection string to connect blob storage in Azure blob storage explorer in my previous article. Enter container name as given in the blob storage. Leave other fields as default.
Click next and skip the cognitive service configuration and you will be landed in “Customize target index”. Here you can configure metadata field as searchable, Retrievable, Filterable, Sortable and Facetable.
The metadata fields are automatically populated based on the metadata added for each file (Files under blob container). I configured the metadata as in the screenshot below. Name the index as “azureblobindex”
Click Create an Indexer and submit.
After submitting the indexer, you can see the index details from Azure search overview section
Click the “myindexer” and monitor the Execution. Wait for completing the indexer. If not, hit Run button.
After successful completion of search index, you can see the result. Sometimes you may encounter “The blob ‘Azure_for_Architects.pdf’ has the size of 18348502 bytes, which exceeds the maximum size for document extraction for your current service tier”. Don’t panic, it is the limitation in Free tier. For now, you delete this document and run again. Otherwise, change the tier.
After removing the large file and re-index, we can see the status as like below.
Now we all set for search for indexed data. Click “Search Explorer” from the action menu
Select the indexer you want to search
Here is the sample query to select fields and filter by language
If you want to configure multiple values for a single metadata with filterable option, then you have to configure the metadata type as multiple in indexer. By doing this, first delete the indexer and again go to import data and select the existing data source. Skip to navigate “Customize target index” and change the filed type as Collection.
After changing the configuration, create the indexer and run.
Select query for multi value filter (String collection)
Use $select=Topics&$search.in(Topics, ‘Education’)