To track deletions, a soft delete policy is required in the data source. Something Azure Storage supports. However, it is not compatible with Azure Search. This should be an out-of-the-box feature in Azure Search.

I would like to suggest to include the function of extract the graph or image in the documents file like PDF, words and PowerPoint .

Could we have a configurable property to set the minimum TLS version that search service will accept (as per Redis Cache and other resource types). With the removal of 1.0 and 1.1 on 6th Feb 2020 it would have been useful to set in advance to ensure that all endpoints talking to it were using 1.2 - I would assume that this same scenario will exist in the future when new versions make 1.2+ obsolete.

Hi Team,
We are using Azure search for 2 yaers now. We have a requirement where we want to show auto suggester in some order. Say if we have 3 fields as auto suggestor[Name, title, city]. We want that the value from Name should come first then the remaining ones

while using "Simple query syntax in Azure Cognitive Search" how can i search for text containing spaces while using wild card?

I am using suggestion API in my application .If there are matching results more than 100 there is no option to identify the match results count.

I want to search by sending partial or full -text along with other parameters. and also is there any option for Phrase search with full text or partial text.

I wrote a technical review of the Knowledge Store preview feature and wanted to share it with you. See attached.

Implement create/delete/update indexes via the Azure CLI in addition to Azure REST Api capabilities. You can create the search service itself of course, but i see no way to create the index or indexers (e.g.) using Cli

When defining datasource of Type Azure Storage Table in the Query field, only options in the examples I see is like (PartitionKey eq '0') or something like that, but there is no example which would allow us to select columns along with the filter conditions.

Currently, Search Explorer provides the full URL of the GET request for the Search REST API. To use a query from Search Explorer in the SDK, you have to parse out the parameters by hand, and sometimes URL decode them.

It would be better if Search Explorer could provide code snippets for each supported language (C# and F# for now, Java, JavaScript/TypeScript, and Python when those SDKs are available) that show how to call the Search API using the parameters that the Portal user has typed in.

Provide ability to get access to and filter full indexer status history.

I have an object class that contains properties that I would not want to persist in elastic but needs them (inherit by an interface)
I get an error since these fields are not mapped.
How can I ignore these fields

I would like to get metrics for our azure search machines cpu in order to get a clear picture of put resources usage

AWS offers up to 52% discount when committing to a 3 year reserved instance for AWS ElasticSearch.

There is no reserved pricing option for Azure Search.

Please provide reserved pricing, similar to Cosmos DB/Azure SQL, at a discount comparable to AWS.

The current SLA is only 99.9%. That is out of sync with typical enterprise SLAs of 99.99% that our own applications must abide to for our customers.

Hence, this lower than typical SLA endangers our own SLAs. Please improve to 99.99%

I came across this forum a bit later after already posting by query on github here https://github.com/MicrosoftDocs/azure-docs/issues/41726

Thoughts?

Correctly it is not possible to rename index or indexer once it is created. It is difficult to admin or developer to create index or indexer from the scratch. Please provide some option on indexer should be update. Also provide some option to create indexer using existing one.

We have been getting the following error when indexing

The request is invalid. Details: parameters : An unexpected 'PrimitiveValue' node was found when reading from the JSON reader. A 'StartArray' node was expected

It would be nice to know what field it failed on in the response and even better which specific object it was trying to work with and the value on that field.