Azure Search

Azure Search is a search-as-a-service solution that allows developers to incorporate a sophisticated search experience into web and mobile applications without having to worry about the complexities of full-text search and without having to deploy, maintain or manage any infrastructure

  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. Ability to identify from what data source the index document was created

    While indexing multiple data sources(azure tables) into same index I'd like to identify what data source produces specific document. This will allow to reference back to the system that can show original document.

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    under review  ·  0 comments  ·  Crawlers  ·  Flag idea as inappropriate…  ·  Admin →
  2. Support correlated search on complex collections

    Now that Azure Search supports complex types, it would be useful to be able to combine full-text search and strict Boolean filters when querying over complex collections. Currently this is not possible -- search.ismatch()/ismatchscoring() are not allowed in lambda expressions, and using sub-fields of complex collections in full Lucene queries has uncorrelated semantics.

    For example, if you want to search for hotels that have deluxe rooms with city views, none of these gives you exactly what you want:

    search=Rooms/Description:(city views) AND Rooms/Type:(Deluxe Room)

    search=Rooms/Description:(city views)&$filter=Rooms/any(room: room/Type eq 'Deluxe Room')

    $filter=search.ismatchscoring('city views') and Rooms/any(room: room/Type eq 'Deluxe Room')

    Instead, what you'd…

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  3. provide top N DISTINCT values from suggester API

    At the moment the suggester API return the top N results, but if I want use this feature on a field that contains duplicate values, this request could show me N identical values.
    For example, my index contains the searchable fields address and city (address as the key) and the city field contains only the cities "Parma" and "Pavia". I want to add autocomplete on the city field on the web app. The suggester API with TOP 5 could give me 5 identical values with the city "pavia". But I like to see both the cities in the TOP 5.…

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  REST API  ·  Flag idea as inappropriate…  ·  Admin →
  4. "Script as JSON" on Index Blade

    On the index blade in the portal I like to be able to create my index with the UI for speed. I would also like to have a script option that I can then use for repeatability.

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Portal  ·  Flag idea as inappropriate…  ·  Admin →
  5. Presto based managed query service

    I want Presto based managed query service like AWS Athena and GCP Bigquery.

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  6. Query Keys Should Be Time-Boxed

    When creating a query key there should be a parameter to say "For How Long". If you look at Blob Storage when you retrieve a SharedAccessBlobPolicy(). I may only want to grant access to the index for a certain period.

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    under review  ·  0 comments  ·  Security  ·  Flag idea as inappropriate…  ·  Admin →
  7. Populate json string in Azure SQL table directly into Azure Search Index such that it is queryable on json attributes

    I have a column in my Azure SQL table that contains json (not array of strings) is there a direct way to populate that into azure search index so that it is queryable (filterable on the json attributes)

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Crawlers  ·  Flag idea as inappropriate…  ·  Admin →
  8. Support multi point Edm.GeographyPoint search

    It would be great if you could sort by nearest where documents could have an array of Edm.GeographyPoints.
    We are currently duplicating documents for every different Edm.GeographyPoint.

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  9. API to check availability of documents for search

    Currently there is short delay before documents become available for search after they were created or updated.

    For better consistency a blocking call that returns after all of the documents updates, creates, etc have been indexed and are ready for search would be great.

    Or provide status api for the set of documents updates, creates, etc sent to understand if they have all been indexed and are ready for search

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Indexing  ·  Flag idea as inappropriate…  ·  Admin →
  10. Fully covered OData filter functions

    Needed for use case like `$filter=length(title) gt 64` and/or `$filter=toupper(firstname)`

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  11. Indexer priority

    We are using CosmosDb as our DB for apps search and we need to have overrides to change apps metadata on-the-fly. This allows us to hide some apps from clients or change privacy statements without the need for the developer to re-upload their work.

    To achieve this, we use a CosmosDb collection for the app data and another for the overrides.
    In order to provide our clients with search, we use Azure Search and to keep indexes up-to-date, we wanted to use Indexers.

    It is indeed possible at the present time to have multiple data sources get uploaded to the…

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Crawlers  ·  Flag idea as inappropriate…  ·  Admin →

    Thank you for your feedback. While it is unlikely we’ll address this suggestion in the near future, we’ll reassess based on the number of votes it receives.

    Also, you may want to start a stack overflow discussion on this topic… may be we can think of other mechanisms (besides indexer priorities) that allow you to get to an acceptable outcome.

    Thanks,
    Luis Cabrera
    Azure Search Product Team

  12. Adding hard delete policy to all of Indexer

    Can we add hard delete policy to all of indexer as same as soft delete policy?
    Some senario in custmer immideiately want to refect source change result. We know we can do it by making push base approach, but it's much easier and simplyer to ensure single data update way.

    Cosmos:
    https://docs.microsoft.com/en-us/azure/search/search-howto-index-cosmosdb#indexing-deleted-documents

    Azure Table:
    https://docs.microsoft.com/en-us/azure/search/search-howto-indexing-azure-tables#incremental-indexing-and-deletion-detection

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Indexing  ·  Flag idea as inappropriate…  ·  Admin →
  13. Support a DFS Query Then Fetch Search Type

    Support DFS Query then Fetch search type in order to get more accurate scores across shards. (Equivalent to search_type=dfs_query_then_fetch in Elasticsearch.)

    https://www.elastic.co/blog/understanding-query-then-fetch-vs-dfs-query-then-fetch

    https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-search-type.html

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  14. Expose write metrics not just read metrics through the portal

    At the moment the portal only shows read metrics (QPS, Latency, Throttled) but this is only half the picture of what can be impacting an Azure Search Service.

    Write metrics (e.g. index updates per second) are just as important especially for high volume re-indexing operations,

    To get anything like this needs an export of logs, which goes to an unusable Json format (i.e. CSV is more useful for analysis) and/or a PowerBI account.

    Neither of which are particularly useful for this scenario where we just need a quick review of what's actually going on with our index.

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Indexing  ·  Flag idea as inappropriate…  ·  Admin →
  15. Index blob container metadata

    We're trying to set up an indexer for groups of related blobs (e.g. multiple formats of a single image file) using containers. Unfortunately, it seems like the blob indexer doesn't extract metadata from containers - only the blobs themselves. As a workaround, we can duplicate metadata across all blobs in a container, but it would be nice if the indexer supported indexing container metadata directly.

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Indexing  ·  Flag idea as inappropriate…  ·  Admin →
  16. Semantics/tag clouds/concepts

    Semantics/tag clouds/concepts

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  17. Support indexing JSON-LD embedded in HTML documents

    Support indexing structured data (e.g. schema.org) in HTML documents.

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Crawlers  ·  Flag idea as inappropriate…  ·  Admin →
  18. increase the number of available free search services

    We would ideally deploy our entire system using a script to provision and configure all the necessary Azure resources. This is helpful for clean per client deployments and also for qa/test deployments, however being limited to 1 free search service means having to have an expensive basic tier search service per qa/test deployment, or having to share a single shared free tier search service for qa/test/dev. Ideally we'd just be able to deploy 3 free tier search services.

    An alternative to this would be a more finely graduated billing scheme based on usage (currently the cost of the search service…

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Pricing and Quotas  ·  Flag idea as inappropriate…  ·  Admin →
  19. SAT support for query tokens

    As with Blog/containers it's possible to have a SAT token and time limit it. This would be a great benefit for azure search too, to allow query keys with the same concept.

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  20. Azure search Suggestions dose not filter duplicate items

    Azure search Suggestions dose not filter duplicate items. This is Suggestions official document https://docs.microsoft.com/en-us/rest/api/searchservice/suggestions.

    Please remove useless duplicate items. Duplicate items are useless in "search-as-you-type" scenario.

    6 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Feedback and Knowledge Base