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. Support startsWith, contains, and endsWith wildcard style queries

    In simple query syntax wildcard style searches are supported. OData is great with its rich query support but is missing support for wildcard style searches like simple query syntax.

    To me the easiest path for implementation would be to support the startsWith, contains, and endsWiths functions in OData.

    178 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    17 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  2. HTTP POST based search query

    I can see only HTTP GET is supported.
    Hope I don't get a 414 (Request-URI Too Long) status if a URI is longer than the server can handle

    79 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    7 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →

    Azure Search now supports invoking Search and Suggest REST APIs via HTTP POST (previously you could only invoke them using GET). This is useful in scenarios where you have large filters and/or many facets and may be constrained by the 8 KB limit on the length of request URIs.

    The new endpoints are now available in the 2015-02-28-Preview API version, documented here:

    https://azure.microsoft.com/documentation/articles/search-api-2015-02-28-preview/#SearchDocs

    https://azure.microsoft.com/documentation/articles/search-api-2015-02-28-preview/#Suggestions

  3. Add OData syntax support for startswith, contains, and endswith queries

    This is essentially a duplicate of another suggestion: https://feedback.azure.com/forums/263029-azure-search/suggestions/6470849

    The original suggestion mentioned odata support, and while it was marked as complete, it was only added to lucene searches.

    I'm sure many of the 178 votes where for specifically odata support.

    58 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  4. Sort results by rank

    We had also requested sort results by rank

    54 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  5. Improve collection matching expressions

    Currently, collection matching only supports simple equality for the 'any' option and inequality for the 'all' option.

    I would like to be able match a collection against a set of values instead of testing a single match individually and then combining the results as shown below

    $filter=tags/any(t: t eq 'x') and tags/any(t: t eq 'y')

    Additionally, it would also be useful to have the capabilities to include simple boolean logic within the lambda to craft expressions that match a collection item against slightly more complex logic for example:

    $filter=tags/any(t: t eq 'x' or t eq 'y')

    50 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 →
  6. Add fuzzy logic on searches

    Please allow for fuzzy searches by enable that feature on the Lucene Simple Query Parser implementation (https://msdn.microsoft.com/en-us/library/azure/dn798920.aspx). It would greatly help to people who misspell some technical words or brand names.

    35 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  7. Support for searching of part of words

    It’s hard to estimate the importance and the frequency of querying on ‘sub-word-value’, because mostly they will be entering the full word

    Querying on ‘similar’ values (stemming) is not something that is very important in our scenarios, because customers will typically search on exact values (order numbers, customer names, etc)
    Things around spaces or not can off course be handy (was the address of that customer Waterstreet, Water str. Or Water Street, for example)
    i.e.,
    https://azsearchp1.cloudapp.net/svc3/indexes/fulltexttestindex/search?text=LOFT&skip=0&take=10 NOT returning matches, while
    https://azsearchp1.cloudapp.net/svc3/indexes/fulltexttestindex/search?text=LOFT1234&skip=0&take=10 is returning matches.

    17 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  8. Make it so SearchMode=All matches if stop words are present even when querying across multiple analyzers

    Not all of the documents in our corpus are localized so we must search the default language (English) and the user's locale with the provided user query. Queries that contain a stop word for either default or user language analyzer causes no results to be returned.

    Example:
    The documents contain 2 searchable fields (plus an ID field).
    Title_EN -> English Analyzer
    Title_ES -> Spanish Analyzer

    A document exists with the following values:
    Title_EN -> "The best document"
    Title_ES -> "Un documento excelente"

    When a user inputs a query, we search on Title_EN and Title_ES. If the user provides the query…

    9 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 →
  9. configurable (or unlimited) hit highlights

    Per https://docs.microsoft.com/en-us/rest/api/searchservice/search-documents "Azure Search returns only up to 5 highlights per field. This limit is not configurable" >> this should be changed. Either by offering a configurable threshold, by pagination or any other way. We'd need to be able to access all hit results and not just a partial subset

    9 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  1 comment  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  10. Allow specifying query time analyzer

    Issueing a query with searchMode=all against multiple fields that use different analyzers may lead to unexpected results.

    Let’s take an index with two fields where one is analyzed with English Lucene analyzer, and the other with standard (default) analyzer.

    {
    "fields": [
    {"name": "docId", "type": "Edm.String", "key": true, "searchable": false},
    {"name": "field1", "type": "Edm.String", "analyzer": "en.lucene" },
    {"name": "field2", "type": "Edm.String" }
    ]
    }

    Let’s add these two documents:

    {
    "value": [
    {
    "docId": "1",
    "field1": "Waiting for a bus",
    "field2": "Exploring cosmos"
    },
    {
    "docId": "2",
    "field1": "Run to the hills",
    "field2": "run for your life"
    }
    ]
    }

    8 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 →
  11. Return all the highlights and not just 5

    Today Azure Search returns only the first 5 highlight for each field. It would be great to get all the highlights.

    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 →
  12. Apply different search terms to different sets of fields

    When I have multiple words to search on, and multiple fields to search, I would like to be able to define which search words should be used to search on which search field.

    To clarify take a look at this request:

    https://[oursearch].search.windows.net/indexes/jobs/docs?$count=true&search=admin* london*&searchMode=all&searchFields=jobTitle,organisation,description,location&api-version=2015-02-28

    This search terms are admin london. The fields I want to search on are title,organisation,description,location. I know london is a location so I only want to search in the location field for London but should search for "admin" in the other three fields (title,organisation,description).

    4 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 →
  13. Provide searching of phrases

    Without this capability, azure search may not be able to effectively search "dense" content (e.g., msw). Another example is an index I have with scientific journal docs. When I search for "back pain", the best I could do (please correct me if I'm wrong) is:

    text=back,pain&textmode=all

    Even though each query result contains "back" and "pain", most of them are irrelevant and have nothing to do with "back pain".

    Boosting the rank scoring of finding multiple adjacent terms could help but the query results would still mostly be irrelevant.

    4 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 →
  14. Hit Highlighting

    We are working out the search frontend and one of the things we really would love to see is getting back a fragment of a matching search when searching on a text field.
    For example, we do a search on text, in documents that can be rather large. We don’t want to download the full message body for the 10 matching results, but it would be good if we could get the part of the document where the matching search text is found. (just like bing/google shows when searching on a term)
    Right now we retrieve all messages and then…

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  15. Customize highlights

    Right now it returns matched text between <em> tags. Instead would it be possible to return it as start and end postion value pairs in a given field. We can convert the current output in that format but wanted to know if there are customization options in the service itself.

    We are finding that consuming the current format is not very useful (<em> means italics, we want to show matched text in bold and in different color) and quite heavy to render (having html controls rather than native). We are currently looking to parse the string and finding start/end index…

    2 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 →
  16. Support geolocalization / spatial search

    Queries like:

    - Records within a range of X kilometers/miles from a specific point.
    - Records within a field (multiple coordinates like a country, ... - just like polygons in google maps)
    - Order by distance
    - Support for facets

    2 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  17. Return the number of results for each query

    It's helpful for the client to know the total number of results that match a query, so e.g.,

    - it could be reported to the user as feedback and if it's too large, the user could then enter more constraints to narrow the search, etc.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
  18. provide query performance feedback on request

    How long did the query run for? (Plus any other appropriate measurements). This would help benchmarking, testing and maintenance of a search service by helping to identify the slowest queries.

    1 vote
    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 →

    Hi Harold, we do return elapsed time for a query in an HTTP response header. Originally this wasn’t documented, but we updated the docs. There’re probably many other measurements and analytics we could provide – if you have specific ideas, please submit there!
    Thanks you,
    Eugene Shvets
    Developer
    Azure Search

  19. Extend the Azure Search indexer for Azure tables to allow a search index to be sourced from multiple tables

    Would it be possible to extend the Azure Search indexer for Azure tables to allow a search index to be sourced from multiple tables? I have a search that, currently, I build from a ProductAnnotation and a ProductWholesale table (both share a common Id for lookup and this Id is stored in the search index).

    1 vote
    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 →

    Hi Keith,

    This is already possible in a slightly different form: you can create multiple datasource / indexer pairs all writing into the same index. As long as you stagger their schedules so that the indexers are unlikely to overlap on the same document at the same time, you’ll get the functionality this suggestion asks for.

    Hope this helps!
    Your Azure Search team

  20. API for managing star/popularity/reputation ratings

    Require API for managing star/popularity/reputation ratings

    0 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    completed  ·  0 comments  ·  Query - Search  ·  Flag idea as inappropriate…  ·  Admin →
← Previous 1
  • Don't see your idea?

Feedback and Knowledge Base