Azure Search Hierarchical Search Field - search

I am evaluating Azure Search for a project. The MSDN articles are having only Flattened schema structure. Below is an example scenario i am looking at.
The below is "Project" class having reference to List of "Question" class. And "Question" has it's own set of fields
public class Project
{
public Guid Id
{
get;
set;
}
public string Owner
{
get;
set;
}
public string Title
{
get;
set;
}
public List<Question> QuestionList
{
get;
set;
}
public bool Disable
{
get;
set;
}
}
public class Question
{
public Guid Id { get; set; }
public string Title { get; set; }
public string Description { get; set; }
}
Below will be Index Schema for Project
Id - Edm.String
Owner - Edm.String
Title - Edm.String
QuestionList - Collection(Edm.String)
Questions
Is it possible to specify each item in QuestionList has hierarchical data?
Is it possible to Search only inside "Title" of "Question"?
The only possibility i see is to create Index for "Question" separately and use it
The only point i see in MSDN relevant is the below paragraph and i can't make much sense of it
Levels in faceted navigation
As noted, there is no direct support for nesting facets in a
hierarchy. Out of the box, faceted navigation only supports one level
of filters. However, workarounds do exist. You can encode a
hierarchical facet structure in a Collection(Edm.String) with one
entry point per hierarchy. Implementing this workaround is beyond the
scope of this article, but you can read about collections in OData by
Example.

For your first question, Azure Search does not allow for hierarchical datatypes, and to search you would need to flatten the data as you did for the QuestionList field which you created as a Collection. If you were asking how to also filter results based on items in this Collection, you can do that using OData Expressions such as $filter=QuestionList/any(t: t eq 'Question1') (https://msdn.microsoft.com/en-us/library/azure/dn798921.aspx)
I think for your second question, you were interested searching only in "Title" or "Question", correct? For this, you can use the SearchFields parameter (https://msdn.microsoft.com/en-us/library/azure/dn798927.aspx).
Liam

Related

Get request with multiple inputs in ServiceStack API

I am creating a endpoint that accepts multiple parameters. I was wondering what the proper way of doing this in ServiceStack is, as I know that the routes can be defined like this:
[Route("/Cars/{EquipmentIds}/{ManufacturerIds}/{ColourIds}")]
But does not that mean that EquipmentIds has to be defined before ManufacturerIds and ColourIds, otherwise the API will interpret it incorrectly?
I would like to specify the parameters I use, and then not include the rest when they are not used.
A unclean url would then look like this ?EquipmentIds=1&ColourIds=1
I found the following solution, but this one was from 2011
Multiple Optional Parameters with ServiceStack.Net.
Is there a new solution for this, or is that still the proper way of doing things?
The URL should be a "resource identifier" where any Query String arguments are modifiers to that resource request.
So you shouldn't put complex types in the /path/info which are unlikely to form part of Resource Identifier for that route.
Your route should be self-describing as to what it's a resource identifier of. Stuffing a an anonymous bunch of ids in the URL doesn't make it a clean URL, it still needs to be semantically correct and self-describing at what the different path components contain.
It's hard to know the right Route to use when it's not clear what Service this is used for, but if this was a Car Search Service the appropriate URL would be something like:
/cars/search?EquipmentIds=1,2,3&ManufacturerIds=4,5,6&ColourIds=7,8,9
Just as when you use a Search Service like Google, they don't try to pack everything in the route, which is only used to identify you're making a search request. Everything else including the search query is added to the query string, e.g;
https://www.google.com/search?q=test
Which in ServiceStack would just be defined as:
[Route("/cars/search")]
public class SearchCars
{
public List<int> EquipmentIds { get; set; }
public List<int> ManufacturerIds { get; set; }
public List<int> ColourIds { get; set; }
}
Or it can be easily auto implemented in AutoQuery with:
[Route("/cars/search")]
public class SearchCars : QueryDb<Car>
{
public List<int> EquipmentIds { get; set; }
public List<int> ManufacturerIds { get; set; }
public List<int> ColourIds { get; set; }
}

Auto Query search

How can I do an auto query with "or" operation like -
http://localhost/rockstars/first_name=MikeORlast_name=Smith
See the Section on changing AutoQuery Behavior where you can change the behavior of every field by annotating Services with [Query(QueryTerm.Or)], e.g:
[Query(QueryTerm.Or)]
public class QueryRockstars : QueryBase<Rockstar> {}
Otherwise if you only want to some of the fields to have "OR" behavior you can decorate them individually with:
public class QueryRockstars : QueryBase<Rockstar>
{
[QueryField(Term=QueryTerm.Or)]
public string FirstName { get; set; }
[QueryField(Term=QueryTerm.Or)]
public string LastName { get; set; }
}
However from a Services design POV applying different behavior is not recommended since ideally each of the fields should have the same semantics.
With the above Request DTO's you can now query it with:
/rockstars?FirstName=Mike&LastName=Smith

ServiceStack AutoQuery, Multiple IJoin

In my example I have the following database structure. Order has many OrderLine, which has one Product.
I am trying to return the following DTO:
public class OrderLineDto {
public int Id { get; set; }
public int Quantity { get; set; }
public string OrderType { get; set; }
public string ProductName { get; set; }
}
This should be possible by use of the following Query Route:
[Route("/orderlines")]
public class FindOrderLines : QueryBase<OrderLine, OrderLineDto>,
IJoin<OrderLine, Order>,
IJoin<OrderLine, Product>
{ }
What I am trying to do here is join OrderLine in both directions to bring in Type from Order, and Name from Product and return it in an OrderLineDto.
I am able to do these things individually by only using one IJoin, however AutoQuery appears only to use the first IJoin interface declaration, and does not perform the second join.
If I attempt to do a join like this: IJoin<OrderLine, Order, Product>
I get the following exception: Could not infer relationship between Order and Product
Is it possible to achieve what I am trying to do here with auto query or should I go back to writing standard REST services, abandoning AutoQuery?
I have submitted a pull request to ServiceStack which will now allow this behavior.
https://github.com/ServiceStack/ServiceStack/pull/955

Orchard: how to persist a record without content

Allright, this should be fairly easy.
I would like to persist some records for my module in Orchard (1.7.2) without those records being also a ContentPartRecord.
In other words, I would like to be able to persist in DB the following objects:
public class LogItemRecord
{
public virtual string Message { get; set; }
}
..which is already mapped on to the db. But notice that this class is not derived from ContentPartRecord, as it is most certainly not one.
However, when I call IRepository instance's .Create method, all I get is a lousy nHibernate exception:
No persister for: MyModule.Models.LogItemRecord
...which disappears if I do declare the LogItem record as having been inherited from ContentPartRecord, but trying to persist that, apart from being hacky-tacky, runs into an exception of its own, where nHibernate again justly complains that the Id value for the record is zero, though in not so many words.
So... how do I play nicely with Orchard and use its API to persist objects of my own that are not ContentParts / ContentItems?
I'm running 1.7.3 (also tested in 1.7.2) and have successfully been able to persist the following class to the DB:
public class ContactRecord
{
public virtual int Id { get; set; }
public virtual string Name { get; set; }
public virtual string JobTitle { get; set; }
public virtual string Email { get; set; }
public virtual string Phone { get; set; }
}
Here are the relevant lines from Migrations.cs
SchemaBuilder.CreateTable(
typeof(ContactRecord).Name,
table => table
.Column<int>("Id", col => col.Identity().PrimaryKey())
.Column<string>("Name")
.Column<string>("JobTitle")
.Column<string>("Email")
.Column<string>("Phone")
);
I'm going to assume that the code you've shown for LogItemRecord is the complete class definition when making the following statement...
I think that any Record class you store in the DB needs an Id property, and that property should be marked as Identity and PrimaryKey in the table definition (as I've done above).
When you create a *Record class which inherits from ContentPartRecord and setup the table like
SchemaBuilder.CreateTable(
"YourRecord",
table => table
.ContentPartRecord()
// more column definitions
);
then you get the Id property/PK "for free" by inheritance and calling .ContentPartRecord() in the Migration.
See the PersonRecord in the Orchard Training Demo Module for another example of storing a standard class as a record in the DB.

Including a base member doesn't seem to work in Entity Framework 5

here are my entities:
public abstract class ResourceBase
{
[Key]
int Id { get; set; }
[ForeignKey("Resource")]
public Guid ResourceId { get; set; }
public virtual Resource Resource { get; set; }
}
public class Resource
{
[Key]
public Guid Id { get; set; }
public string Type { get; set; }
}
public class Message : ResourceBase
{
[MaxLength(300)]
public string Text { get; set; }
}
And then my query is something like this:
var msgs = messages.Where(x=>x.Id == someRangeOfIds).Include(m=>m.Resource).Select(x => new
{
message = x,
replyCount = msgs.Count(msg => msg.Id = magicNumber)
});
I am running this with proxy creation disabled, and the result is all the messages BUT with all the Resource properties as NULL. I checked the database and the Resources with matching Guids are there.
I drastically simplified my real life scenario for illustration purposes, but I think you'll find you can reproduce the issue with just this.
Entity Framework 5 handles inherited properties well (by flattening the inheritence tree and including all the properties as columns for the entity table).
The reason this query didn't work was due to the projection after the include. Unfortunately, the include statement only really works when you are returning entities. Although, I did see mention of a solution which is tricky and involves invoking the "include" after the shape of the return data is specified... If anyone has more information on this please reply.
The solution I came up with was to just rephrase the query so I get all messages in one query, and then in another trip to the database another query that gets all the reply counts.
2 round trips when it really should only be 1.

Resources