I'm trying to define an index in RavenDb that uses the output of another index as it's input but I can't get it to work.
I have the following entities & indexes defined.
SquadIndex produces the result I expect it to do but SquadSizeIndex doesn't even seem to execute.
Have I done something wrong or is this not supported?
class Country
{
public string Id { get; private set; }
public string Name { get; set; }
}
class Player
{
public string Id { get; private set; }
public string Name { get; set; }
public string CountryId { get; set; }
}
class Reference
{
public string Id { get; set; }
public string Name { get; set; }
}
class SquadIndex : AbstractIndexCreationTask<Player, SquadIndex.Result>
{
public SquadIndex()
{
Map = players => from player in players
let country = LoadDocument<Country>(player.CountryId)
select new Result
{
Country = new Reference
{
Id = country.Id,
Name = country.Name
},
Players = new[]
{
new Reference
{
Id = player.Id,
Name = player.Name
}
}
};
Reduce = results => from result in results
group result by result.Country
into g
select new Result
{
Country = g.Key,
Players = g.SelectMany(x => x.Players)
};
}
internal class Result
{
public Reference Country { get; set; }
public IEnumerable<Reference> Players { get; set; }
}
}
class SquadSizeIndex : AbstractIndexCreationTask<SquadIndex.Result, SquadSizeIndex.Result>
{
public SquadSizeIndex()
{
Map = squads => from squad in squads
select new Result
{
Country = squad.Country,
PlayerCount = squad.Players.Count()
};
Reduce = results => from result in results
group result by result.Country
into g
select new Result
{
Country = g.Key,
PlayerCount = g.Sum(x => x.PlayerCount)
};
}
internal class Result
{
public Reference Country { get; set; }
public int PlayerCount { get; set; }
}
}
No, you can't. The output of indexes are not documents to be indexed.
You can use the scripted index results to chain indexes, but that isn't trivial.
Related
I have the following structure:
public class Order
{
public string Id { get; set; }
public string CustomerId { get; set; }
public decimal Amount { get; set; }
public DateTimeOffset CreatedDate { get; set; }
}
public class Customer
{
public string Id { get; set; }
public string Name { get; set; }
}
I want to export all customers (RavenDB Stream) with order turnover and last ordered date.
I do already have an index (Customers_ByTurnover) which outputs this data (map=Orders, reduce by CustomerId). Although this does only list customers which have already ordered something.
I need an index for all Customers and load these details into each row.
Here is the code I want to write (the Query method is pseudo and doesn't really exist):
public class Customers_ByOrders : AbstractIndexCreationTask<Customer, Customers_ByOrders.Result>
{
public class Result
{
public string Id { get; set; }
public string Name { get; set; }
public decimal Turnover { get; set; }
public DateTimeOffset? LastOrderedDate { get; set; }
}
public Customers_ByOrders()
{
Map = items => items.Select(item => new Result()
{
Id = item.Id,
Name = item.Name,
Turnover = Query<Order>().Where(x => x.CustomerId == item.Id).Sum(x => x.Amount),
LastOrderedDate = Query<Order>().Where(x => x.CustomerId == item.Id).Select(x => x.CreatedDate).OrderByDescending(x => x).FirstOrDefault()
});
}
}
How can I solve this issue?
You cannot create a query inside an index, to get the desired info you will have to create a map-reduce index on the Orders collection, group by on CustomerId and in the reduce function apply Sum() on Amount field and order the LastOrderedDate. To get the Name name field you will have to use LoadDocument extension.
public class Customers_ByOrders : AbstractIndexCreationTask<Orders, Customers_ByOrders.Result>
{
public class Result
{
public string Id { get; set; }
public string Name { get; set; }
public decimal Turnover { get; set; }
public DateTimeOffset? LastOrderedDate { get; set; }
}
public Customers_ByOrders()
{
Map = orders => from o in orders
select new Result
{
Id = o.CustomerId,
Turnover = o.Amount,
LastOrderedDate = o.CreatedDate
};
Reduce = results => from result in results
group result by result.Id
into g
select new Result
{
Id = g.Key,
Turnover = g.Sum(x => x.Turnover),
LastOrderedDate = g.OrderByDescending(x => x.LastOrderedDate).Select(x => x.LastOrderedDate).FirstOrDefault()
};
}
}
Query Code:
var query = session.IndexQuery<App_OrgSearch.IndexResult, App_OrgSearch>();
var organizationUnitResults = query.Statistics(out stats)
.Skip(0)
.Take(5)
.AsProjection<Org>().ToList();
public static IRavenQueryable<TResult> IndexQuery<TResult, TIndex>(this IDocumentSession session)
where TIndex : AbstractIndexCreationTask, new()
{
return session.Query<TResult, TIndex>();
}
App_OrgSearch is the index I defined as below:
public class App_OrgSearch : AbstractIndexCreationTask<Org, App_OrgSearch.IndexResult>
{
public class IndexResult
{
public string Id { get; set; }
public string BusinessName { get; set; }
public string ShortName { get; set; }
public IList<string> Names { get; set; }
public List<string> PhoneNumbers { get; set; }
public List<OrganizationUnitPhone> OrganizationUnitPhones { get; set; }
}
public App_OrganizationUnitSearch()
{
Map = docs => from doc in docs
select new
{
Id = doc.Id,
Names = new List<string>
{
doc.BusinessName,
doc.ShortName,
},
BusinessName = doc.BusinessName,
ShortName = doc.ShortName,
PhoneNumbers = doc.OrganizationUnitPhones.Where(x => x != null && x.Phone != null).Select(x => x.Phone.Number),
};
Indexes.Add(x => x.Names, FieldIndexing.Analyzed);
}
}
I have 27 records in database. I want to take 5, but after query, all 27 records are returned. Why does Take function not work?
Your sample code seems wrong.
var query = session.IndexQuery<App_OrgSearch.IndexResult, App_OrgSearch>();
var organizationUnitResults = organizationUnitsQuery.Statistics(out stats)
What is organizationUnitsQuery ? You have the query as query, but there is no IndexQuery method on the session
Assuming the following class:
public class Thing
{
public string Id { get; set; }
public string Title{ get; set; }
public string KeyId { get; set; }
public CultureInfo Culture { get; set; }
}
and the following result class:
public class ProjectedThing
{
public string Id { get; set; }
public string Title{ get; set; }
public string KeyId { get; set; }
public CultureInfo Culture { get; set; }
public IEnumerable<Thing> Things { get; set; }
}
How can I build an index that holds the result class?
The closest I've come is with the following index definition:
public class ProjectedThings : AbstractIndexCreationTask<Thing,ProjectedThing>
{
public ProjectedThings()
{
Map = docs => from doc in docs
select new
{
Title = doc.Title,
KeyId = doc.KeyId,
Culture = doc.Culture,
Things = new[] {
new Thing{
Id = doc.Id,
Title = doc.Title,
KeyId = doc.KeyId,
Culture = doc.Culture,
TitlePluralized = doc.TitlePluralized
}
}
};
Reduce = results => from r in results
group r by r.KeyId into g
select new
{
Title = g.FirstOrDefault(x => x.Id == x.KeyId).Title,
KeyId = g.Key,
Culture = g.FirstOrDefault(x => x.Id == x.KeyId).Culture,
Things = from thing in g.SelectMany(x => x.Things).Where(x => x.Id != x.KeyId)
select new
{
Id = thing.Id,
Title = thing.Title,
KeyId = thing.Key,
Culture = thing.Culture
}
};
}
}
That's almost doing the trick, but I can't collect the Title, KeyId, and Culture in the reduction. Only the Things property is being populated.
Look at your code:
g.FirstOrDefault(x => x.Id == x.KeyId)
I don't understand this, but this is likely to be always false.
You probably want:
g.FirstOrDefault().Title,
I have two documents Ticket and MenuItem i have created index with TransformResults but problem is i am getting null value for Loaded document in transform
public class Ticket
{
public int ID { get; set; }
public int ItemId { get; set; }
public string ItemName { get; set; }
public int Price { get; set; }
}
public class MenuItem
{
public int ID { get; set; }
public string ItemName { get; set; }
public string PriceCategory { get; set; }
}
i have created a index like
public class TicketItemGross : AbstractIndexCreationTask<Ticket, TicketItemGross.TicketItemDetails>
{
public class TicketItemDetails
{
public string ID { get; set; }
public string ItemId { get; set; }
public string ItemName { get; set; }
public int Price { get; set; }
public string PriceCategory { get; set; }
}
public TicketItemGross()
{
Map = docs => from doc in docs
select new
{
ID = doc.ID,
ItemId=doc.ItemId,
ItemName=doc.ItemName,
Price=doc.Price
};
TransformResults = (database, docs) => from m in docs
let d = database.Load<MenuItem>(m.ID)
select new
{
ID = m.ID,
ItemId = m.ItemId,
ItemName = m.ItemName,
Price = m.Price,
PriceCategory=d.PriceCategory
};
}
}
and the problem is that when i query data. I get null for PriceCategory but for all other fields i get correct value
here is query
IEnumerable<TicketItemGross.TicketItemDetails> list;
using (var session = store.OpenSession())
{
list = session.Query<TicketItemGross.TicketItemDetails, TicketItemGross>();
}
This is happening because you are using integer IDs. When you call database.Load in your transform, you'll need to manually convert it to a string ID.
database.Load<MenuItem>("MenuItems/" + m.ID)
This is one of several places where Raven gets confused if you use integer or guid IDs. If you use string ids, you won't have this problem.
Also, you might consider using a results transformer instead. They are easier than index transformers, which are now obsolete.
I'm trying to use the TransformResults feature, and I can't get it to work. I'm not totally sure I understand this feature, perhaps there is another way to solve this problem. What I want is just the Id from the Order and the email addesses from the Customer and the Entrepreneur. I am happy for all tips that can take me in the right direction. Here is my code.
Document
public class OrderDocument
public string Id {get; set }
public EntrepreneurInfo EntrepreneurInfo { get; set; }
public CustomerInfo CustomerInfo { get; set; }
public OrderStatus CurrentOrderStatus { get; set; }
}
Info classes
public class EntrepreneurInfo
{
public string EntrepreneurDocumentId { get; set; }
public string Number { get; set; }
public string Name { get; set; }
}
public class CustomerInfo
{
public string CustomerDocumentId { get; set; }
public string Number { get; set; }
public string Name { get; set; }
}
The info classes are just subsets of a Customer and Entrepreneur documents respectively.
The Customer and Entrepreneur documents inherits from a base class ( AbstractOrganizationDocument) that has the EmailAddress property.
My Index
public class OrdersApprovedBroadcastingData :
AbstractIndexCreationTask<OrderDocument, OrdersApprovedBroadcastingData.ReduceResult>
{
public OrdersApprovedBroadcastingData()
{
this.Map = docs => from d in docs
where d.CurrentOrderStatus == OrderStatus.Approved
select new
{
Id = d.Id,
CustomerId = d.CustomerInfo.CustomerDocumentId,
EntrepreneurId = d.EntrepreneurInfo.EntrepreneurDocumentId
};
this.TransformResults = (db, orders) => from o in orders
let customer = db.Load<CustomerDocument>(o.CustomerId)
let entrepreneur = db.Load<EntrepreneurDocument>(o.EntrepreneurId)
select
new
{
o.Id,
o.CustomerId,
CustomerEmail = customer.EmailAddress,
o.EntrepreneurId,
EntrepreneurEmail = entrepreneur.EmailAddress
};
}
public class ReduceResult
{
public string Id { get; set; }
public string CustomerId { get; set; }
public string CustomerEmail { get; set; }
public string EntrepreneurId { get; set; }
public string EntrepreneurEmail { get; set; }
}
}
If I look at the result of this Index in Raven Studio I get null values for all fields except the Id. And finally here is my query.
Query
var items =
this.documentSession.Query<OrdersApprovedBroadcastingData.ReduceResult, OrdersApprovedBroadcastingData>()
.Select(x => new OrdersToBroadcastListItem
{
Id = x.Id,
CustomerEmailAddress = x.CustomerEmail,
EntrepreneurEmailAddress = x.EntrepreneurEmail
}).ToList();
Change your index to:
public class OrdersApprovedBroadcastingData : AbstractIndexCreationTask<OrderDocument>
{
public OrdersApprovedBroadcastingData()
{
Map = docs => from d in docs
where d.CurrentOrderStatus == OrderStatus.Approved
select new
{
};
TransformResults = (db, orders) =>
from o in orders
let customer = db.Load<CustomerDocument>(o.CustomerInfo.CustomerDocumentId)
let entrepreneur = db.Load<EntrepreneurDocument>(o.EntrepreneurInfo.EntrepreneurDocumentId)
select new
{
o.Id,
CustomerEmailAddress = customer.EmailAddress,
EntrepreneurEmailAddress = entrepreneur.EmailAddress
};
}
}
Your result class can simply be the final form of the projection, you don't need the intermediate step:
public class Result
{
public string Id { get; set; }
public string CustomerEmailAddress { get; set; }
public string EntrepreneurEmailAddress { get; set; }
}
You don't have to nest this class in the index if you don't want to. It doesn't matter either way. You can query either with:
var items = session.Query<Result, OrdersApprovedBroadcastingData>();
Or with
var items = session.Query<OrderDocument, OrdersApprovedBroadcastingData>().As<Result>();
Though, with the first way, the convention tends to be to nest the result class, so really it would be
var items = session.Query<OrderDocument.Result, OrdersApprovedBroadcastingData>();
Note in the index map, I am not including any properties at all. None are required for what you asked. However, if you want to add a Where or OrderBy clause to your query, any fields you might want to filter or sort on should be put in there.
One last thing - the convention you're using of OrderDocument, CustomerDocument, EntrepreneurDocument, is a bit strange. The usual convention is just Order, Customer, Entrepreneur. Think of your documents as the persisted form of the entities themselves. The convention you are using will work, it's just not the one usually used.