In Example II of Indexing Related Documents, an index is built over Authors by Name and Book title. The relevant entities look like so:
public class Book {
public string Id { get; set; }
public string Name { get; set; }
}
public class Author {
public string Id { get; set; }
public string Name { get; set; }
public IList<string> BookIds { get; set; }
}
I.e. only the Author holds information about the relation. This information is used in constructing said index.
But how would I construct an index over Books by Authors (assuming a book could have multiple authors)?
Edit:
The book/author analogy only goes so far. I'll make an example that's closer to my actual use case:
Suppose we have some tasks that are tied to locations:
public class Location {
public string Id { get; set; }
public double Latitude { get; set; }
public double Longitude { get; set; }
}
public class Task {
public string Id { get; set; }
public string Name { get; set; }
public string LocationId { get; set; }
public Status TaskStatus { get; set; }
}
I have an endpoint serving Locations as GeoJson to a map view in a client. I want to color the Locations depending on status of Tasks associated with them. The map would typically show 500-2000 locations.
The query on locations is implemented as a streaming query.
Using the query-method indicated in Ayende's initial answer, I might do something like:
foreach (var location in locationsInView)
{
var completedTaskIds = await RavenSession.Query<Task>()
.Where(t => t.LocationId == location.Id && t.TaskStatus == Status.Completed)
.ToListAsync();
//
// Construct geoJson from location and completedTaskIds
//
}
This results in 500-2000 queries being executed against RavenDB, which doesn't seem right.
This is why I initially thought I needed an index to construct my result.
I have since read that RavenDB caches everything by default, so that might be a non-issue. On the other hand, having implemented this approach, I get an error ("...maximum number of requests (30) allowed for this session...").
What is a good way of fixing this?
You cannot index them in this manner.
But you also don't need to.
If you want to find all the books by an author, you load the author and you have the full list.
You can do this using a multi map/reduce index.
All sources of truth about the objects of interest are mapped to a common object type (Result). This mapping is then reduced grouping by Id and keeping just the relevant pieces, creating a "merge" of truths about each object (Book in this case). So using the Book/Author example, where several Authors might have contributed to the same book, you could do something like the following.
Note that the map and reduce steps must output the same type of object, which is why author.Id is wrapped in a list during the mapping from author.
Author.Names are excluded for brevity, but could be included in the exact same way as Author.Id.
public class BooksWithAuthors : AbstractMultiMapIndexCreationTask<BooksWithAuthors.Result>
{
public class Result
{
string Id;
string Title;
IEnumerable<string> AuthorIds;
}
public BooksWithAuthors()
{
AddMap<Book>(book => from book in books
select new
{
Id = book.Id,
Title = book.Title,
AuthorIds = null;
});
AddMap<Author>(author => from author in authors
from bookId in author.bookIds
select new
{
Id = bookId,
Title = null,
AuthorIds = new List<string>(){ author.Id };
});
Reduce = results => from result in results
group result by result.Id
into g
select new
{
Id = g.Key,
Title = g.Select(r => r.Title).Where(t => t != null).First(),
AuthorIds = g.Where(r => r.AuthorIds != null).SelectMany(r => r.AuthorIds)
};
}
}
Related
TLDR:
Is it possible to:
store a document (Article) with a relation (Tags) in one Store/StoreAsync call to RavenDB, but into separate collections?
then fetch the parent document (Article) including the related documents (Tags) in one query (without including/loading the tags separately)?
Explanation
AFAIK the only way to store data to RavenDB which have a relation into separate collections is to store them individually. When you read the data, you need to Include the related documents and call Load to get them.
I wonder if there is a way to simplify this by storing and querying Articles and related Tags in one go.
I have a idea how I wish it would work (but it does not), as well a working but cumbersome example.
The examples below are split into these steps
POCOs
Storing data
Index definition
Querying the index
I put the broken-but-I-wish-it-would-work and the working examples next to each other. I think it is easier to understand it that way.
POCOs
broken-but-I-wish-it-would-work
namespace Articles
{
public class ArticlePersistance
{
public string Id { get; set; }
public string Title { get; set; }
public List<TagPersistance> Tags { get; set; } // Specify TagPersistance here
}
[DearRavenDBStoreToSeparateCollectionPlease] // Does not exist
public class TagPersistance
{
public string Id { get; set; }
public string Name { get; set; }
}
}
working
namespace Articles
{
public class ArticlePersistance
{
public string Id { get; set; }
public string Title { get; set; }
public List<string> Tags { get; set; } // Specify string here
}
public class TagPersistance
{
public string Id { get; set; }
public string Name { get; set; }
}
}
Storing data
broken-but-I-wish-it-would-work
Storing ArticlePersistance and TagPersistance into their own collections with one call to StoreAsync. AFAIK this stores the Article and Tags into the same collection.
var tag = new TagPersistance() { Name = "Tag1" };
var article = new ArticlePersistance()
{
Title = "aaa",
Tags = new List<TagPersistance> { tag } // Embed the full Tag here
};
await session.StoreAsync(article); // Only one call to StoreAsync
await session.SaveChangesAsync();
working
Storing the Article and Tags separately:
var tag = new TagPersistance() { Name = "Tag1" };
await session.StoreAsync(tag); // Store Tag separately
var article = new ArticlePersistance()
{
Title = "aaa",
Tags = new List<string> { tag.Id } // Embed only the tag id
};
await session.StoreAsync(article);
await session.SaveChangesAsync();
Index definition
broken-but-I-wish-it-would-work
Index on ArticlePersistance which stores the full Tag objects
public class Articles_Test : AbstractIndexCreationTask<ArticlePersistance>
{
public Articles_Test()
{
Map = articles =>
from article in articles
let tags = article.Tags.Select(t => LoadDocument<TagPersistance>(t)) // Load the related Tags
select new
{
Title = article.Title,
Tags = tags // Store the full Tag objects here
};
}
}
working
Index which holds only the Tag names, not the full Tag objects:
public class Articles_Test : AbstractIndexCreationTask<ArticlePersistance>
{
public Articles_Test()
{
Map = articles =>
from article in articles
let tags = article.Tags.Select(t => LoadDocument<TagPersistance>(t)) // Load the related Tags
select new
{
Title = article.Title,
Tags = tags.Select(t => t.Name) // Store only the Tag name
};
}
}
Querying the index
broken-but-I-wish-it-would-work
Finally querying the index and getting the article with the tags back.
I hoped for fetching the Article and the Tags in one go here
// This does not work
var article = await session
.Query<ArticlePersistance, Articles_Test>()
.Where(a => a.Title == "aaa")
.ToListAsync();
working
This is working, but cumbersome.
You need to care about the relation between Article and Tags which could already be specified in the Index definition.
var article = await session
.Query<ArticlePersistance, Articles_Test>()
.Where(a => a.Title == "aaa")
.Include(t => t.Tags) // Include the tags
.ToListAsync();
// Query the tags separatelly
var tags = await session.LoadAsync<TagPersistance>(article.SelectMany(a => a.Tags));
I have referenced numerous questions on this site related to calculated fields and ViewModels, but I can't seem to extrapolate from examples given. I hope that laying out a specific scenario would allow someone to pin point what I can't see. I am new to WebApp design in general. Please take that into consideration. Also, if I've left off any relevant information, please let me know and I will update the question.
Here is the scenario:
I have a complex query that is spanning multiple tables to return data used in calculations. Specifically, I store units for a recipe converted to a base unit and then convert the quantity to the units specified by the user.
I am using AutoMapper to map from entities to ViewModels and vice versa, but I am not sure how to handle the calculated values. Especially with the nested ViewModel Collection thrown into the mix.
Option 1
Do I return an autonomous set of data? Like the following... and then somehow use AutoMapper to do the mapping? Perhaps I would need to do the mapping manually, which I haven't found a solid example which includes nested ViewModels. At this point, I'm not even sure if the following code handles the nested collection correctly for the autonomous data.
var userId = User.Identity.GetUserId();
var recipes = from u in db.Users.Where(u => u.Id == userId)
from c in db.Categories
from r in db.Recipes
join ur in db.UserRecipes.Where(u => u.UserId == userId) on r.Id equals ur.RecipeId
join mus in db.MeasUnitSystems on ur.RecipeYieldUnitSysId equals mus.Id
join muc in db.MeasUnitConvs on mus.Id equals muc.UnitSysId
join mu in db.MeasUnits on mus.UnitId equals mu.Id
join msy in db.MeasUnitSymbols on mu.Id equals msy.UnitId
select new
{
Id = c.Id,
ParentId = c.ParentId,
Name = c.Name,
Descr = c.Descr,
Category1 = c.Category1,
Category2 = c.Category2,
Recipes = new
{
Id = r.Id,
Title = r.Title,
Descr = r.Descr,
Yield = String.Format("{0} {1}", ((r.Yield * muc.UnitBaseConvDiv / muc.UnitBaseConvMult) - muc.UnitBaseConvOffset), msy.Symbol)
}
};
Option 2
Another option that crossed my mind was to return the entities and use AutoMapper as I normally would. Then iterate through the collections and perform the calculations there. I feel like I could make this work, but it seems inefficient to me because it would result in many queries back to the database.
Option 3
???? I can't think of any other method to do this. But, please, if you have suggestions, I am more than willing to hear them.
Relevant Data
Here is the query returning the data I want in SQL Server (more or less).
declare #uid as nvarchar(128) = 'da5435ae-5198-4690-b502-ea3723a9b217'
SELECT c.[Name] as [Category]
,r.Title
,r.Descr
,(r.Yield*rmuc.UnitBaseConvDiv/rmuc.UnitBaseConvMult)-rmuc.UnitBaseConvOffset as [Yield]
,rmsy.Symbol
FROM Category as c
inner join RecipeCat as rc on c.Id = rc.CategoryId
inner join Recipe as r on rc.RecipeId = r.Id
inner join UserRecipe as ur on r.Id = ur.RecipeId and ur.UserId = #uid
inner join MeasUnitSystem as rmus on ur.RecipeYieldUnitSysId = rmus.Id
inner join MeasUnitConv as rmuc on rmus.Id = rmuc.UnitSysId
inner join MeasUnit as rmu on rmus.UnitId = rmu.Id
inner join MeasUnitSymbol as rmsy on rmu.Id = rmsy.UnitId
inner join UserUnitSymbol as ruus on rmsy.UnitId = ruus.UnitId and rmsy.SymIndex = ruus.UnitSymIndex and ruus.UserId = #uid
ViewModels
public class CategoryRecipeIndexViewModel
{
public int Id { get; set; }
public int ParentId { get; set; }
[Display(Name = "Category")]
public string Name { get; set; }
[Display(Name = "Description")]
public string Descr { get; set; }
public ICollection<CategoryRecipeIndexViewModel> Category1 { get; set; }
public CategoryRecipeIndexViewModel Category2 { get; set; }
public ICollection<RecipeIndexViewModel> Recipes { get; set; }
}
public class RecipeIndexViewModel
{
public int Id { get; set; }
[Display(Name = "Recipe")]
public string Title { get; set; }
[Display(Name = "Description")]
public string Descr { get; set; }
[Display(Name = "YieldUnit")]
public string Yield { get; set; }
}
UPDATE 2/10/2018
I found an answer here that does a very good job of explaining exactly what I'm looking at. Particularly under the A Better solution ? section. Mapping queries directly to my ViewModels looks like it would allow me to get my calculated values as well. Problem is, the example given is once again too simplistic.
He gives the following DTO's
public class UserDto
{
public int Id {get;set;}
public string Name {get;set;}
public UserTypeDto UserType { set; get; }
}
public class UserTypeDto
{
public int Id { set; get; }
public string Name { set; get; }
}
And does the following for mapping:
var users = dbContext.Users.Select(s => new UserDto
{
Id = s.Id,
Name = s.Name,
UserType = new UserTypeDto
{
Id = s.UserType.Id,
Name = s.UserType.Name
}
});
Now what if the UserDTO looked like this:
public class UserDto
{
public int Id {get;set;}
public string Name {get;set;}
public ICollection<UserTypeDto> UserTypes { set; get; }
}
How would the mapping be done if the UserTypes were a collection?
Update 2/13/2018
I feel I am making progress, but am currently headed in the wrong direction. I found this and came up with the following (which currently errors because of the method call in the linq query):
*Note: I removed Category2 from the ViewModel as I found it was not needed and only complicated this further.
query inside index controller method
IEnumerable<CategoryRecipeIndexViewModel> recipesVM = db.Categories
.Where(x => x.ParentId == null)
.Select(x => new CategoryRecipeIndexViewModel()
{
Id = x.Id,
ParentId = x.ParentId,
Name = x.Name,
Descr = x.Descr,
Category1 = MapCategoryRecipeIndexViewModelChildren(x.Category1),
Recipes = x.Recipes.Select(y => new RecipeIndexViewModel()
{
Id = y.Id,
Title = y.Title,
Descr = y.Descr
})
});
Recursive Method
private static IEnumerable<CategoryRecipeIndexViewModel> MapCategoryRecipeIndexViewModelChildren(ICollection<Category> categories)
{
return categories
.Select(c => new CategoryRecipeIndexViewModel
{
Id = c.Id,
ParentId = c.ParentId,
Name = c.Name,
Descr = c.Descr,
Category1 = MapCategoryRecipeIndexViewModelChildren(c.Category1),
Recipes = c.Recipes.Select(r => new RecipeIndexViewModel()
{
Id = r.Id,
Title = r.Title,
Descr = r.Descr
})
});
}
At this point, I don't even have the calculations I require, but that doesn't matter until I get this working (small steps). I quickly discovered you can't really call a method inside a Linq Query. Then a thought occurs to me, if I need to force the Linq Query to execute and then perform all the mapping on the in memory data, then I would essentially be doing the same thing as Option 2 (above), but I could perform the calculations within the ViewModel. This is the solution I will pursue and will keep everyone posted.
You have to iterate over UserType Collection and map the value to UserType dto's collection.
Use this code.
var users = dbContext.Users.Select(s => new UserDto
Id = s.Id,
Name = s.FullName,
UserType = s.UserType.Select(t => new UserTypeDto
{
Id = t.Id,
Name = t.Name
}).ToList()
Hope this will help.
I got it working! ...I think. ...Maybe. If anything, I'm querying the data, mapping it to my ViewModels and I have the calculations too. I do have additional questions, but they are a lot more specific. I will layout the solution I followed and where I think it requires work below.
I basically implemented my Option 2 from above, but instead of iterating through the collections, I just performed the calculations within the ViewModels.
Controller Method
public ActionResult Index()
{
var userId = User.Identity.GetUserId();
var recipes = db.Categories.Where(u => u.Users.Any(x => x.Id == userId))
.Include(c => c.Category1)
.Include(r => r.Recipes
.Select(u => u.UserRecipes
.Select(s => s.MeasUnitSystem.MeasUnitConv)))
.Include(r => r.Recipes
.Select(u => u.UserRecipes
.Select(s => s.MeasUnitSystem.MeasUnit.MeasUnitSymbols)));
IEnumerable<CategoryRecipeIndexViewModel> recipesVM = Mapper.Map<IEnumerable<Category>, IEnumerable<CategoryRecipeIndexViewModel>>(recipes.ToList());
return View(recipesVM);
}
View Models
public class CategoryRecipeIndexViewModel
{
public int Id { get; set; }
public int ParentId { get; set; }
[Display(Name = "Category")]
public string Name { get; set; }
[Display(Name = "Description")]
public string Descr { get; set; }
public ICollection<CategoryRecipeIndexViewModel> Category1 { get; set; }
public ICollection<RecipeIndexViewModel> Recipes { get; set; }
}
public class RecipeIndexViewModel
{
public int Id { get; set; }
[Display(Name = "Recipe")]
public string Title { get; set; }
[Display(Name = "Description")]
public string Descr { get; set; }
public double Yield { get; set; }
public ICollection<UserRecipeIndexViewModel> UserRecipes { get; set; }
[Display(Name = "Yield")]
public string UserYieldUnit
{
get
{
return System.String.Format("{0} {1}", ((Yield *
UserRecipes.FirstOrDefault().MeasUnitSystem.MeasUnitConv.UnitBaseConvDiv /
UserRecipes.FirstOrDefault().MeasUnitSystem.MeasUnitConv.UnitBaseConvMult) -
UserRecipes.FirstOrDefault().MeasUnitSystem.MeasUnitConv.UnitBaseConvOffset).ToString("n1"),
UserRecipes.FirstOrDefault().MeasUnitSystem.MeasUnit.MeasUnitSymbols.FirstOrDefault().Symbol);
}
}
}
public class UserRecipeIndexViewModel
{
public MeasUnitSystemIndexViewModel MeasUnitSystem { get; set; }
}
public class MeasUnitSystemIndexViewModel
{
public MeasUnitIndexViewModel MeasUnit { get; set; }
public MeasUnitConvIndexViewModel MeasUnitConv { get; set; }
}
public class MeasUnitIndexViewModel
{
public ICollection<MeasUnitSymbolIndexViewModel> MeasUnitSymbols { get; set; }
}
public class MeasUnitConvIndexViewModel
{
public double UnitBaseConvMult { get; set; }
public double UnitBaseConvDiv { get; set; }
public double UnitBaseConvOffset { get; set; }
}
public class MeasUnitSymbolIndexViewModel
{
public string Symbol { get; set; }
}
This appears to be working, but I know it needs some work.
For instance, the relation shown between the Recipe and UserRecipe shows one to many. In reality, if the UserRecipe were filtered by the current user, the relationship would be one to one. Also, the same goes for the MeasUnit and the MeasUnitSymbol entities. Currently, I'm relying on the FirstOrDefault of those collections to actually perform the calculations.
Also, I have seen numerous posts that state that calculations should not be done in the View Models. Except for some who say it's okay if it is only a requirement of the View.
Last I will say that paying attention to variable names within the ViewModels would have saved me some headaches. And I thought I knew how to utilize Linq Queries, but had issues with the data returned. It was easier to rely on the eager loading provided by Entity Framework to bring back the hierarchical data structure needed, versus the flat table structures I'm used to working with.
I'm still new to a lot of this and wrapping my head around some of the quirks of MVC and Entity Framework leaves me brain dead after a few hours, but I will continue to optimize and adopt better programming methods as I go.
I have the code below which works:
List<XXTJobTableModel> xjobs = filexxts.GroupBy(x=> x.job)
.Select(fx => new JobTableModel
{
job_no = fx.First().job,
emps = fx.GroupBy(x=>x.emp_id).Select(x => new EmployeeTableModel
{
eid = x.First().emp_id,
heds = x.GroupBy(h => h.HED).Select(h => new HEDModel
{
hed = h.First().HED,
hours = h.Sum(c => c.HOURS),
amt = h.Sum(c => c.AMOUNTRATE)
}).ToList()
}).ToList()
}).ToList();
public class JobTableModel
{
public string job_no { get; set; }
public List<EmployeeTableModel> emps { get; set; }
}
public class EmployeeTableModel
{
public string emp_id { get; set; }
public List<HEDModel> heds { get; set; }
}
public class THEDModel
{
public string hed { get; set; }
public decimal hours { get; set; }
public decimal amt { get; set; }
}
I have another List<XXTJobTableModel> yjobs already loaded from another data source. What I would like to do, if the job_no is not found in yjobs, then add the job (with employee and he data) to yjobs. If the job_no is found in yjobs, and the emp_id is not found in the Employee table for that job, then add the employee (and hed data) to that jobs. If the employee is found in that job, then just add the HED data to the employees list.
I do not have a preference if I somehow merge xjobs and yjobs or if I load the xjobs directly into the yjobs list.
thanks
Are you looking for a very concise set of LINQ statements to do this merge operation or just any solution at all? What I would do is loop through one set and use LINQ to compare each item to the other set, and add items resulting from the comparison to a third set. Problems like this you have to decompose into manageable pieces, and then recompose it into a larger solution once the smaller pieces are working. Make an attempt and if you get stuck someone on here will surely be able to fill in the blanks.
Need a help with RavenDB.
In my web page I want to have such list:
item1 category1
item2 category2
...
and another one:
category1, number of items
category2, number of items
...
My data structures:
public class Item
{
public string Id { get; set; }
public string Name { get; set; }
public string CategoryId { get; set; }
}
public class Category
{
public string Id { get; set; }
public string Name { get; set; }
public bool IsActive { get; set; }
}
Index for the first list:
public class Item_WithCategory : AbstractIndexCreationTask<Item>
{
public class Result
{
public string Name { get; set; }
public string CategoryName { get; set; }
}
public Item_WithCategory()
{
Map = items => from item in items
select new
{
Name = item.Name,
CategoryName = LoadDocument<Category>(item.CategoryId).Name
};
}
}
Is this data structure suitable for my case, or will it be better to have Category instead of CategoryId in item structure?
Should I use my index or is there a better solution to take category name?
If my index is good, how to write a correct query? My current try:
Item_WithCategory.Result[] all;
using (var session = DocumentStoreHolder.Store.OpenSession())
{
all = session.Query<Item_WithCategory.Result, Item_WithCategory>().ToArray();
}
but it throws exception stating that return type is item, not result. How to fix it?
You have a couple of options here. You could store both the CategoryId and the CategoryName on the Item entity. This will of course lead to duplicated data (if you still need to store the Category entity), but "storage is cheap" is a popular term these days.The downside of this is that you need to update each Item document of a given category if the category name changes to keep things consistent. A benefit is that you need to do less work to get your desired result.
If you store Category Name on the item as well you don't need a special index to handle the first list, just query on the Items and return what you need. For the second list you need to create a Map/Reduce index on the Item entity that groups on the category.
However, if you need to use the data structure you've given, there are a couple of ways of solving this. First, it's really not recommended to use a LoadDocument inside of an index definition, especially not in a select statement. This might affect indexing performance in a negative way.
Instead, just index the properties you need to query on (or use an auto index) and then use a Result Transformer to fetch information from related documents:
public class ItemCategoryTransformer : AbstractTransformerCreationTask<Item>
{
public ItemCategoryTransformer()
{
TransformResults = results => from item in results
let category = LoadDocument<Category>(item.CategoryId)
select new ItemCategoryViewModel
{
Name = item.Name,
CategoryName = category.Name
};
}
}
public class ItemCategoryViewModel
{
public string Name { get; set; }
public string CategoryName { get; set; }
}
You can use this Transformer with a Query on the Item entity:
using (var session = documentStore.OpenSession())
{
var items = session.Query<Item>()
.TransformWith<ItemCategoryTransformer, ItemCategoryViewModel>()
.ToList();
}
As for the second list, still using your data structure, you have to use a couple of things. First, a Map/Reduce index over the Items, grouped by CategoryId:
public class Category_Items_Count : AbstractIndexCreationTask<Item, Category_Items_Count.Result>
{
public class Result
{
public string CategoryId { get; set; }
public int Count { get; set; }
}
public Category_Items_Count()
{
Map = items => from item in items
select new Result
{
CategoryId = item.CategoryId,
Count = 1
};
Reduce = results => from result in results
group result by result.CategoryId
into c
select new Result
{
CategoryId = c.Key,
Count = c.Sum(x => x.Count)
};
}
}
But as you only have the CategoryId on the Item entity, you have to use a similar transformer as in the first list:
public class CategoryItemsCountTransformer : AbstractTransformerCreationTask<Category_Items_Count.Result>
{
public CategoryItemsCountTransformer()
{
TransformResults = results => from result in results
let category = LoadDocument<Category>(result.CategoryId)
select new CategoryItemsCountViewModel
{
CategoryName = category.Name,
NumberOfItems = result.Count
};
}
}
public class CategoryItemsCountViewModel
{
public string CategoryName { get; set; }
public int NumberOfItems { get; set; }
}
And lastly, query for it like this:
using (var session = documentStore.OpenSession())
{
var items = session.Query<Category_Items_Count.Result, Category_Items_Count>()
.TransformWith<CategoryItemsCountTransformer, CategoryItemsCountViewModel>()
.ToList();
}
As you can see, there are quite a difference in work needed depending on what data structure you're using. If you stored the Category Name on the Item entity directly you wouldn't need any Result Transformers to achieve the results you're after (you would only need a Map/Reduce index).
However, Result Transformers are executed server side and are only executed on request, instead of using LoadDocument inside of an index which is executed every time indexing occurs. Also, and maybe why LoadDocuments inside of index definitions isn't recommended, every change to a document that's referenced with a LoadDocument in an index will cause that index to have to be rewritten. This might lead to a lot of work for the index engine.
Lastly, to answer your last question about why you get an exception when querying: As the actual return type of your index is the document that that's being indexed (in this case Item). To use something else you need to project your result to something else. This can be done by using ".As()" in the query:
Item_WithCategory.Result[] all;
using (var session = DocumentStoreHolder.Store.OpenSession())
{
all = session.Query<Item_WithCategory.Result, Item_WithCategory>()
.As<Item_WithCategory.Result>()
.ToArray();
}
Long post, but hope it helps!
I have the following entity collections in RavenDB:
public class EntityA
{
public string Id { get; set; }
public string Name { get; set; }
public string[] Tags { get; set; }
}
public class EntityB
{
public string Id { get; set; }
public string Name { get; set; }
public string[] Tags { get; set; }
}
The only thing shared is the Tags collection: a tag of EntityA may exist in EntityB, so that they may intersect.
How can I retrieve every EntityA that has intersecting tags with EntityB where the Name property of EntityB is equal to a given value?
Well, this is a difficult one. To do it right, you would need two levels of reducing - one by the tag which would expand out your results, and another by the id to collapse it back. Raven doesn't have an easy way to do this.
You can fake it out though using a Transform. The only problem is that you will have skipped items in your result set, so make sure you know how to deal with those.
public class TestIndex : AbstractMultiMapIndexCreationTask<TestIndex.Result>
{
public class Result
{
public string[] Ids { get; set; }
public string Name { get; set; }
public string Tag { get; set; }
}
public TestIndex()
{
AddMap<EntityA>(entities => from a in entities
from tag in a.Tags.DefaultIfEmpty("_")
select new
{
Ids = new[] { a.Id },
Name = (string) null,
Tag = tag
});
AddMap<EntityB>(entities => from b in entities
from tag in b.Tags
select new
{
Ids = new string[0],
b.Name,
Tag = tag
});
Reduce = results => from result in results
group result by result.Tag
into g
select new
{
Ids = g.SelectMany(x => x.Ids),
g.First(x => x.Name != null).Name,
Tag = g.Key
};
TransformResults = (database, results) =>
results.SelectMany(x => x.Ids)
.Distinct()
.Select(x => database.Load<EntityA>(x));
}
}
See also the full unit test here.
There is another approach, but I haven't tested it yet. That would be to use the Indexed Properties Bundle to do the first pass, and then map those results for the second pass. I am experimenting with this in general, and if it works, I will update this answer with the results.