I am using Database First EF to generate model from the existing database. When I first generated the models, it ignores only one of the table, the entity was not added to EDMX, no model file is created for the table and no context is created for the entity.
When I tried to explicitly add the table to EDMX (when generating the model, selected the specific table first and then updated the model with all the other tables from the database), it complained with the following error.
Two entities with possibly different keys are mapped to the same row. Ensure these two mapping fragments map both ends of the AssociationSet to the corresponding columns.
This specific table has two columns which are primary keys of some other tables and both the columns are specified as Primary keys for the table.
Am I doing something wrong or should I handle this table differently since it has two columns defined as Primary Keys? Any suggestions greatly appreciated!
You are not doing anything wrong. Your table is junction table for many-to-many relation. You don't need that table in the model because EF (in contrast to database) can handle many-to-many relation directly without any intermediate. The table is actually mapped on behind of the many-to-many relation - you will see that in mapping details window.
Btw. you are not using code first. Code first = no EDMX.
Related
I have two fact tables: FactSales & FactInvoices. Both have an foreign key relationship with DimDate.Datekey. In VS, the SSAS DSV displays these relationships (the lines are drawn between the tables).
In the DSV I decided to to create a named query that limits the dimdate to 2021. After doing this, I still see the relationships between the two fact tables and dimdate (which is now a named query).
At the DB-level, I created a 3rd fact table called FactExpenses. FactExpenses also has an FK relationship with DimDate.Datekey. The problem is that my dsv (in SSAS) does not recognize this relationship (ie. It doesn't draw the line between both tables).
Two questions: why doesn't VS display the relationship between my 3rd fact table with the named query but it does with the other two fact tables? I understand that the relationship isn't with the named query, but the relationship should disappear in all the fact tables.
When I want to limit the amount of data displayed in dimdate, should I use a named query?
The relationships in the DSV are separate to the foreign keys on the base tables, but they get added automatically based on the database schema when you add tables to the DSV. My guess would be when you added the initial dim and fact tables to the DSV in Visual Studio it automatically added the relationships based on the foreign keys that exist on the base tables, but this may not occur automatically for named queries. You can manually add the relationship yourself for the third table to get the same result.
I think a named query is a reasonable approach for the filtering you want to do. An alternative would be to create a view in the source database if you need to do more intense or complex filtering.
I have an SQL database with 200 tables.
I changed the primary key on one "basic" table.
I changed ALL foreign keys referring to that table, in more than 20 tables.
I Updated the LinQ model (right-click, update model from DB) to reflect the changes in the model itself.
As a result, 3 tables disappeared from the model (they are not visible in the graphical view of the model), and I get errors: Error 3013: mapping problem from line xxxx: no mapping fot table "tablename"
Trying to update again, if I look on the details of the tables to add, I can't see the 3 tables "lost": they are in the DB, I'm sure, and SEEM to be in the model, because they aren't in the list of tables that I can add, but they aren't in the model.
What could I do in this situation?
It seems solved ... one friend suggested to open manually the edmx file with an editor (I used Notepad++), and remove any reference at the 3 tables (they were there), save, and try the update again, and the tables appeared again.
As indicated in the Entity Framework image below, I have 3 tables, tblModel, tblModelFetish and tblFetish. A record in tblModel can have multiple records in the tblModelFetish table linked by the modelid column. The tblModelFetish table links to the tblFetish table via the fetishID column to get the fetish description stored in the fetish column. This Entity Model was generated with VS 2010 from an existing database including foreign keys.
a Entity Data Model http://spreadthenudes.com/efmodel.jpg
Using the odata syntax, I'm able to access all models (http://localhost:51157/WcfDataService.svc/tblModels) or a specific model (http://localhost:51157/WcfDataService.svc/tblModels(11)) successfully. I'm having trouble accessing the related tables data via odata, I've tried many permutations including expand etc.
What I want is a result set of all the columns in the tblModel and the related tblFetish records including the fetish column from the tblFetish table. In other words, Mary (a modelname in tableModel) has 3 fetishes (3 records in tblModelFetish) named beach, travel and coffee (stored in tblFetish, fetish column).
What is the odata syntax to acquire this?
thanks for reading! Bob
Try either:
http://localhost:51157/WcfDataService.svc/tblModels(11)?$expand=tblModelFetishes/tblFetish
or
http://localhost:51157/WcfDataService.svc/tblModels?$filter=id eq 11&$expand=tblModelFetishes/tblFetish
Just to make it clear you will not get one huge result set with all columns but the entity graph consisted of your entities.
I am just getting into Entity Framework for the first time beyond simple examples.
I am using the model-first approach and am querying the data source with LINQ-to-Entities.
I have created an entity model that I am exposing as an OData service against a database where I do not control the schema. In my model, I have two entities that are based off of two views in this database. I've created an association between the two entities. Both views have a column with the same name.
I am getting the error:
Ambiguous column name 'columnname'. Could not use view or function 'viewname' because of binding errors.
If I was writing the SQL statement myself, I'd qualify one of the column names with an alias to prevent this issue. EF apparently isn't doing that. How do I fix this, short of changing the view? (which I cannot do) I think this does have something to do with these entities being mapped to views, instead of being mapped to actual tables.
Assuming you can change the model have you tried going into the model and just changing one of the column names? I can still see how it might be problematic if the two views are pulling back the same column from the same table. I can tell that when working directly with a model mapped to tables, having identically named columns is not a problem. Even having multiple associations to the same table is handled correctly, the Navigation Properties are automatically given unique names. Depending on which version of EF you used you should be able to dig into the cs file either under the model or under the t4 template file and see what's going on. Then you can always create a partial class to bend it to your will.
Entity framework returning only one value but the list size is correct
I have a table that does not have primary id and I need to get or select all the values in it.
What I see is when I do the selection with linq the number of objects is correct but it is the first row over and over.
I am simply doing something like this
List<MyValueType> valuesInDB = myDb.MyValueTypes.ToList();
Problem is I may get thousands of rows (which is correct) but the rows all have the same exact data.
I am using VS 2010 and used the wizard to create my EF object.
The problem is that entity framework is not able to work with entity without a key. So if your table doesn't specify a key, entity framework will infer its own. The key created by EF is composed of all non-nullable non-binary columns.
So if you for example have single non-nullable column in your entity which have only very small set of values (like enum) you will be able to load only single entity "per value". The reason is an inner implementation of the context and the state manager which uses Identity map pattern. When data record is retrieved from database, EF will first check an entity key and tries to find an object with the same key in its internal storage. If an object is found it will use that object instead of data record retrieved (despite of different data). If an object with the key is not found a new object is materialized and added to internal storage.
That is the purpose of Identity map - object with given key should be created only once by each context. Identity map is core pattern in ORM.
I wrote about Identity map also in this question.
I would suggest searching for the word "Warning" in your EDM's designer.cs file. It might tell you if Entity Framework is having any issues with your table.
I really can't comment much in the absence of the table design. I tried replicating your problem but wasn't able to do so. Here is what I did:
Created a table with no primary key but it had a unique key on an ID column. Entity Framework was able to infer a primary key and when I fetched the data, I not only got the correct number of rows but also the corrects data in those rows.
Created a table with no primary key and no unique key. Also there was no column called ID. Entity Framework excluded this table in the EDM that was generated. Consequently I wasn't able to query this table at all.This was displayed as a warning in the EDM designer file.
It would be better if you can share the create script for your table.