Building LINQ Expression Getting Ignored - linq

I'm trying to build a linq query coming from a table grid from the client side, so im expecting page offset, page start, order and the traditional paging parameters. I have the following code:
[Route("api/settings/logs")]
public Rest.DatatablesResponse GetLogs(int draw, int start, int length) {
var query_string = Request.GetQueryNameValuePairs().ToDictionary(x => x.Key, x => x.Value);
var search = query_string["search.value"];
int order_column = int.Parse(query_string["order[0].column"]);
var order_direction = query_string["order[0].dir"];
var count = db.Logs.Count(q => q.Mode == 2);
var logs = (from l in db.Logs
where l.Mode == 2
select new {
id = l.ID,
mode = l.Mode,
phase_id = l.Phase.ID,
created = l.Created,
user = l.User.Name,
blender_name = l.Blender.Name,
oil_name = l.Oil,
oil_quantity = l.OilQuantity,
production_cycle_name = l.ProductionCycle.Name
});
if (order_direction == "asc") {
if (order_column == 0) logs.OrderBy(q => q.created);
else if (order_column == 2) logs.OrderBy(q => q.production_cycle_name);
} else {
if (order_column == 0) logs.OrderByDescending(q => q.created);
else if (order_column == 2) logs.OrderByDescending(q => q.production_cycle_name);
};
if (!string.IsNullOrEmpty(search)) {
logs.Where(q => q.blender_name.Contains(search) ||
q.oil_name.Contains(search) ||
SqlFunctions.StringConvert((decimal)q.id).Contains(search));
}
logs.Skip(start).Take(length);
DateTime dtDateTime = new DateTime(1970,1,1,0,0,0,0,System.DateTimeKind.Utc);
var steps = from l in logs.ToList()
select new {
id = l.id,
message = StringHelpers.FormatWith(_tpl_message[l.phase_id.ToString() + l.mode.ToString() ], l) ,
created = dtDateTime.AddSeconds(l.created).ToString("h:mmtt - MMMM d, yyyy"),
production_cycle_name = l.production_cycle_name
};
return new Rest.DatatablesResponse {
draw = draw,
recordsTotal = count,
recordsFiltered = count,
data = steps.ToArray()
};
}
My problem is the skip and take and orderby expressions are getting ignored for some reason, and this is the SQL code generated just before converting my linq expressions to a list. From my understanding, the query should not be executed or evaluated until my logs.ToList() call, so the ordering and take/skip should be taken into account, but it's not:
{SELECT
[Extent1].[ID] AS [ID],
[Extent1].[Mode] AS [Mode],
[Extent1].[Phase_ID] AS [Phase_ID],
[Extent1].[Created] AS [Created],
[Extent2].[Name] AS [Name],
[Extent3].[Name] AS [Name1],
[Extent1].[Oil] AS [Oil],
[Extent1].[OilQuantity] AS [OilQuantity],
[Extent4].[Name] AS [Name2]
FROM [dbo].[Steps] AS [Extent1]
LEFT OUTER JOIN [dbo].[Users] AS [Extent2] ON [Extent1].[User_Id] = [Extent2].[Id]
LEFT OUTER JOIN [dbo].[Blenders] AS [Extent3] ON [Extent1].[Blender_ID] = [Extent3].[ID]
LEFT OUTER JOIN [dbo].[ProductionCycles] AS [Extent4] ON [Extent1].[ProductionCycle_ID] = [Extent4].[ID]
WHERE 2 = [Extent1].[Mode]}
Irrelevant P.S. I'm using the not so clever ifs for building the order expression instead of using DynamicLINQ since i have only two sortable columns.

logs.Skip(start).Take(length);
Creates a IQueryable<T> where T is the same anonymous type of which logs is an IQueryable<T> but with start items skipped. Then from that it creates a similarly typed IQueryable<T> where lenght items are the most that will be taken.
Then it throws that away and lets it be garbage collected. (Or ideally the compiler or jitter steps will realise it's thrown away and cut out the whole thing).
Then logs.ToList() goes back to the logs you still have and creates a list from it.
You should replace the Skip and Take line with:
logs = logs.Skip(start).Take(length);
So that you are actually making use of this skipping and taking.
I'm using the not so clever ifs for building the order expression instead of using DynamicLINQ since i have only two sortable columns.
There's nothing particularly not-clever about that, except that you make the same mistake; apply the OrderBy and then throwing away the result instead of using it. Likewise with the Where. You need logs = logs.OrderBy(...) etc.
I'd also question from l in logs.ToList() select new {…} here.
It might be the best approach, if obtaining that list in one step has some advantage. However otherwise:
from l in logs select new {…}
Do the select work on the database, retrieving just what you need.
from l in logs.AsEnumerable() select new {…}
Do the select work in the application, appropriate if part of it cannot be converted to database work, but do it as it comes rather than loading it all into memory first.
from l in await logs.ToListAsync() select new {…}
Has the downside of ToList() but in asynchronous uses, then (assuming your provider has a ToListAsync() method) allows for awaiting.
ToList() is rarely the best option here.

Related

EF Linq query with conditional include

So I have the following Linq query:
var member = (from mem in
context.Members.Include(m =>
m.MemberProjects.Select(mp => mp.Project))
where mem.MemberId == memberId
select mem).FirstOrDefault();
This returns a Member entity, with a set of MemberProjects that have a Project child. I would like to limit the MemberProjects to only those for which the Project child has a property
ProjectIdParent == null.
One of my failed attempts might make the intent clearer:
var member = (from mem in context.Members
.Include(m => m.MemberProjects
.Where(mp =>
mp.Project.ProjectIdParent == null)
.Select(proj => proj.Project))
where mem.MemberId == memberId
select mem).FirstOrDefault();
This of course complains of an invalid Include expression because of the Where clause.
Any thoughts on how to do this would be great :)
DISCLAIMER: I havent tested this. This is just an idea. If you let me know the results, I will update this accordingly. (Skip to the update part for the tested solutions)
var member = (from mps in context.MemberProjects
.Include(m => m.Members)
.Include(m => m.Projects)
where mps.Project.ProjectIdParent == null
select mps)
.FirstOrDefault(mprojs => mprojs.Member.MemberId == memberId);
I'd also analyze the queries using something like EFProfiler to make sure the generated queries dont leave the realm of sanity.
You can also take a look at this post by Jimmy Bogard on Many to Many relationships with ORMs.
Update
I came up with multiple tested solutions for this with EF 6.1.3. My Edmx looked like below:
The setup data is like below:
I was able to run code below to get the MemberFive correctly
var member = context.Members.FirstOrDefault
(m => m.MemberId == memberId
&& m.Projects.Any(p => p.ProjectParentId == null));
The generated SQL looked like this:
SELECT TOP (1) [Extent1].[MemberId] AS [MemberId],
[Extent1].[MemberName] AS [MemberName]
FROM [dbo].[Members] AS [Extent1]
WHERE ([Extent1].[MemberId] = 1)
AND (EXISTS (SELECT 1 AS [C1]
FROM (SELECT [MemberProjects].[MemberId] AS [MemberId],
[MemberProjects].[ProjectId] AS [ProjectId]
FROM [dbo].[MemberProjects] AS [MemberProjects])
AS [Extent2]
INNER JOIN [dbo].[Projects] AS [Extent3]
ON [Extent3].[ProjectId] = [Extent2].[ProjectId]
WHERE ([Extent1].[MemberId] = [Extent2].[MemberId])
AND ([Extent3].[ProjectParentId] IS NULL)))
If you dont like the generated query you can use this:
var memberQuery = #"Select M.* from Members M
inner join MemberProjects MP on M.MemberId = Mp.ProjectId
inner join Projects P on MP.ProjectId = P.ProjectId
where M.MemberId = #MemberId and P.ProjectParentId is NULL";
var memberParams = new[]
{
new SqlParameter("#MemberId", 1)
};
var member3 = context.Members.SqlQuery(memberQuery, memberParams)
.FirstOrDefault();
The later consistently returned under 20ms vs the other one hovered around 60ms (if that matters to you).
I hope this helps.

Linq to SQL not querying memory

I am using Linq for SQL and have always thought that I was querying the results of a Query in memory. I have just looked at the database and it is showing many thousands of queries rather than 1 which is what I expected.
My approach has been to run a query and then use linq to search within the resultset.
IQueryable<mapping> fieldList = from mm in db.mappings
join mi in db.metaItems on mm.secondaryMetaItemId equals mi.metaItemId
join mo in db.metaObjects on mi.metaObjectId equals mo.metaObjectId
where mm.linkageId == 277
select mm;
for (int i=0;i<100;i++)
{
mapping thisVar = fieldList.FirstOrDefault(m => m.primaryItem == info.Name);
}
How can I stop Linq requerying everytime I access my resultset...
Thanks for your help!
When you write a LINQ query, the query doesn't get actually get executed until you perform an action that actually enumerates over it (deferred execution). Calling FirstOrDefault() is an example of one such method that enumerates over the result (the first result has to be found). You'll want to call a method or otherwise enumerate over the results once. That way when you want to refer to those results throughout your program, you do it on a stored copy.
The easiest way you can do that is to convert it to a list. That will put the results in memory as a list. You could then use that.
IQueryable<mapping> fieldList =
from mm in db.mappings
join mi in db.metaItems on mm.secondaryMetaItemId equals mi.metaItemId
join mo in db.metaObjects on mi.metaObjectId equals mo.metaObjectId
where mm.linkageId == 277
select mm;
// save it!
var result = fieldList.ToList(); // query is processed only once here
// do stuff with result
for (int i=0;i<100;i++)
{
// using the stored result
thisVar = result.FirstOrDefault(m => m.primaryItem == info.Name);
}
try this :
var fieldList = (from mm in db.mappings
join mi in db.metaItems on mm.secondaryMetaItemId equals mi.metaItemId
join mo in db.metaObjects on mi.metaObjectId equals mo.metaObjectId
where mm.linkageId == 277
select mm).AsEnumerable();
foreach (int i=0;i<100;i++)
{
mapping thisVar = fieldList.FirstOrDefault(m => m.primaryItem == info.Name);
}

Multiple Counts within a single query

I want a list of counts for some of my data (count the number of open.closed tasks etc), I want to get all counts inside 1 query, so I am not sure what I do with my linq statement below...
_user is an object that returns info about the current loggedon user
_repo is am object that returns an IQueryable of whichever table I want to select
var counters = (from task in _repo.All<InstructionTask>()
where task.AssignedToCompanyID == _user.CompanyID || task.CompanyID == _user.CompanyID
join instructions in _repo.GetAllMyInstructions(_user) on task.InstructionID equals
instructions.InstructionID
group new {task, instructions}
by new
{
task
}
into g
select new
{
TotalEveryone = g.Count(),
TotalMine = g.Count(),
TotalOpen = g.Count(x => x.task.IsOpen),
TotalClosed = g.Count(c => !c.task.IsOpen)
}).SingleOrDefault();
Do I convert my object to single or default? The exception I am getting is, this sequence contains more than one element
Note: I want overall stats, not for each task, but for all tasks - not sure how to get that?
You need to dump everything into a single group, and use a regular Single. I am not sure if LINQ-to-SQL would be able to translate it correctly, but it's definitely worth a try.
var counters = (from task in _repo.All<InstructionTask>()
where task.AssignedToCompanyID == _user.CompanyID || task.CompanyID == _user.CompanyID
join instructions in _repo.GetAllMyInstructions(_user) on task.InstructionID == instructions.InstructionID
group task by 1 /* <<=== All tasks go into one group */ into g select new {
TotalEveryone = task.Count(),
TotalMine = task.Count(), // <<=== You probably need a condition here
TotalOpen = task.Count(x => x.task.IsOpen),
TotalClosed = task.Count(c => !c.task.IsOpen)
}).Single();
From MSDN
Returns the only element of a sequence, or a default value if the
sequence is empty; this method throws an exception if there is more
than one element in the sequence.
You need to use FirstOrDefault. SingleOrDefault is designed for collections that contains exactly 1 element (or none).

Getting Last rows from the result of Linq to Sql statement

I couldn't get last articles of every writers in this statement.
List<Editor> lstEditors = dataContext.GetTable<Editor>().Where(t => t.M_Active).Select(t => t).ToList();
var lstArticles = from article in DAO.context.GetTable<Article>().ToList()
join editor in lstEditors on article.RefEditorId equals editor.EditorId
select
new
{
article.M_ArticleId,
article.M_Subject,
article.M_Text,
editor.M_EditorId,
editor.M_Member.M_EditorPicture,
M_NameSurname = editor.M_Member.M_Fname + " " + editor.M_Member.M_Lname
};
Be careful, your query is fetching all the contents of both the Editor and the Yazi tables and then performs Linq-to-Objects on it.
I'm not sure what you ask exactly either, do you want to obtain the list of all writers (editors) along with the last article of each one of these writers?
Do you want to get the writers that did not write any articles yet also?
Edit:
explanation of methods causing an immediate query
Any time you call one of the methods listed below on an IQueryable object (tables or other queries), it performs the actual query to SQL server:
ToList(), ToArray(), ToLookup(), ToDictionay()
Count(), Sum(), Avg(), Aggregate(), Min(), Max()
First(), FirstOrDefault(), Last(), LastOrDefault()
getting last article written by each writer
//create a subquery that returns an editor and its last article date
var editorLastArticleDates =
from article in DAO.context.GetTable<Article>()
group article by article.RefEditor into g
let lastArticleDate= g.Max(x => x.Date)
select new
{
Editor = g.Key,
LastArticleDate = lastArticleDate,
};
//Note: We did not do a ToList() here so the query is not executed
// The editorLastArticleDates object is a IQueryable<>
var query =
from article in DAO.context.GetTable<Article>()
join editorLastArticleDate in editorLastArticleDates
on new { article.Editor, article.Date } // 1
equals new { editorLastArticleDate.Editor, // 2
Date = editorLastArticleDate.LastArticleDate } // 3
select new
{
article.M_ArticleId,
article.M_Subject,
article.M_Text,
article.RefEditor.M_EditorId,
article.RefEditor.M_Member.M_EditorPicture,
M_NameSurname = article.RefEditor.M_Member.M_Fname + " "
+ article.RefEditor.M_Member.M_Lname,
};
//Note: We did not do a ToList() yet so the query is not executed
// The query object is a IQueryable<>
Console.WriteLine(query.ToString()); //Displays SQL query on the console
var results = query.ToList(); // SQL query is executed on this line.
In the code above, I left some remarks on things I had problems with:
When using join, the section between new and equals access only variables declared before the join keyword while the section after the equals keyword has access to the variable defined between join and in.
When writing your join condition, make sure you use equals and not ==.
When using new { XXX, YYY } syntax in your join condition, you declare anonymous types. If the property names are not identical on both sides, it will not compile. In order to have identical property names in this sample, I added the Date = before my value.
By the way, you should use LinqPad to test your queries, it is really a nice tool.

Is there a better way to code this LINQ fragment?

I have this fragment of code:
SmsDataClassesDataContext dc = new SmsDataClassesDataContext();
// Get the customer
Customer currentCustomer = dc.Customers.Single( c => c.Hash1 == forThisHash );
// Get from Name (LINQ to XML)
var q = from c in thisSmsPack.Descendants("from")
select c;
string from = q.First().Value;
foreach ( XElement element in thisSmsPack.Descendants("to") )
{
// Create the queue
SmsQueue sq = new SmsQueue();
sq.CustomerId = currentCustomer.CustomerId;
sq.MsgFrom = from;
sq.MsgTo = element.Attribute("name").Value;
sq.MsgPhone = element.Attribute("phone").Value;
sq.MsgBody = element.Attribute("msg").Value;
sq.Priority = currentCustomer.SendsSmsAtPriority;
sq.DontSendUntil = GetNextSendDate();
// sq.TimeCreated = System.DateTime.Now;
currentCustomer.SmsQueues.Add(sq);
}
dc.SubmitChanges();
I am creating new instances of "SmsQueues", populating the values and when the foreach loop is finished I submit the changes. Given the new lambda/linq/anonymous types that .NET 3.5 has, is there a more "modern" way to accomplish the above?
As a side question, maybe related, can I return an existing type composed of different columns in the select part of the linq expression?
Suppose you have three tables:
T1 == T1.Id, T1.Name
T2 == T2.Id, T2.Phone
T3 == T3.Name, T3.Phone, T3.SomethingElse
Can I perform a LINQ query that returns:
T1.Name, T2.Phone, SomethingElseNew
And let .NET know that that is of Type T3 (and it's a new instance of it)?
That way when I SubmitChanges, new T3 instances are inserted in the DB?
I don't know if I make myself clear :S
I don't have a system available to test this, but I think this (or something very close) should work.
CustomerId = currentCustomer.CustomerId;
var sqrange = from element in thisSmsPack.Descendants("to") )
select new SmsQueue
{
// Create the queue
MsgFrom = from,
MsgTo = element.Attribute("name").Value,
MsgPhone = element.Attribute("phone").Value,
MsgBody = element.Attribute("msg").Value,
Priority = currentCustomer.SendsSmsAtPriority,
DontSendUntil = GetNextSendDate()
// TimeCreated = System.DateTime.Now
};
currentCustomer.SmsQueues.AddRange(sqrange);
EDIT: Fixed the numerous syntax errors (as delineated in the comments)
You could do something like this (syntax may be off slightly, no intellisense here):
var q = T1.Join(T2, t => t.Id, t2 => t2.Id)
select new T3{Name=t.Name,Phone=t2.Phone,SomethingElseNew="Chickens"};

Resources