Develop Reference

How to process XML file into C# List<T> on conditional basis?

Here is my sample XML schema that is generated by SQL Server. I want it to assign to List<class> in C# using LINQ. I tried it but I am getting six rows but after parsing this XML. I should get only one rows in list.
There some conditions to parse this XML.
TYPE == 1 then SMS,
TYPE == 3 then DATA
TYPE == 2 then MINUTES
XML:
<summary>
<period>Jul-2016</period>
<providerid>7</providerid>
<type>1</type>
<volume>2981655</volume>
</summary>
<summary>
<period>Jul-2016</period>
<providerid>7</providerid>
<type>2</type>
<volume>6449570</volume>
</summary>
<summary>
<period>Jul-2016</period>
<providerid>7</providerid>
<type>3</type>
<volume>7702484</volume>
</summary>
Here is my C# class structure. I want to parse this XML schema into a List<UsageSummary>. After parsing this there will be only two rows into List<UsageSummary>. I have used XDocument.Parse for XML parsing. After that, I am using Linq .Descendants method but I am getting six rows but output should contain only one rows.
public class UsageSummary
{
public int carrierID { get; set; }
public Int64 minutes { get; set; }
public Int64 sms { get; set; }
public Int64 data { get; set; }
public string period { get; set; }
}
I have used this code:
List<UsageSummary> obj = new List<UsageSummary>();
obj = (from res in xmlDoc.Descendants("summary")
select new UsageSummary
{
carrierID = (Convert.ToInt16(res.Element("providerid").Value)),
period = res.Element("period").Value.ToString(),
sms = (Convert.ToInt64(res.Element("SMS").Value)),
data = (Convert.ToInt64(res.Element("DATA").Value)),
minutes = (Convert.ToInt64(res.Element("MINUTES").Value))
}).ToList();
Kindly help me with this problem.

You need to group your elements. You can split the problem into 3 parts so you're not trying to this all at once.
First, map your XML onto a structure that mirrors the XML structure:
var items =
from summary in doc.Descendants("summary")
select new
{
Period = (string) summary.Element("period"),
CarrierId = (int) summary.Element("providerid"),
Type = (int) summary.Element("type"),
Volume = (long) summary.Element("volume")
};
Then group these by CarrierId and Period. For Volume, you want a lookup of all the volumes by Type:
var volumesByPeriodAndCarrier =
from item in items
group item by new {item.CarrierId, item.Period}
into grouping
select new
{
grouping.Key.CarrierId,
grouping.Key.Period,
VolumeByType = grouping.ToLookup(x => x.Type, x => x.Volume)
};
Then you can easily create your summaries. CarrierId and Period map directly to your class, and the 3 volumes are where you use your mapping from types to fields and sum the totals:
var summaries = volumesByPeriodAndCarrier
.Select(x => new UsageSummary
{
carrierID = x.CarrierId,
minutes = x.VolumeByType[2].Sum(),
sms = x.VolumeByType[1].Sum(),
data = x.VolumeByType[3].Sum(),
period = x.Period
}).ToList();
See this fiddle for a working demo.

This solution is implemented using a linq "pivoting" technic.
Pivoting consists in collecting different types of data, each of them corresponding to a given data record (or xml nodes in our example), in the fields of one record (or object..).
In our case, each pivot object contains the "set" of different types of data (on which you can operate several aggregation operators) of a specific grouping of the orginal data set.
In particular, it allows to sum volume values by type, providerid and period, and it gets zero(0) for missing type xml element values.
List<UsageSummary> obj = new List<UsageSummary>();
obj = (from xmlsummary in doc.Descendants("summary")
group xmlsummary by new { id = xmlsummary.Element("providerid").Value, period = xmlsummary.Element("period").Value } into summaryGrouped
select new UsageSummary{
carrierID = Convert.ToInt16(summaryGrouped.Key.id),
period = summaryGrouped.Key.period,
sms = (summaryGrouped.Where(sg => sg.Element("type").Value.Equals("1")).Select(v => Convert.ToInt64(v.Element("volume").Value)).Sum()),
minutes = (summaryGrouped.Where(sg => sg.Element("type").Value.Equals("2")).Select(v => Convert.ToInt64(v.Element("volume").Value)).Sum()),
data = (summaryGrouped.Where(sg => sg.Element("type").Value.Equals("3")).Select(v => Convert.ToInt64(v.Element("volume").Value)).Sum())
}
).ToList();

Lambda SQL Query / Manipulate String At Query Or After Result

I'm using C#, EF5, and Lambda style queries against SQL.
I have the usual scenario of binding data to gridviews. Some of the results for my columns may be too long (character count) and so I only want to display the first 'n' characters. Let's say 10 characters for this example. When I truncate a result, I'd like to indicate this by appending "...". So, let's say the following last names are returned:
Mercer, Smith, Garcia-Jones
I'd like them to be returned like this:
Mercer, Smith, Garcia-Jon...
I was doing something like this:
using (var context = new iaiEntityConnection())
{
var query = context.applications.Where(c => c.id == applicationPrimaryKey);
var results = query.ToList();
foreach (var row in results)
{
if (row.employerName.Length > 10)
{
row.employerName = row.employerName.Substring(0, Math.Min(10, row.employerName.ToString().Length)) + "...";
}
if (row.jobTitle.Length > 10)
{
row.jobTitle = row.jobTitle.Substring(0, Math.Min(10, row.jobTitle.ToString().Length)) + "...";
}
}
gdvWorkHistory.DataSource = results;
gdvWorkHistory.DataBind();
However, if I change my query to select specific columns like this:
var query2 = context.applications.Select(c => new
{
c.id,
c.applicationCode,
c.applicationCategoryLong,
c.applicationType,
c.renew_certification.PGI_nameLast,
c.renew_certification.PGI_nameFirst,
c.renew_certification.PAI_homeCity,
c.renew_certification.PAI_homeState,
c.reviewStatusUser,
c.dateTimeSubmittedByUser
})
The result appears to become read-only if specific columns are selected, and I really should be selecting just the columns I need. I'm losing my ability to edit the result set.
So, I'm rethinking the entire approach. There must be away to select the first 'n' characters on select, right? Is there anyway to append the "..." if the length is > 10 on select? That seems trickier. Also, I guess I could parse through the gridview after bind and make this adjustment. Or, perhaps there is a way to maintain my ability to edit the result set when selecting specific columns?
I welcome your thoughts. Thanks!

To quote MSDN
Anonymous types provide a convenient way to encapsulate a set of read-only properties into a single object without having to explicitly define a type first.
So you would have to define a class and select into that if you want read write capability.
e.g.
public class MyClass {
public int id { get; set; }
public string applicationCode {get; set; }
// rest of property defintions.
}
var query2 = context.applications.Select(c => new MyClass {
id = c.id,
applicationCode = c.applicationCode,
// Rest of assignments
};
As to just providing 10 character limit with ... appended. I'm going to assume you mean on the applicationcategoryLog field but you can use the same logic on other fields.
var query2 = context.applications.Select(c => new
{
c.id,
c.applicationCode,
applicationCategoryLong = (c.applicationCategoryLong ?? string.Empty).Length <= 10 ?
c.applicationCategoryLong :
c.applicationCategoryLong.Substring(0,10) + "...",
c.applicationType,
c.renew_certification.PGI_nameLast,
c.renew_certification.PGI_nameFirst,
c.renew_certification.PAI_homeCity,
c.renew_certification.PAI_homeState,
c.reviewStatusUser,
c.dateTimeSubmittedByUser
})

Cannot implicitly convert type '.List<AnonymousType#1>' to '.List<WebApplication2.Customer>'

In the following code that returns a list:
public List<Customer> GeAllCust()
{
var results = db.Customers
.Select(x => new { x.CustName, x.CustEmail, x.CustAddress, x.CustContactNo })
.ToList()
return results;
}
I get an error reporting that C# can't convert the list:
Error: Cannot implicitly convert type System.Collections.Generic.List<AnonymousType#1> to System.Collections.Generic.List<WebApplication2.Customer>
Why is that?
Here's a screenshot showing some additional information that Visual Studio provides in a tooltip for the error:
Is it right way to return some columns instead of whole table....?
public object GeAllCust()
{
var results = db.Customers.Select(x => new { x.CustName, x.CustEmail, x.CustAddress, x.CustContactNo }).ToList();
return results;
}

When you look the code:
x => new { ... }
This creates a new anonymous type. If you don't need to pull back only a particular set of columns, you can just do the following:
return db.Customers.ToList();
This assumes that Customers is an IEnumerable<Customer>, which should match up with what you are trying to return.
Edit
You have noted that you only want to return a certain subset of columns. If you want any sort of compiler help when coding this, you need to make a custom class to hold the values:
public class CustomerMinInfo
{
public string Name { get; set; }
public string Email { get; set; }
public string Address { get; set; }
public int? ContactNumber { get; set; }
}
Then change your function to the following:
public List<CustomerMinInfo> GetAllCust()
{
var results = db.Customers.Select(x => new CustomerMinInfo()
{
Name = x.CustName,
Email = x.Email,
Address = x.Address,
ContactNumber = x.CustContactNo
})
.ToList();
return results;
}
This will work, however, you will lose all relationship to the database context. This means if you update the returned values, it will not stick it back into the database.
Also, just to repeat my comment, returning more columns (with the exception of byte arrays) does not necessarily mean longer execution time. Returning a lot of rows means more execution time. Your function is returning every single customer in the database, which when your system grows, will start to hang your program, even with the reduced amount of columns.

You are selecting to an anonymous type, which is not a Customer.
If you want to do (sort of) this, you can write it like this:
return db.Customers.Select(x => new Customer { Name = x.CustName, Email = x.CustEmail, Address = x.CustAddress, ContactNo = x.ContactNo }).ToList();
This assumes the properties on your Customer object are what I called them.
** EDIT ** Per your comment,
If you want to return a subset of the table, you can do one of two things:
Return the translated form of Customer as I specified above, or:
Create a new class for your business layer that only has only those four fields, and change your method to return a List<ShrunkenCustomer> (assuming ShunkenCustomer is the name that you choose for your new class.)

GetAllCust() is supposed to return a List of Customer, Select New will create a list of Anonymous Types, you need to return a list of Customer from your query.
try:
var results = db.Customers.Select( new Customer{CustName = x.CustName}).ToList(); //include other fields

I guess Customer is a class you have defined yourself?
The my suggestion would be to do something like the following:
var results = db.Customers.Select(x => new Customer(x.Custname, x.CustEmail, x.CustAddress, x.CustContactNo)).ToList();
The reason is that you are trying to return a list of Customer but the results from your link is an anonymous class containing those four values.
This would of course require that you have a constructor that takes those four values.

Basically whatever u got in var type, loop on that and store it in list<> object then loop and achieve ur target.Here I m posting code for Master details.
List obj = new List();
var orderlist = (from a in db.Order_Master
join b in db.UserAccounts on a.User_Id equals b.Id into abc
from b in abc.DefaultIfEmpty()
select new
{
Order_Id = a.Order_Id,
User_Name = b.FirstName,
Order_Date = a.Order_Date,
Tot_Qty = a.Tot_Qty,
Tot_Price = a.Tot_Price,
Order_Status = a.Order_Status,
Payment_Mode = a.Payment_Mode,
Address_Id = a.Address_Id
});
List<MasterOrder> ob = new List<MasterOrder>();
foreach (var item in orderlist)
{
MasterOrder clr = new MasterOrder();
clr.Order_Id = item.Order_Id;
clr.User_Name = item.User_Name;
clr.Order_Date = item.Order_Date;
clr.Tot_Qty = item.Tot_Qty;
clr.Tot_Price = item.Tot_Price;
clr.Order_Status = item.Order_Status;
clr.Payment_Mode = item.Payment_Mode;
clr.Address_Id = item.Address_Id;
ob.Add(clr);
}
using(ecom_storeEntities en=new ecom_storeEntities())
{
var Masterlist = en.Order_Master.OrderByDescending(a => a.Order_Id).ToList();
foreach (var i in ob)
{
var Child = en.Order_Child.Where(a => a.Order_Id==i.Order_Id).ToList();
obj.Add(new OrderMasterChild
{
Master = i,
Childs = Child
});
}
}

Linq To SQL Select Dynamic Columns

Is it possible to dynamically limit the number of columns returned from a LINQ to SQL query?
I have a database SQL View with over 50 columns. My app has a domain object with over 50 properties, one for each column. In my winforms project I bind a list of domain objects to a grid. By default only a few of the columns are visible however the user can turn on/off any of the columns.
Users are complaining the grid takes too long to load. I captured the LINQ generated SQL query then executed it within SQL Server Management Studio and verified its slow. If I alter the SQL statement, removing all the invisible columns, it runs almost instantly. There is a direct correlation between performance and the number of columns in the query.
I'm wondering if its possible to dynamically alter the number of columns returned from the LINQ generated SQL query? For example, here is what my code currently looks like:
public List<Entity> GetEntities()
{
using (var context = new CensusEntities())
{
return (from e in context.Entities
select e).ToList();
}
}
The context.Entities object was generated from a SQL View that contains over 50 columns so when the above executes it generates SQL like "SELECT Col1, Col2, Col3, ... Col50 FROM Entity INNER JOIN...". I would like to change the method signature to look like this:
public List<Entity> GetEntities(string[] visibleColumns)
{
using (var context = new CensusEntities())
{
return (from e in context.Entities
select e).ToList();
}
}
I'm not sure how to alter the body of this method to change the generated SQL statement to only return the column values I care about, all others can be NULL.

Something like this should work:
List<string> columns = new List<string>();
columns.Add("EmployeeID");
columns.Add("HireDate");
columns.Add("City");
Add columns to your list ^.
var result = Class.ReturnList(columns);
Pass the List to a method ^.
public static List<Entity> ReturnList(List<string> VisibleColumns)
{
StringBuilder SqlStatement = new StringBuilder();
SqlStatement.Append("Select ");
for (int i = 0; i < VisibleColumns.Count; i++)
{
if (i == VisibleColumns.Count - 1)
{
SqlStatement.Append(VisibleColumns[i]);
}
else
{
SqlStatement.Append(VisibleColumns[i]);
SqlStatement.Append(",");
}
}
SqlStatement.Append(" FROM Entity");
using (var ctx = new DataClasses1DataContext())
{
var result = ctx.ExecuteQuery<Entity>(SqlStatement.ToString());
return result.ToList();
}
}
This basically just makes a SELECT statement with all the fields you passed in with the VisibleColumns list.
In this case, the SQL statement that will be generated by the strings in the VisibleColumns list is:
Select EmployeeID, HireDate, City From Employee
(note: i used the Northwind database to try this out, hence the EmployeeID etc column names. You should replace them with your own, obviously.)

It is not trivial to do this dynamically, but if you have a limited set of combinations of columns you want to retreive you can do an explicit select like this:
public List<Entity> GetEntities()
{
using (var context = new CensusEntities())
{
return (from e in context.Entities
select new
{
col1 = e.col1,
col4 = e.col4,
col5 = e.col5,
}
).ToList()
.Select(x=>new Entity{col1 = x.col1, col4 = x.col4, col5 = x.col5}).ToList();
}
}
The extra select step is necessary because LINQ2SQL won't create partial entities for you.
Create a method for each common combination of columns (especially the initial) the users wants to retrieve.
However to make this dynamic you can build a query with you entity stored as a property in an anonymous class and collect your result properties in another anonymous class in second property in the same anonymous class. Finally you select your entities from the collected objects into objects of the correct type.
public List<Entity> GetEntities()
{
using (var context = new CensusEntities())
{
var combinedResult = (from e in context.Entities
select new {
Entity = e,
CollectedValues = new
{
// Insert default values of the correct type as placeholders
col1 = 0, // or "" for string or false for bool
col2 = 0, // or "" for string or false for bool
// ...
col49 = 0, // or "" for string or false for bool
col50 = 0, // or "" for string or false for bool
}
);
// Then copy each requested property
// col1
if (useCol1)
{
var combinedResult = (from e in combinedResult
select new {
Entity = e,
CollectedValues = new
{
col1 = e.Enitity.col1, // <-- here we update with the real value
col2 = e.CollectedValues.col2, // <-- here we just use any previous value
// ...
col49 = e.CollectedValues.col49, // <-- here we just use any previous value
col50 = e.CollectedValues.col50, // <-- here we just use any previous value }
);
}
// col2
if (useCol2)
{
// same as last time
col1 = e.CollectedValues.col1, // <-- here we just use any previous value
col2 = e.Enitity.col2, // <-- here we update with the real value
// ...
}
// repeat for all columns, update the column you want to fetch
// Just get the collected objects, discard the temporary
// Entity property. When the query is executed here only
// The properties we actually have used from the Entity object
// will be fetched from the database and mapped.
return combinedResult.Select(x => x.CollectedValues).ToList()
.Select(x=>new Entity{col1 = x.col1, col2 = x.col2, ... col50 = x.col50}).ToList();
}
}
There will be lots of code, and a pain to maintain, but it should work.
If you are going this route I suggest that you build a code generator that builds this code with reflection from your LINQ context.

Try something like this
using (var context = new CensusEntities())
{
var q = from e in context.Entities
select e.myfield1,e.myfield2;
return q.Tolist();
}
The resulting query should be lighter and also all the data conversion that goes underneath.
But if you really need to build dynamic input, I think some dynamic sql should be involved. So
build the dynamic SQL and get a data table
use a datatable to a dynamic object conversion as shown here
How can I convert a DataTable into a Dynamic object?
BTW a lot of hard work, I think you should considered using the first block of code.

Efficient Way To Query Nested Data

I have need to select a number of 'master' rows from a table, also returning for each result a number of detail rows from another table. What is a good way of achieving this without multiple queries (one for the master rows and one per result to get the detail rows).
For example, with a database structure like below:
MasterTable:
- MasterId BIGINT
- Name NVARCHAR(100)
DetailTable:
- DetailId BIGINT
- MasterId BIGINT
- Amount MONEY
How would I most efficiently populate the data object below?
IList<MasterDetail> data;
public class Master
{
private readonly List<Detail> _details = new List<Detail>();
public long MasterId
{
get; set;
}
public string Name
{
get; set;
}
public IList<Detail> Details
{
get
{
return _details;
}
}
}
public class Detail
{
public long DetailId
{
get; set;
}
public decimal Amount
{
get; set;
}
}

Normally, I'd go for the two grids approach - however, you might also want to look at FOR XML - it is fairly easy (in SQL Server 2005 and above) to shape the parent/child data as xml, and load it from there.
SELECT parent.*,
(SELECT * FROM child
WHERE child.parentid = parent.id FOR XML PATH('child'), TYPE)
FROM parent
FOR XML PATH('parent')
Also - LINQ-to-SQL supports this type of model, but you need to tell it which data you want ahead of time. Via DataLoadOptions.LoadWith:
// sample from MSDN
Northwnd db = new Northwnd(#"c:\northwnd.mdf");
DataLoadOptions dlo = new DataLoadOptions();
dlo.LoadWith<Customer>(c => c.Orders);
db.LoadOptions = dlo;
var londonCustomers =
from cust in db.Customers
where cust.City == "London"
select cust;
foreach (var custObj in londonCustomers)
{
Console.WriteLine(custObj.CustomerID);
}
If you don't use LoadWith, you will get n+1 queries - one master, and one child list per master row.

It can be done with a single query like this:
select MasterTable.MasterId,
MasterTable.Name,
DetailTable.DetailId,
DetailTable.Amount
from MasterTable
inner join
DetailTable
on MasterTable.MasterId = DetailTable.MasterId
order by MasterTable.MasterId
Then in psuedo code
foreach(row in result)
{
if (row.MasterId != currentMaster.MasterId)
{
list.Add(currentMaster);
currentMaster = new Master { MasterId = row.MasterId, Name = row.Name };
}
currentMaster.Details.Add(new Detail { DetailId = row.DetailId, Amount = row.Amount});
}
list.Add(currentMaster);
There's a few edges to knock off that but it should give you the general idea.

select < columns > from master
select < columns > from master M join Child C on M.Id = C.MasterID

You can do it with two queries and one pass on each result set:
Query for all masters ordered by MasterId then query for all Details also ordered by MasterId. Then, with two nested loops, iterate the master data and create a new Master object foreach row in the main loop, and iterate the details while they have the same MasterId as the current Master object and populate its _details collection in the nested loop.

Depending on the size of your dataset you can pull all of the data into your application in memory with two queries (one for all masters and one for all nested data) and then use that to programatically create your sublists for each of your objects giving something like:
List<Master> allMasters = GetAllMasters();
List<Detail> allDetail = getAllDetail();
foreach (Master m in allMasters)
m.Details.Add(allDetail.FindAll(delegate (Detail d) { return d.MasterId==m.MasterId });
You're essentially trading memory footprint for speed with this approach. You can easily adapt this so that GetAllMasters and GetAllDetail only return the master and detail items you're interested in. Also note for this to be effective you need to add the MasterId to the detail class

This is an alternative you might consider. It does cost $150 per developer, but time is money too...
We use an object persistence layer called Entity Spaces that generates the code for you to do exactly what you want, and you can regenerate whenever your schema changes. Populating the objects with data is transparent. Using the objects you described above would look like this (excuse my VB, but it works in C# too):
Dim master as New BusinessObjects.Master
master.LoadByPrimaryKey(43)
Console.PrintLine(master.Name)
For Each detail as BusinessObjects.Detail in master.DetailCollectionByMasterId
Console.PrintLine(detail.Amount)
detail.Amount *= 1.15
End For
With master.DetailCollectionByMasterId.AddNew
.Amount = 13
End With
master.Save()