Linq query to do group by and min - c#

I have a datastructure as shown below. For each (Person, Game) pairs, I need to find the lastest score in the past 24 hours. Is it possible to do that in LINQ? Something like (Person, Game, LatestScore)
+----------------+-----------------+---------------+-+------------+
| Person | Game | Score |EventTime |
+-----------------------------------------------------------------+
| | | | |
| | | | |
+----------------+-----------------+---------------+--------------+
Any hints would be very helpful.

Assuming you have a class like this:
class GameInfo
{
public DateTime EventTime { get; set; }
public String Game { get; set; }
public String Person { get; set; }
public double Score { get; set; }
}
You could do:
List<GameInfo> data = new List<GameInfo> {
new GameInfo { Game = "G1", Person = "A", Score = 10, EventTime = new DateTime(2014, 10, 10, 10, 10, 10) },
new GameInfo { Game = "G1", Person = "A", Score = 10, EventTime = new DateTime(2014, 10, 10, 10, 10, 10) },
new GameInfo { Game = "G2", Person = "B", Score = 11, EventTime = new DateTime(2014, 10, 10, 20, 10, 10) },
new GameInfo { Game = "G2", Person = "B", Score = 11, EventTime = new DateTime(2014, 10, 10, 20, 10, 10) }
};
var q = from game in data
group game by new { G = game.Game, P = game.Person } into g
select new {
Person = g.Key.P,
Game = g.Key.G,
Score = g.Aggregate((curmin, x) => (curmin == null || (x.EventTime) < curmin.EventTime ? x : curmin)).Score
};
foreach (var item in q)
{
Console.WriteLine("{0}, {1}, {2}", item.Game, item.Person, item.Score);
}
As #Rawling pointed out, getting the scoce can be quite costly if you have lage data sets. So doing that efficiently might save alot of time in getting the desired output.

var dt = DateTime.Now.AddDays(-1);
var results = context.Where(x=>x.EventTime >= dt)
.GroupBy(x=>new {x.Person,x.Game})
.Select(x=>new
{
x.Key.Person,
x.Key.Game,
LatestScore = x.Where(d=>d.EventTime == x.Max(l=>l.EventTime))
.Select(d=>d.Score)
.FirstOrDefault()
});

Related

Can I merge two lists using Linq?

I'm trying to merge two lists and I thought I had a solution but if there are two PackItems with the same length the results are not as expected.
Expectations/requirements.
Both lists contain the same total number of pieces for each length.
EDIT: Added code to clarify the input requirements.
The same length can be used in multiple PacksItems.
The same lengths can be produced out of multiple CoilNums.
The goal is to contain a list the contains a unique entry for each PackItem.ID/CoilNum.
Requirement for the output is that the total number of pieces for each length matched the input lists.
Here is the code I have so far.
public class PackItem
{
public int ID { get; set; }
public int Quantity { get; set; }
public string Length { get; set; }
}
public class ProductionInfo
{
public ProductionInfo AddID(PackItem item)
{
LineID = item.ID;
Quantity = Math.Min(Quantity, item.Quantity);
return this;
}
public int LineID { get; set; }
public string CoilNum { get; set; }
public int Quantity { get; set; }
public string Length { get; set; }
}
private void DoTest()
{
var packItems = new List<PackItem>()
{
new PackItem() {ID = 4, Quantity = 5, Length = "10"},
new PackItem() {ID = 5, Quantity = 2, Length = "4"},
new PackItem() {ID = 6, Quantity = 1, Length = "4"}
};
var productionInfoList = new List<ProductionInfo>()
{
new ProductionInfo() { CoilNum = "A", Quantity = 4, Length = "10"},
new ProductionInfo() { CoilNum = "B", Quantity = 1, Length = "10"},
new ProductionInfo() { CoilNum = "B", Quantity = 2, Length = "4"},
new ProductionInfo() { CoilNum = "A", Quantity = 1, Length = "4"},
};
//assert that both lists meet input requirements
var result1 = "";
var sum1 = packItems.GroupBy(i => i.Length);
foreach (var group in sum1) result1 += $"{group.Sum(i=>i.Quantity)} | {group.Key}\n";
var input2 = "";
var result2 = "";
var sum2 = productionInfoList.GroupBy(i => i.Length);
foreach (var group in sum2) result2 += $"{group.Sum(i => i.Quantity)} | {group.Key}\n";
Console.WriteLine("packItems: \nSum(Quantity) | Length");
Console.WriteLine(result1);
Console.WriteLine("productionInfoList: \nSum(Quantity) | Length");
Console.WriteLine(result2);
if (result1 == result2)
{
Console.WriteLine("Both Lists have the same quantity of each length");
}
else
{
Console.WriteLine("Error: Both Lists do not have the same quantity of each length");
return;
}
var merged = productionInfoList.SelectMany(x => packItems, (x, y) => new { x, y })
.Where(i => i.x.Length == i.y.Length)
.Select(i => i.x.AddID(i.y));
Console.WriteLine("ID | Coil | Qty | Length");
foreach (var item in merged)
{
Console.WriteLine($"{item.LineID} | {item.CoilNum} | {item.Quantity} | {item.Length}");
}
}
//expected output
ID | Coil | Qty | Length
4 | A | 4 | 10
4 | B | 1 | 10
5 | B | 2 | 4
6 | A | 1 | 4
//actual output
ID | Coil | Qty | Length
4 | A | 4 | 10
4 | B | 1 | 10
5 | B | 2 | 4
6 | B | 1 | 4
5 | A | 1 | 4
6 | A | 1 | 4
I'm stuck at this point and they only way I can think of is splitting each of these lists into individual items of one each, and then compiling a list by looping through them one by one.
Is there a way this can be done with Linq?
Here is a method that produces the correct output. Is there an easier way to do this? Can this be done with Linq only?
private void DoTest()
{
var packItems = new List<PackItem>()
{
new PackItem() {ID = 4, Quantity = 5, Length = "10"},
new PackItem() {ID = 5, Quantity = 2, Length = "4"},
new PackItem() {ID = 6, Quantity = 1, Length = "4"}
};
var productionInfoList = new List<ProductionInfo>()
{
new ProductionInfo() { CoilNum = "A", Quantity = 4, Length = "10"},
new ProductionInfo() { CoilNum = "B", Quantity = 1, Length = "10"},
new ProductionInfo() { CoilNum = "B", Quantity = 2, Length = "4"},
new ProductionInfo() { CoilNum = "A", Quantity = 1, Length = "4"},
};
//first create a list with one item for each pieces
var individualProduction = new List<ProductionInfo>();
foreach (var item in productionInfoList)
{
for (int i = 0; i < item.Quantity; i++)
{
individualProduction.Add(new ProductionInfo()
{
Quantity = 1,
Length = item.Length,
CoilNum = item.CoilNum
});
}
}
//next loop through and assign all the pack line ids
foreach (var item in individualProduction)
{
var packItem = packItems.FirstOrDefault(i => i.Quantity > 0 && i.Length == item.Length);
if (packItem != null)
{
packItem.Quantity -= 1;
item.LineID = packItem.ID;
}
else
{
item.Quantity = 0;
}
}
//now group them back into a merged list
var grouped = individualProduction.GroupBy(i => (i.CoilNum, i.LineID, i.Length));
//output the merged list
var merged1 = grouped.Select(g => new ProductionInfo()
{
LineID = g.Key.LineID,
CoilNum = g.Key.CoilNum,
Length = g.Key.Length,
Quantity = g.Count()
});
}
Quite unclear ...
This one is closed of the desired result but doesn't take into consideration any quantity so that the fist PackItem is always choosed. If decreasing the pItem.Quantity this would select the next available pItem.ID where Quantity > 0. But this will require more code :)
var results = productionInfoList.Select(pInfo =>
{
var pItem = packItems.First(z => z.Length == pInfo.Length);
return new { pItem.ID, pInfo.CoilNum, pInfo.Quantity, pInfo.Length };
}).ToList();
When you have a goal of : The goal is to contain a list the contains a unique entry for each PackItem.ID/CoilNum. your bottom answer is correct, since it has a unique id coilNum pair. What you are looking for is a different uniquenes.
var l = packItems.Join(productionInfoList, x => x.Length, y => y.Length, (x, y) => { y.AddID(x); return y; }).GroupBy(x => new { x.CoilNum, x.Length }).Select(x => x.First());
It is unclear on the exact rules of the case, but here I am using Length as a unique key to perform a join operation (Would recommend to have a different unique key for join operations).

Complex linq join in EF6

There are two entities, for example, job and solution.
Each of them has a date field and a level field and a quantity field.
It is necessary to combine them so that they are grouped first by level, then by month, and at the same time, their quantity must be summed up.
I tried different options, but nothing comes out at all. The main problem is grouping by months and summing the numbers in the enclosed sheets.
That is, the output should be one sequence of summed numbers, grouped by level, and then by month.
For example:
var jobs = new List<Job>()
{
new Job { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 111 },
new Job { Level = 1, Date = new DateTime(2019, 1, 20), Quantity = 222 },
new Job { Level = 2, Date = new DateTime(2019, 2, 1), Quantity = 333 },
new Job { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 444 }
};
var solutions = new List<Solution>()
{
new Solution { Level = 1, Date = new DateTime(2019, 2, 1), Quantity = 555 },
new Solution { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 666 },
new Solution { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 777 },
new Solution { Level = 2, Date = new DateTime(2019, 1, 20), Quantity = 888 }
};
Output:
Level 1 -> 1 Jan 2019 -> 1110 (111 + 222 + 777)
Level 1 -> 1 Feb 2019 -> 555
Level 2 -> 1 Jan 2019 -> 888
Level 2 -> 1 Feb 2019 -> 1443 (333 + 444 + 666)
And so on. And yes, all this is in EF6.
Try following which uses Concat. I create a class for the merging. It can also be done anonymously.
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
namespace ConsoleApplication116
{
class Program
{
static void Main(string[] args)
{
var jobs = new List<Job>()
{
new Job { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 111 },
new Job { Level = 1, Date = new DateTime(2019, 1, 20), Quantity = 222 },
new Job { Level = 2, Date = new DateTime(2019, 2, 1), Quantity = 333 },
new Job { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 444 }
};
var solutions = new List<Solution>()
{
new Solution { Level = 1, Date = new DateTime(2019, 2, 1), Quantity = 555 },
new Solution { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 666 },
new Solution { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 777 },
new Solution { Level = 2, Date = new DateTime(2019, 1, 20), Quantity = 888 }
};
List<LevelDateQuantity> concat = jobs.Select(x => new LevelDateQuantity() { Date = x.Date, Level = x.Level, Quantity = x.Quantity})
.Concat( solutions.Select(x => new LevelDateQuantity() { Date = x.Date, Level = x.Level, Quantity = x.Quantity})).ToList();
List<LevelDateQuantity> results = concat.OrderBy(x => x.Level).ThenBy(x => x.Date)
.GroupBy(x => new { level = x.Level, date = new DateTime(x.Date.Year, x.Date.Month,1)})
.Select(x => new LevelDateQuantity() { Level = x.Key.level, Date = x.Key.date, Quantity = x.Sum(y => y.Quantity)})
.ToList();
}
}
public class LevelDateQuantity
{
public int Level { get; set; }
public DateTime Date { get; set; }
public int Quantity { get; set; }
}
public class Job : LevelDateQuantity
{
public int Level { get; set; }
public DateTime Date { get; set; }
public int Quantity { get; set; }
}
public class Solution : LevelDateQuantity
{
public int Level { get; set; }
public DateTime Date { get; set; }
public int Quantity { get; set; }
}
}
Oath, because we can not see your poco class structure we don't know if the two tables are seperate or has a one to many relation to a master table, so by the code you have provided I would do this ;
var jobs = new List<Job>()
{
new Job { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 111 },
new Job { Level = 1, Date = new DateTime(2019, 1, 20), Quantity = 222 },
new Job { Level = 2, Date = new DateTime(2019, 2, 1), Quantity = 333 },
new Job { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 444 }
};
var solutions = new List<Solution>()
{
new Solution { Level = 1, Date = new DateTime(2019, 2, 1), Quantity = 555 },
new Solution { Level = 2, Date = new DateTime(2019, 2, 20), Quantity = 666 },
new Solution { Level = 1, Date = new DateTime(2019, 1, 1), Quantity = 777 },
new Solution { Level = 2, Date = new DateTime(2019, 1, 20), Quantity = 888 }
};
foreach (var sol in solutions)
{
var jb = new Job();
jb.Level = sol.Level;
jb.Date = sol.Date ;
jb.Quantity= sol.Quantity;
jobs.Add(jb);
}
var result = Jobs.GroupBy(x=> new { x.Level, x.Date}).Select(x=> new
{
level = x.Key.Level,
date = x.Key.Date,
sumQ = x.Sum(y => y.Quantity )
});
I haven't tested the code and not wrote in in a compiler so there might be some typeerrors apart from that this should solve your problem.

Join three list using multiple columns c# linq lambda

I have these lists:
var subjects = new List<SubjectModel>
{
new SubjectModel { subjId = 1, subjName = "Math" },
new SubjectModel { subjId = 2, subjName = "Science" },
new SubjectModel { subjId = 3, subjName = "History" },
new SubjectModel { subjId = 4, subjName = "Language" }
};
var quizzes = new List<QuizModel>
{
new QuizModel { quizId = 1, quizDate = DateTime.Parse("2016-11-25"), quizScore = 10, subjectId = 1 },
new QuizModel { quizId = 2, quizDate = DateTime.Parse("2016-11-25"), quizScore = 15, subjectId = 1 },
new QuizModel { quizId = 3, quizDate = DateTime.Parse("2016-11-25"), quizScore = 8, subjectId = 2 },
new QuizModel { quizId = 4, quizDate = DateTime.Parse("2016-11-26"), quizScore = 13, subjectId = 1 },
new QuizModel { quizId = 5, quizDate = DateTime.Parse("2016-11-26"), quizScore = 20, subjectId = 2 }
};
var exams = new List<ExamModel>
{
new ExamModel { examId = 1, examDate = DateTime.Parse("2016-11-25"), examScore = 90, subjectId = 1 },
new ExamModel { examId = 2, examDate = DateTime.Parse("2016-11-25"), examScore = 88, subjectId = 2 },
new ExamModel { examId = 3, examDate = DateTime.Parse("2016-11-25"), examScore = 92, subjectId = 4 },
new ExamModel { examId = , examDate = DateTime.Parse("2016-11-26"), examScore = 84, subjectId = 1 },
};
var exercises = new List<ExerciseModel>
{
new ExerciseModel { exerciseId = 1, exerciseDate = DateTime.Parse("2016-11-25"), exerciseScore = 17, subjectId = 1 },
new ExerciseModel { exerciseId = 2, exerciseDate = DateTime.Parse("2016-11-25"), exerciseScore = 15, subjectId = 2 },
new ExerciseModel { exerciseId = 3, exerciseDate = DateTime.Parse("2016-11-26"), exerciseScore = 15, subjectId = 1 },
new ExerciseModel { exerciseId = 4, exerciseDate = DateTime.Parse("2016-11-26"), exerciseScore = 12, subjectId = 4 },
new ExerciseModel { exerciseId = 5, exerciseDate = DateTime.Parse("2016-11-26"), exerciseScore = 10, subjectId = 1 },
};
I was able to successfully group each of them by date and by subject.
var allQuizzes = quizzes.GroupBy(qz => qz.quizDate, (q, values) =>
new
{
Date = q,
Quizzes = values.GroupBy(v => v.subjectId, (c, values2) =>
new {
SubjectId = c,
QuizSum = values2.Sum(v2 => v2.quizScore)
})
});
var allExercises = exercises.GroupBy(ex => ex.exerciseDate, (e, values) =>
new {
Date = e,
Exercises = values.GroupBy(x => x.subjectId, (z, values2) =>
new {
SubjectId = z,
ExerSum = values2.Sum(r => r.exerciseScore)
})
});
var allExams = exams.GroupBy(ex => ex.examDate, (e, values) =>
new
{
Date = e,
Exercises = values.GroupBy(x => x.subjectId, (z, values2) =>
new
{
SubjectId = z,
ExamSum = values2.Sum(r => r.examScore)
})
});
However, I need to join all three of them to get the sum of all scores. The final table should display like this.
-----------------------------------------------------------------
| Date | Math | Science | History | Language |
| 11/25/2016 | 132 | 111 | 0 | 92 |
| 11/26/2016 | 122 | 20 | 0 | 12 |
-----------------------------------------------------------------
I tried to join them, but it can't seem to join by multiple columns.
I select from all 3 collections results in form of the same anonymous class (the same Idea had Andrei in first answer), that allows me just to collect all results together in all list, without mapping and converting.
var allQuiz = quizzes.GroupBy(x => new { x.subjectId, x.quizDate })
.Select(x => new {
Date = x.Key.quizDate,
Subj = x.Key.subjectId,
Sum = x.Sum(r=>r.quizScore)});
var allExam= exams.GroupBy(x => new { x.subjectId, x.examDate })
.Select(x => new {
Date = x.Key.examDate,
Subj = x.Key.subjectId,
Sum = x.Sum(r=>r.examScore)});
var allExc = exercises.GroupBy(x => new { x.subjectId, x.exerciseDate })
.Select(x => new {
Date = x.Key.exerciseDate,
Subj = x.Key.subjectId,
Sum = x.Sum(r=>r.exerciseScore)});
Combining of all results together:
var all = allQuiz.ToList();
all.AddRange(allExam.ToList());
all.AddRange(allExc.ToList());
var result = all.GroupBy(x => new { x.Date, x.Subj })
.Select(x => new { x.Key.Date, x.Key.Subj, Sum = x.Sum(s => s.Sum)});
var list = result.GroupBy(r => r.Date).Select(x => new {
Date = x.Key,
Math = x.SingleOrDefault(t=>t.Subj==1)?.Sum ?? 0,
Science = x.SingleOrDefault(t=>t.Subj==2)?.Sum ?? 0,
History = x.SingleOrDefault(t=>t.Subj==3)?.Sum ?? 0,
Language = x.SingleOrDefault(t=>t.Subj==4)?.Sum ?? 0,
});
Output in LinqPad:
Here is an idea. Instead of keeping the distinction while grouping, you could convert all three to the same structure. For instance:
var allQuizzes = quizzes.GroupBy(qz => qz.quizDate, (q, values) =>
new
{
Date = q,
Results = values.GroupBy(v => v.subjectId, (c, values2) =>
new {
SubjectId = c,
Sum = values2.Sum(v2 => v2.quizScore)
})
});
Notice names "Results" and "Sum" - you can use the same for the other two objects. And now you have three collections, all of the same structure:
{
Date:
Results: [
{SubjectId, Sum}
{SubjectId, Sum}
...
]
}
Since they are all the same now, you can stop treating them differently, use UNION to merge all three, group them by date and within that by subject. Then you could probably iterate through subject list to get necessary info, depends on what you mean by "final table".
This is what i came up with.
It may not be best optimized, but might be enough for you.
I rendered the results into a StringBuilder in my test.
var result =
quizzes.Select(q => new {SubjectId = q.subjectId, Date = q.quizDate, Score = q.quizScore})
.Union(exams.Select(e => new {SubjectId = e.subjectId, Date = e.examDate, Score = e.examScore}))
.Union(exercises.Select(e => new {SubjectId = e.subjectId, Date = e.exerciseDate, Score = e.exerciseScore}))
.GroupBy(arg => arg.Date,
(key, values)=>
new
{
Key = key,
Scores = values.GroupBy(v => v.SubjectId, (s, values2) => new { SubjectId = s, SumScore = values2.Sum(v2 => v2.Score) })
});
StringBuilder sb = new StringBuilder("Date\t\t");
foreach (SubjectModel subject in subjects)
{
sb.Append($"{subject.subjName}\t");
}
sb.AppendLine();
foreach (var record in result)
{
sb.Append($"{record.Key.ToShortDateString()}\t");
foreach (SubjectModel subject in subjects)
{
int sum = record.Scores.Where(s => s.SubjectId == subject.subjId).Select(s => s.SumScore).DefaultIfEmpty(0).Single();
sb.Append($"{sum}\t");
}
sb.AppendLine();
}
string finalTable = sb.ToString();
Instead of using three different anonymous objects to hold the results, make your own class:
public enum TestType
{
Quiz,
Exam,
Exercise,
}
public class TestScore
{
public TestType Type { get; set; }
public DateTime Date { get; set; }
public int Score { get; set; }
public int SubjectId { get; set; }
// Constructors - make a TestScore object
public TestScore(QuizModel q)
{
Type = TestType.Quiz;
Date = q.quizDate;
Score = q.quizScore;
SubjectId = q.SubjectId;
}
public TestScore(ExamModel e)
{
Type = TestType.Exam;
Date = e.examDate;
Score = e.examScore;
SubjectId = e.SubjectId;
}
public TestScore(ExerciseModel e)
{
Type = TestType.Exercise;
Date = e.exerciseDate;
Score = e.exerciseScore;
SubjectId = e.SubjectId;
}
}
Convert to TestScore:
List<TestScore> scores = new List<TestScore>();
scores.AddRange(quizzes.Select(q => new TestScore(q));
scores.AddRange(exams.Select(e => new TestScore(e));
scores.AddRange(exercises.Select(e => new TestScore(e));
Now you have one datasource instead of three, displaying the results becomes easy.

Translate foreach to linq to solve running balance

I'm new in linq and read some stuff on the web about them.
Now, below is a query works fine which is to calculate the project 12-month running balance from the current date. Is it possible to translate this to linq?
It would help me understand the linq better.
var firstDayMonth = new DateTimeOffset(new DateTime(DateTime.Now.Year, DateTime.Now.Month, 1));
var months = Enumerable.Range(0, 12)
.Select(m => firstDayMonth.AddMonths(m));
List<SomeDate> SomeDates = new List<SomeDate>()
{
new SomeDate { Id = 7, Month = firstDayMonth.AddMonths(0), Balance = 1m },
new SomeDate { Id = 7, Month = firstDayMonth.AddMonths(0), Balance = 3m },
new SomeDate { Id = 8, Month = firstDayMonth.AddMonths(1), Balance = 6m },
new SomeDate { Id = 8, Month = firstDayMonth.AddMonths(1), Balance = 5m },
new SomeDate { Id = 8, Month = firstDayMonth.AddMonths(1), Balance = 3m },
new SomeDate { Id = 9, Month = firstDayMonth.AddMonths(2), Balance = 5m },
new SomeDate { Id = 10, Month = firstDayMonth.AddMonths(3), Balance = 3m },
new SomeDate { Id = 12, Month = firstDayMonth.AddMonths(5), Balance = 15m },
new SomeDate { Id = 13, Month = firstDayMonth.AddMonths(6), Balance = 16m },
new SomeDate { Id = 13, Month = firstDayMonth.AddMonths(6), Balance = 12m },
};
var projected12MonthsBalance = new List<SomeDate>();
foreach(var month in months)
{
projected12MonthsBalance.Add(new SomeDate { Month = month, Balance = SomeDates.TakeWhile(s => s.Month <= month).Sum(s => s.Balance) });
}
Console.WriteLine(projected12MonthsBalance);
public class SomeDate
{
public int Id { get; set; }
public DateTimeOffset Month { get; set; }
public decimal Balance { get; set; }
}
Try this:
var projected12MonthsBalance = months.Select(x => new SomeDate
{
Month = x,
Balance = SomeDates.TakeWhile(s => s.Month <= x).Sum(s => s.Balance)
}).ToList();

Group by using linq (range + count)

var data = new[] {
new { Id = 0, Cat = 1, Price = 2 },
new { Id = 1, Cat = 1, Price = 10 },
new { Id = 2, Cat = 1, Price = 30 },
new { Id = 3, Cat = 2, Price = 50 },
new { Id = 4, Cat = 2, Price = 120 },
new { Id = 5, Cat = 2, Price = 200 },
new { Id = 6, Cat = 2, Price = 1024 },
};
var ranges = new[] { 10, 50, 100, 500 };
Needed output is grouped price count by equal or greater than the range used according categories.
(in one linq statement)
cat range count
-------------------------------------
1 10 2 (In 1. categories there is 2 item that price >= 10(range) [10;30])
2 10 4 (In 2. categories there is 4 item that price >= 10(range) [50;120;200;1024])
2 50 4 ....
2 100 3 ....
2 500 1 (In 2. categories there is 1 item that price >= 500(range) [1024])
Try this:
var data = new[] {
new { Id = 0, Cat = 1, Price = 2 },
new { Id = 1, Cat = 1, Price = 10 },
new { Id = 2, Cat = 1, Price = 30 },
new { Id = 3, Cat = 2, Price = 50 },
new { Id = 4, Cat = 2, Price = 120 },
new { Id = 5, Cat = 2, Price = 200 },
new { Id = 6, Cat = 2, Price = 1024 },
};
var ranges = new[] { 10, 50, 100, 500 };
var result = from r in ranges
from g in data
where g.Price >= r
select new {g.Cat, Price=r};
var groupedData =
from d in result
group d by new{d.Cat, d.Price} into g
select new{Cat=g.Key.Cat, Price=g.Key.Price, TotalCount=g.Count()};
This should work:
var values =
data.SelectMany(x => ranges.Where(y => x.Price >= y)
.Select(y => new { Record = x, Range = y }))
.GroupBy(x => new { Cat = x.Record.Cat, Range = x.Range })
.Select(x => new { Cat = x.Key.Cat, Range = x.Key.Range, Count = x.Count()});
Results:
{ Cat = 1, Range = 10, Count = 2 }
{ Cat = 2, Range = 10, Count = 4 }
{ Cat = 2, Range = 50, Count = 4 }
{ Cat = 2, Range = 100, Count = 3 }
{ Cat = 2, Range = 500, Count = 1 }

Categories

Resources