Develop Reference

SQL to LINQ - Getting Wrong Results

If a user selects 07-07-2016, I want to count how many times billdays > 30 for that billdate month, for the past 12 months.
Query is this:
select COUNT(*) as 'BillsOver30' from pssuite_web.pujhaccd
where billdays>30 and DATEDIFF(month,'07-07-2016', GETDATE()) <= 13
Group By Month(billdate)
Result is this:
1784 (July)
1509 (June)
2986 (May)
2196 (etc)
5853
3994
1753
1954
869
1932
629
1673
LINQ query is:
DateStart = '07-07-2016' (from a textbox on view)
DateTime earliestDate = objdate1.DateStart.Value.AddMonths(-13);
var custQuery9 = (from m in DataContext.pujhaccds
where m.billdays > 30 &&
m.billdate >= earliestDate &&
m.billdate <= objdate1.DateStart
group m by m.billdate.Value.Month into p
select new Date1()
{
theMonth = p.Key,
theCount = p.Count()
});
Results are:
Month Count
1 1029
2 1018
3 1972
4 1519
5 2657
6 2019
7 1206
8 1023
9 761
10 1620
11 354
12 931
You can see that the LINQ query is way off.
Doesn't matter what date I put in, the result always stays the same. Rather than 12 months from starting date.
I must be writing it wrong, thanks.

Lambda Linq Iqueryable group - add another grouping

I have a method that I use for generating a summary report, based on the options a user selects in pick lists, to count the number of a type of test that is completed in a region, with subset locations, within a date range.
Here's what the data looks like:
Completed Test Counts by Location
From 1 FEB 2015 to 1 MAR 2015
Total TestType Location Region Division
455 24 Hour Lab 1 City 1 Division A
28 24 Hour Lab 2 City 1 Division A
95 24 Hour Clinic Z City 2 Division A
189 24 Hour Clinic Y City 2 Division A
Here's what it might look like for a different test type, at the same time period:
Total TestType Location Region Division
285 48 Hour Lab 1 City 1 Division A
12 48 Hour Lab 2 City 1 Division A
75 48 Hour Clinic Z City 2 Division A
106 48 Hour Clinic Y City 2 Division A
Now, the users would like to see the summary breakdown by testType in one report (in SQL, adding another attribute to a group by). In my pick list of testTypeId, I am using 0 as my "All" item. In a perfect world, I'd have another addition to the IQueryable if testTypeId == 0, so that I can keep the same private method, rather than writing a new query.
Here's what we'd like the data to look like in the "ALL" situation:
Total TestType Location Region Division
455 24 Hour Lab 1 City 1 Division A
285 48 Hour Lab 1 City 1 Division A
59 12 Lead Lab 1 City 1 Division A
28 24 Hour Lab 2 City 1 Division A
12 48 Hour Lab 2 City 1 Division A
95 24 Hour Clinic Z City 2 Division A
75 48 Hour Clinic Z City 2 Division A
5 12 Lead Clinic Z City 2 Division A
189 24 Hour Clinic Y City 2 Division A
106 48 Hour Clinic Y City 2 Division A
8 12 Lead Clinic Y City 2 Division A
The example below shows what I have with what I'd like to add in it, but the syntax tells me I can't take an IQueryable containing an IGrouping and convert it to the target type of IQueryable.
Can anyone point me in a good direction here?
private CompletedCountReport GetCompletedCountsByRegion(int regionId, DateTime? startDate, DateTime? endDate, int testTypeId)
{
var ccRpt = new CompletedCountReport { CompletedCounts = new List<CompletedCount>(), StartDate = startDate, EndDate = endDate, SelectedRegionID = regionId};
var query = HolterTestDao.FindAll(new GetCompletedByRegionIdAndDates(regionId, startDate.Value, endDate.Value));
// a specific test type was selected
if (testTypeId > 0)
{
query = query.Where(x => x.HolterTestType == testTypeId);
}
// THIS IS WHAT I WANT TO ADD -> otherwise group by test type <- THIS IS WHAT I WANT TO ADD.
if (testTypeId == 0)
{
query = query.GroupBy(x => x.HolterTestType);
}
// order by locationID within the region.
query = query.OrderBy(x => x.Location.ID);
var htList = query.ToList();
// now Group by Location, to get counts.
var reportContents = htList.GroupBy(x => x.Location.ID);
foreach (var r in reportContents)
{
var rList = r.ToList();
var cc = new CompletedCount();
var loc = LocationDao.FindById(r.Key);
cc.Description = loc.Description;
cc.RegionId = loc.Region.ID;
cc.DivisionId = loc.Region.Division.ID;
cc.TestTypeId = testTypeId;
cc.Count = rList.Count;
ccRpt.CompletedCounts.Add(cc);
}
return ccRpt;
}

My solution was to OrderBy Location and HolterTestType, then GroupBy location for the basic contents, and then add another grouping in my loop through the results to get the counts by test type:
query = query.OrderBy(x => x.Location.ID).ThenBy(x => x.HolterTestType);
var htList = query.ToList();
// now Group by Location.
var reportContents = htList.GroupBy(x => x.Location.ID);
foreach (var r in reportContents)
{
//Group by TestType, to get counts.
var t = r.GroupBy(x => x.HolterTestType);
var tList = t.ToList();
foreach(var u in tList)
{
var cc = new CompletedCount();
var loc = LocationDao.FindById(r.Key);
cc.Description = loc.Description;
cc.RegionId = loc.Region.ID;
cc.DivisionId = loc.Region.Division.ID;
cc.TestTypeId = u.Key;
cc.Count = u.Count();
ccRpt.CompletedCounts.Add(cc);
}
}

How to select hour in priority

I'm using C# and SQL Server 2008 R2.
I have hour data like this:
Client_ID TheHour HourMin HourMAx
53026 09:00 7 9
53026 12:00 10 12
53026 15:00 13 15
53026 18:00 16 19
I will put TheHour into my combobox, that depends on computer hour.
When the computer hour is 10:00 then the value of my combobox:
12:00
15:00
18:00
09:00
My linq is :
int The_Hour = DateTime.Now.Hour;
var query = from o in oEntite_T.LS_CLIENTHORRAIRE
where o.CLIENT_ID == CLIENT_ID &&
The_Hour >= o.HEURE_MIN && The_Hour <= o.HEURE_MAX
select o;
LesListe = query.ToList();
It return only 1 value, which is 12:00.
That meant my combobox will select hour that depend on computer hour, but it leave the possibility for the user to select another hour.

Here is a working example in LINQPad, I also added a flag to indicate selected value.:
void Main()
{
var LS_CLIENTHORRAIREs = new List<LS_CLIENTHORRAIRE>
{
new LS_CLIENTHORRAIRE{TheHour="09:00",HourMin=7,HourMAx=9},
new LS_CLIENTHORRAIRE{TheHour="12:00",HourMin=10,HourMAx=12},
new LS_CLIENTHORRAIRE{TheHour="15:00",HourMin=13,HourMAx=15},
new LS_CLIENTHORRAIRE{TheHour="18:00",HourMin=16,HourMAx=19}
};
LS_CLIENTHORRAIREs.Dump();
int The_Hour = 10;
var query = from o in LS_CLIENTHORRAIREs
select new {o.TheHour,o.HourMin,o.HourMAx,selected = The_Hour >= o.HourMin && The_Hour <= o.HourMAx}
;
query.OrderBy (q => q.HourMAx>The_Hour ? 0 : 1).ThenBy (q => q.HourMAx).Dump();
}
public class LS_CLIENTHORRAIRE
{
public string TheHour{get;set;}
public int HourMin{get;set;}
public int HourMAx{get;set;}
}
Results

So, basically, you don't want to filter the list, you want to sort it, right?
First, you want to sort it by
whether the MaxHour is in the past
and then by
TheHour itself.
Example: Let's assume it is 10:00:
Client_ID TheHour HourMin HourMax MaxHourIsInThePast
53026 09:00 7 9 x
53026 12:00 10 12
53026 15:00 13 15
53026 18:00 16 19
Now you return a list that orders first by MaxHourIsInThePast (first those without x and then those with x) and then by TheHour.
Implementation is left as an exercise. Hint: You will need the LINQ order by keyword, and an expression similar to (o.HourMax < currentHour ? 2 : 1) will come in handy.

Let assume the computer hour is 10.
Then: your condition would be: HourMin <= 10 && HoureMax >= 10.
The first condition would return row 1 and 2. The second one would return row 2. The intersect will return only the second row, which is 12:00.
If you want to get all rows, just remove these conditions.

C# LINQ query to use previous result if empty

I am having a datatable that is populated from an Access DB. The result looks like
Month | Sum
--------------
1 | 1464
2 | 1716
3 | 2125
4 | 2271
5 | 2451
6 | 2583
7 | 2671
9 | 2823
10 | 2975
You are right - nothing for august!
What I want is, that for august the same value as for july is used.
Currently I am using this LINQ query to add the data to a linechart:
for (int i = 1; i <= System.DateTime.Now.Month; i++)
{
var numbers = (from p in dTable.AsEnumerable()
where p.Field<int>("M") >= i
select p).First();
series2.Points.Add(new DataPoint { AxisLabel = i.ToString(), YValues = new double[] { Convert.ToDouble(numbers["Sum"]) } });
}
The chart is shown, but for august the september value is used. I assume it is something very basic that I am doing wrong but I simply can´t figure it out.
Thanks in advance!

You are requesting all the months greater than the current month.
where p.Field<int>("M") >= i
So for August (8), your are retrieving September and greater (9, 10, 11, 12), not July (7).
You have to invert your constraint, and order by descending month:
var numbers = (from p in dTable.AsEnumerable()
where p.Field<int>("M") <= i
select p)
.OrderByDesc(p => p.Month)
.First();

You have to invert your logic:
var numbers = (from p in dTable.AsEnumerable()
where p.Field<int>("M") <= i
select p).Last();
It goes without saying that this doesn't work when there is no previous month.
UPDATE:
the above asumes that the table you are reading from is ordered. If that is not the case, you have to order yourself (as mentioned by Cyril Gandon):
var numbers = (from p in dTable.AsEnumerable()
where p.Field<int>("M") <= i
orderby p.Field<int>("M") descending
select p).First();

Put value in datatable group by month

I have this datatable in c#
Date Employee Job1 Job2 Job3
1/1/2012 a 1 1 1
1/1/2012 b 2 2 2
1/1/2012 c 2 1 4
1/1/2012 d 4 2 1
1/2/2012 a 3 2 5
1/2/2012 b 2 2 2
1/2/2012 c 3 3 3
1/2/2012 d 1 1 1
1/3/2012 a 5 5 5
1/3/2012 b 2 2 6
1/3/2012 c 1 1 1
1/3/2012 d 2 3 4
2/1/2012 a 2 2.5 2
2/1/2012 b 5 5 2
2/1/2012 c 2 2 2
2/2/2012 a 3 3 3
2/2/2012 b 2 3 3
3/1/2012 a 4 4 11
3/5/2012 a 14 42.5 15
3/6/2012 a 21 12.143 22
3/8/2012 a 8.9 45 27
3/8/2012 b 4.4 25 31
I want to loop through the values monthwise such that i can store the value in a datatable and will do the other calculation with that. Here with this example it will have three datatable, first with january values, another one with feb and last one with march rows. How can this be done by Linq. Please suggest the Linq syntax which will group the results month wise.
var monthEmpGroups = dt.AsEnumerable()
.Select(r => new
{
Year = DateTime.Parse(r.Field<string>("DATE")).Year,
Month = DateTime.Parse(r.Field<string>("DATE")).Month.ToString()
})
.GroupBy(x => x.Month);
var dtf = CultureInfo.CurrentCulture.DateTimeFormat;
foreach (var empGroup in monthEmpGroups)
{
int month = int.Parse(empGroup.Key);
string colName = dtf.GetMonthName(month);
// Here i want to get all the rows where month is colName (i.e.january, feb, march)
}
Please suggest if there is other way around to get the values month wise using LINQ.

You should move getting items for every group into monthEmpGroups query:
var monthEmpGroups = (from r in dt.AsEnumerable()
let month = DateTime.Parse(r.Field<string>("DATE")).Month
group r by month into g
select new { Month = g.Key, Items = g.ToList() });
With that you can easily get desired results:
var dtf = System.Globalization.CultureInfo.CurrentCulture.DateTimeFormat;
foreach (var empGroup in monthEmpGroups)
{
int month = empGroup.Month;
string colName = dtf.GetMonthName(month);
List<DataRow> items = empGroup.Items;
}