We Keep Coding

How is it possible that selecting first after GroupBy on a DataTable does not return one value for every group?

I am trying to get one row per id from a DataTable, and I do not care which row I take. The same id can exist on several rows in the table.
Here's the expression that's giving me trouble:
dt.AsEnumerable().GroupBy(i => i.Field<int>("id")).Select(i => i.First())
Running just this section dt.AsEnumerable().GroupBy(i => i.Field<int>("id") correctly gives me a result of 22 groupings for my DataTable. (I have 22 ids with data in this table)
However, when adding on the .Select(i => i.First()), I am only seeing 10 data rows.
To me this doesn't seem to make any sense. If the GroupBy function managed to find 22 distinct id values, I would expect this logic to grab one of each.
My only other thought is that maybe it's just a weird side effect of viewing this data through a watch in Visual Studio rather than assigning to a variable.

If you think it's just weird side effects of viewing the data in a watch, which can happen with LINQ statements, then split it out into
var groups = dt.AsEnumerable().GroupBy(i => i.Field<int>("id")).ToList();
var firstOfGroups = groups.Select(i => i.First()).ToList();
and then look at groups and firstOfGroups in the debugger. Temporarily evaluating items with .ToList() can help a lot with viewing things in the debugger.

I think it is possible, can double check the count of each group items
.Select(g=>new { k = g.Key, c = g.Count() })

Get "lowest" date of parents great grand child with Linq

I'm trying to learn LINQ - the hard way.
I have a few entities which all are connected in sequence
Department -> Groups -> Works -> project
project has a startdate (and an end date)
I'm trying to get the startdate of the Group with the first startdate so to speak.
I've tried:
Department.Groups.Select(g => g.Works.Select(w => w.project.StartDate).Min())
and various variations thereof.
The problem being it returns a list in a list in a list, and I'm getting dizzy from just thinking about it :)
I've tried to work my way backwards from
g.Works.Select(w=> w.Project.StartDate).Min() which gives me the lowest date for Works
Any help is greatly appreciated

Flatten your List with SelectMany and then apply Min like:
var minDate = Department.Groups
.SelectMany(g => g.Works.Select(w => w.project.StartDate))
.Min();
That will return the minimum date for all the Works under all Groups.

If you just want the date then Habib's flatten solution is good. If you want the whole record you can order by and then take the first record, like this:
Department
.Groups
.OrderBy(g => g.Works.Select(w => w.project.StartDate))
.FirstOrDefault()

Linq using sum() in list within a list

The relationship between them is one order to many journals. Now I want to get the sum of all pending orders (or records that are flagged as false under IsDelivered in Order entity).
So far, I have this query but can't seem to get working when I add .Sum()
Orders
.Where(o => o.IsDelivered == false)
.Select(o => new {
pendingOrders = o.Journals.Sum(j => j.TotalAmount)
})
So far this results to:
In a nutshell, how can I get the sum of them? If query needs to be altered or should be a new one. It is welcome. Any help would be much appreciated. Thanks!

You can do it in several ways:
You could add Sum(n => n.pendingOrders) to the end of your query to add up the values, or
You could use SelectMany before Select, and use Sum instead of Select.
Either of the two approaches is going to work.

In this case, it is easiest to use SelectMany (MSDN) to flatten your collections, then use Sum:
Orders.Where(o => !o.IsDelivered).SelectMany(o => o.Journals).Sum(j => j.TotalAmount);

How to add empty groups to a Linq query result set?

I have a very simple query which selects items from a table based on matching the month and then grouping by day. These groups are then used as a data source for a repeater which outputs the group elements as entries "per day".
The problem is that days that don't exist (i.e. there's no group for) will naturally not be displayed, so things for the 6th and 8th, when there's nothing for the 7th, will be seen directly next to each other (think of a calendar view). The question is, given my query below, how could I insert groups with no elements, even when there's no entry for that day?
IQueryable events =
Events
.Where(i => i.Date.Month == date.Month)
.GroupBy(i => i.Date.Day);
I can do this figuring out after the fact, but can I account for it to get the result set at once? Or can a previous tried & tested approach be recommended?

Create your result set like this:
var date = ...;
var events = Enumerable.Range(1, DateTime.DaysInMonth(date.Year, date.Month))
.ToDictionary(
day => new DateTime(date.Year, date.Month, day),
day => new List<Event>());
Then insert into it like this:
var query = Events
.Where(e => e.Date.Year == date.Year)
.Where(e => e.Date.Month == date.Month);
foreach (var e in query)
events[e.Date].Add(e);
If you really want to do this server-side as part of the query, you'll either need to (A) send the list of dates you're interested in as part of the query, or (B) use DBMS-specific functions to construct the set of days you're interested in server-side.
Specific to (B) and SQL, there are things like this: Get the first and last date of next month in MySQL
I personally would just do this client-side. The database query gets whatever data your implementation needs to extract from the database; the user gets whatever information your view needs. To query the database for the data you need to create your view, you don't actually need to know the list of dates in your month, so it really shouldn't be part of the query.

LINQ: Getting the row with the maximum value of a given attribute

I have a bunch of rows grouped on an attribute called MyID. Now I want the one row from each group where the StatusDate attribute is the highest in that one group.
This is what I've come up with.
rows.Select(x => x.Where(y => y.StatusDate == x.Max(z => z.StatusDate)).First())
With a bit more explanation:
rows.Select(x => // x is a group
x.Where(y => // get all rows in that group where...
// the status date is equal to the largest
// status date in the group
y.StatusDate == x.Max(z => z.StatusDate)
).First()) // and then get the first one of those rows
Is there any faster or more idiomatic way to do this?

One alternative would be to use:
rows.Select(x => x.OrderByDescending(y => y.StatusDate).First());
... and check that the query optimiser knows that it doesn't really need to sort everything. (This would be disastrous in LINQ to Objects, but you could use MaxBy from MoreLINQ in that case :)
(Apologies for previous version - I hadn't fully comprehended the grouping bit.)

Don't know if this is Linq to SQL, but if it is, you could alternatively accomplish via a rank() function in SQL (rank each group by date, then select the first ranked row from each), then call this as a stored proc from LINQ. I think that's an approch that is becoming more idiomatic as people hit the bounderies of LINQ2SQL...