Reputation: 32798
Following up for my workmate I have the following class:
public Pages {
primary_key { get; set; }
}
Data for the primary key looks like this:
010000100001
010000100001
010000100001
010000100002
010000100002
010000100002
010000200003
010000200003
020000300004
020000300005
I am trying to understand how I can group this data using LINQ. What I need to do is to group by:
Column 1-2 (2 columns that I'll call subject)
Column 3-7 (5 columns that I'll call chapter)
Column 8-12 (5 columns that I'll call book)
Every time the subject changes I need to get a count of how many chapters and how many books are in that subject.
There was a suggestion link this. It gives me some ideas but I am still left confused.
from page in pages
group page by new {
page.subjectId
, page.bookId
, page.chapterId
} into group
select new {
group.key.SubjectId
, group.key.bookId
, group.Key.chapterId
, total = pages.sum(s => s.Page)
}
I tried to implement this as:
var a = from Page in rowData
group /*Page*/ by new {
SubjectId = Page.PartitionKey.Substring(0,2),
chapterId = Page.PartitionKey.Substring(2,6),
bookId = Page.PartitionKey.Substring(8)
} into group
select new {
group.key.SubjectId
, group.key.bookId
, group.Key.chapterId
, total = rowData.sum(s => s.Page)
};
However I am getting an error message saying:
Error 1 Invalid anonymous type member declarator. Anonymous type members must be declared with a member assignment, simple name or member access.
Upvotes: 3
Views: 5145
Reputation: 392954
This should get you the idea:
new [] {
"010000100001", "010000100001", "010000100001", "010000100002",
"010000100002", "010000100002", "010000200003", "010000200003",
"020000300004", "020000300005" }
.GroupBy(s => new {
Subject = s.Substring(0,2),
Chapter = s.Substring(2,6),
Book = s.Substring(8) });
Edit See a demo live on https://ideone.com/A6RQU
Upvotes: 4