CSV to object model mapping
You can use a simple code like this, which ignores the header and doesn't work with quotes, but may be sufficient for your needs.
from line in File.ReadAllLines(fileName).Skip(1)
let columns = line.Split(',')
select new
{
Plant = columns[0],
Material = int.Parse(columns[1]),
Density = float.Parse(columns[2]),
StorageLocation = int.Parse(columns[3])
}
Or you can use a library, like others suggested.
For the specific data shown in your question...
var yourData = File.ReadAllLines("yourFile.csv")
.Skip(1)
.Select(x => x.Split(','))
.Select(x => new
{
Plant = x[0],
Material = x[1],
Density = double.Parse(x[2]),
StorageLocation = int.Parse(x[3])
});
If you already have a type declared for your data then you can use that rather than the anonymous type.
Note that this code isn't robust at all. It won't correctly handle values containing commas/newlines etc, quoted string values, or any of the other esoteric stuff that is often found in CSV files.
There's a codeproject for that:
http://www.codeproject.com/KB/linq/LINQtoCSV.aspx
However, you may want to look here:
Linq and streamreader getting lines
I wrote a simple library to allow developers to use LINQ on CSV files. Here is my blog post about it: http://procbits.com/2010/10/11/using-linq-with-csv-files/
In your case, you would have to change your header string to look like this:
Plant,Material,DensityLbft3,StorageLocation
And then you could parse the file like this:
var linqCSV = new CsvToXml("csvfile", true);
linqCsv.TextQualifier = null;
linqCsv.ColumnTypes.Add("Plant", typeof(string));
linqCsv.ColumnTypes.Add("Material", typeof(int));
linqCsv.ColumnTypes.Add("DensityLbft3", typeof(double));
linqCsv.ColumnTypes.Add("StorageLocation", typeof(int));
linqCsv.Convert();
You could then use LINQ like this:
var items = from item in linqCsv.DynamicRecords
where item.Plant == "Fred" && item.DensityLbft3 >= 62.6
orderby item.StorageLocation
select item;
Hope that helps or works for you.