Develop Reference

Linq Update - Cannot perform runtime binding on a null reference

I'm using LinqToExcel along with C# to read data from a MS Excel spreadsheet and then update data records in my MS SQL database.
The Excel file has these headers: COURSE_ID, PROVIDER_COURSE_TITLE
My code is like this:
public class TestDataCourse
{
[ExcelColumn("PROVIDER_COURSE_TITLE")]
public string cTitle
{
get;
set;
}
}
///////////////////////////////
string pathToExcelFile = #"C:\\O_COURSES.xlsx";
ConnexionExcel ConxObject = new ConnexionExcel(pathToExcelFile);
//read data from excel
var query1 = (from a in ConxObject.UrlConnexion.Worksheet<TestDataCourse>("O_COURSES")
select a).Take(2000).ToList();
//Get data from MS SQL database that need updated
var courses = _courseService.GetAllCoursesFromDB().Take(100).ToList();
int count = 0;
foreach (var course in courses)
{
//Iterate through the excel doc and assign the
TestDataCourse fakeData = query1.Skip(count).Take(1).FirstOrDefault();
course.CourseTitle = fakeData.cTitle;
count++;
}
_courseService.Save();
When I run this code I can see that it does update some of the records in my database, but as the code execution continues, I get a Source Not Available tab open within my Visual Studio and a Cannot perform runtime binding on a null reference.
The null reference exception had me thinking that maybe there was some null data in the Excel doc, so I put this line of code into my for loop:
course.CourseTitle = fakeData == null ? "Course Test" : fakeData.cTitle;
But I still get the same problem.
Could anyone please help?
Thanks.

Send the contents of a datatable to a csv file with headers

I am trying to send the contents of a datatable to a csv file, with headers. There is a duplicate question but the the accepted answers only seem to work half way. At this point I have mixed and matched the upvoted answers with no luck and need a point in the right direction.
I can write the columns to the file just fine, and I can write data just fine but not together. Also the data never comes out quoted, only comma delimited without quotes.
//This is how the FileHelpers class is built
public class ScannedFileInfo
{
//this prefix will handle the double quotes issue
//delimiters must reside between double quotes
//Must specify FieldOrder too
[FieldQuoted('"', QuoteMode.AlwaysQuoted)]
[FieldOrder(1)]
public string DA_Id;
[FieldQuoted('"', QuoteMode.AlwaysQuoted)]
[FieldOrder(2)]
public string Name;
[FieldQuoted('"', QuoteMode.AlwaysQuoted)]
[FieldOrder(3)]
public string Extension;
[FieldQuoted('"', QuoteMode.AlwaysQuoted)]
[FieldOrder(4)]
public string Fullname;
[FieldQuoted('"', QuoteMode.AlwaysQuoted)]
[FieldOrder(5)]
public string PathLength;
[FieldQuoted('"', QuoteMode.OptionalForBoth)]
[FieldOrder(6)]
public string Directory;
}
//this is how I send it to the file
public static void ImportDirectory(string result, SqlConnection myconn, SqlConnection destConn ,ListBox lbInfo, DataGridView dgView)
{
//create data table code here - works fine...
MessageBox.Show("Scan complete. Starting Import...");
//build file
var engine = new FileHelperEngine<ScannedFileInfo>();
var orders = new List<ScannedFileInfo>();
engine.HeaderText = engine.GetFileHeader();
engine.WriteFile(#"C:\DirectoryScan.csv", orders);
MessageBox.Show("You now have proper labeled columns in your file.");
//now the data import is successful but it overwrites the column info
CommonEngine.DataTableToCsv(dt, #"C:\DirectoryScan.csv", ',');
MessageBox.Show("You now have data in your file, but no columns");
}

The CommonEngine.DataTableToCsv() call does not support all of the features of FileHelpersEngine. In particular, the field quoting and the column headers are missing. The CommonEngine is not as full featured as the FileHelpersEngine<T>.
Instead, you should create a list of ScannedFileInfo instead of a datatable and then use FileHelpersEngine<ScannedFileInfo>.WriteFile().
//create list of ScannedFileInfo here
var records = new List<ScannedFileInfo>();
records.Add(new ScannedFileInfo() { Name = ..., etc... });
// ( or if you prefer, loop through your datatable to build the list)
MessageBox.Show("Scan complete. Starting Import...");
// build file
var engine = new FileHelperEngine<ScannedFileInfo>();
engine.HeaderText = engine.GetFileHeader();
engine.WriteFile(#"C:\DirectoryScan.csv", records);
MessageBox.Show("You now have data and properly labelled columns in your file.");

I cannot recommend that you look into CSVHelper enough. This nuget package has saved me from pulling my hair out on so many occasions, I can't count them all.
If you're into using it, you can use the configuration settings to quickly set a header record, delimiter, ignore quotes or not, etc.

LinqToExcel Duplicate Column Names

I have a machine generated excel file that has a few columns with the same name. e.g.:
A B C D
Group 1           Group 2
Period | Name     Period | Name
And i got a DTO like this:
[ExcelColumn("Period")]
public string FirstPeriod { get; set; }
[ExcelColumn("Name")]
public string FirstName { get; set; }
[ExcelColumn("Period")]
public string SecondPeriod { get; set; }
[ExcelColumn("Name")]
public string SecondName { get; set; }
I use the following command to read the lines:
var excel = new ExcelQueryFactory(filePath);
excel.WorksheetRange<T>(beginCell, endColl + linesCount.ToString(), sheetIndex);
It reads the file just fine, but when i check the content of my DTO i saw that all the 'Second' properties have the same values of the 'First' ones.
This post was the closest thing that i found in my searches and i think the problem could be solved with something like this:
excel.AddMapping<MyDto>(x => x.FirstPeriod, "A");
excel.AddMapping<MyDto>(x => x.FirstName, "B");
excel.AddMapping<MyDto>(x => x.SecondPeriod, "C");
excel.AddMapping<MyDto>(x => x.SecondName, "D");
But i don't know how to get the excel column letters...
Obs: I got a few more code behind this but i don't think its relevant to the problem.

The problem that you're having is not possible to solve today with LinqToExcel because it wraps the OleDb functions and then they map properties based on columns names, so you lose the OleDb options like "FN" for specify columns (like "F1" for "A").
There's a issue on LinqToExcel github repo about this. https://github.com/paulyoder/LinqToExcel/issues/85
I recommend you to change the name of the columns to not duplicates names (e.g. Period1, Name1, Period2, Name2) if it's not possible to change because its machine generated, try change the header names in runtime.
Another option is to make more than one query in excel file, with ranges splitted your groups and then merging the results later.
var excel = new ExcelQueryFactory(filePath);
var group1 = excel.WorksheetRange<T>(A1, B + rowCount);
var group2 = excel.WorksheetRange<T>(C1, D + rowCount);
Edit: I'll work on a feature to try solve this problem in a elegant manner, so maybe in future you have a more flexible option to map columns and properties (if they accept my Pull Request)

LinqToExcel returns null

I have an excel sheet of xls format, I am using LinqToExcel to read it and then import it to my DB.
This sheet consist of about 3K row and only 6 cols. I am using .addmapping to map my class properties to the column names
The problem i have is: the cells of column "web-code" are SOMETIMES coming back as null although there are data in the cells.
Here is a sample data that is coming as null!
My Code watch
And here is a sample data where the data coming correct:
My Code Watch
I have tried applying ExcelColumn attribute for mapping, but no luck!
code:
var factory = new ExcelQueryFactory(_excelFilePath);
factory.AddMapping<ExcelPriceEntity>(x => x.WebCode, "WEB-CODE");
factory.AddMapping<ExcelPriceEntity>(x => x.Type, "TYPE");
factory.AddMapping<ExcelPriceEntity>(x => x.Style, "STYLE");
factory.AddMapping<ExcelPriceEntity>(x => x.Qty, "QTY");
factory.AddMapping<ExcelPriceEntity>(x => x.UnitPrice, "Unit Price");
factory.AddMapping<ExcelPriceEntity>(x => x.Bucket, "WEBCODE W/BUCKET");
factory.StrictMapping = StrictMappingType.ClassStrict;
factory.TrimSpaces = TrimSpacesType.Both;
factory.ReadOnly = true;
var prices = factory.Worksheet<ExcelPriceEntity>(_allPricesSheetName).ToList();
var priccerNP = prices.Where(p => p.Type.Contains("900 ARROW TAPE")).ToList();
My PriceEntity Class:
public class ExcelPriceEntity
{
//[ExcelColumn("TYPE")]
public string Type { get; set; }
public string WebCode { get; set; }
//[ExcelColumn("STYLE")]
public string Style { get; set; }
//[ExcelColumn("QTY")]
public string Qty { get; set; }
//[ExcelColumn("Unit Price")]
public string UnitPrice { get; set; }
//[ExcelColumn("WEBCODE W/BUCKET")]
public string Bucket { get; set; }
}

Alternate Solution:
I ended up saving the excel sheet as csv file, then import to SQL table.Then i used linq-to-sql to read the data.
Root Cause:
After researching i found out the problem was that the first cell of this column(web-code) was interger number, and excel trys to figure out the datatype of the column by looking at the first rows!
So next rows of (web-code) column was some text data. So excel couldn't parse it as integer, and assign null value to it!
What I could've done is, assing text value to the first cell so excel would guess the data type as string. But I didn't test that. For anyone reading this answer, try having text value in you first row if you came across the same problem

here, the Contains is not like string contains. It compares a list of cell values to the exact value u give inside the contains method. just try with the full text "900 AMMONIA STOCK OR CUST...)

Another alternative solution to #alsafoo is to convert the column from "general" to "text".
These are the steps:
1. Right click on any cell in the column.
2. In Number tab, select text.
3. Select Format cell
4. Press Ok.
After then, the library will read all values as string.

exporting a comma separated values text file to use in excel

In my C# winforms program I want to make an export function that will create a comma separated text file or csv. I am not sure about the logic of how to do this the best way. My exported file will be like this :
Family Name, First Name, Sex, Age
Dekker, Sean, Male, 23
Doe, John, Male, 40
So the first line I want to be the name of the columns, and the rest should be treated as values. Is it ok in this way for later usage? Or I should not include column names?
Would be nice to hear your experiences about this!

Sean,
sorry don't have enough privilege points to comment directly on your post. I think you may be confusing CSV and Excel files here. A CSV is simply a text file where each value is separated by a comma, there is no special formating etc. Excel will display CSV files since it knows how to open them but you can just as easily open them in notepad.
Excel .xslx files are different and can contain all sorts of different formats, charts etc. To format these files its important to understand that .xslx files are essentially zips. So the first place to start is to create an excel file with some data, save it and then rename the extension to .zip
Open the zip file created now and you will see a number of different folders and files, of these the most important for your purposes is the XL directory. In this folder you will see a shared strings xml file and a worksheets folder.
Lets start by going into the worksheet folder and opening sheet1.xml. Look for the line that says "
If there is text in this column, i.e. data that excel should read as text then you will have something like 0. This indicates that cell A1 is of type string t="s" and that the value is to be found as the first value in the SharedStrings.xml file 0
If there is a number in the cell then you may have something like 234. In this case Excel knows to use the value 234 in this cell.
So in your case you will need to do the following:
1: create the excel document in C# - there are a number of libraries available for this
2: Open the excel file as a zip
3: Modify in your case the styles and worksheets xml files
4: Save the document

That is absolutely fine to do (to state the obvious....). Excel has a little checkbox that allows the user importing to use the first line as column headers if they select it.
I would also suggest that you leave out the spaces at the start of each piece of data, it isn't necessary.

In general its best practice to include the column headers, the only reason not to do so would be an external program over which you have no control accessing your data which doesn't realise that the first row are the column headers and which can't be changed.
To create the export function something like this should work:
private static List<Person> people = new List<Person>();
static void Main(string[] args)
{
// add some people
people.Add(
new Person() { firstName = "John", familyName = "Smith", sex = Sex.Male, age = 12 }
);
people.Add(
new Person() { firstName = "Mary", familyName = "Doe", sex = Sex.Female, age = 25 }
);
// write the data
Write();
}
static void Write()
{
using (TextWriter tw = new StreamWriter(#"c:\junk1\test.csv", false))
{
// write the header
tw.WriteLine("Family Name, First Name, Sex, Age");
// write the details
foreach(Person person in people)
{
tw.WriteLine(String.Format("{0}, {1}, {2}, {3}", person.familyName, person.firstName, person.sex.ToString(), person.age.ToString()));
}
}
}
}
/// <summary>
/// Applicable sexes
/// </summary>
public enum Sex
{
Male,
Female
}
/// <summary>
/// holds details about a person
/// </summary>
public class Person
{
public string familyName { get; set; }
public string firstName { get; set; }
public Sex sex { get; set; }
public int age { get; set; }
}

You can use Dataset to do this.
Please refer here

//Why not save the lines to a List<string> object, first List<sting>Object.Add//("your headers"), use the string.Join("," "your Header Array string[]" do not add //(+",") the .Join extension menthod will handle that for you.
// here is an example
//if you were to retreive the header values from a database using a SQL Reader
var reader = sqlcmdSomeQueryCommand.ExecuteReader();
var columns = new List<string>();
//get all the field names from the Columns var
for (int intCounter = 0; intCounter < reader.FieldCount; intCounter++)
{
columns.Add(reader.GetName(intCounter));
}
strarryTmpString = columns.ToArray();
string TmpFields = string.Join(", ", strarryTmpString);
columns.Clear();
columns.Add(TmpFields);
//you can save the TmpFieldList to later add the rest of your comma delimited fields
write line by line in a foreach loop or use the List<string> object .foreach(delegate(string delString)
{
someStreamWriterObject.WriteLine(delString)
});