skip header row when parsing .csv file

I’m using a StreamReader object to provide source data for the LinqtoCSV library. Something like this…

CsvFileDescription fileDescription = new CsvFileDescription() 
{ 
            SeparatorChar = ';', 
            MaximumNbrExceptions = 20,                 
};  
      开发者_运维技巧  CsvContext context = new CsvContext();           
        System.IO.MemoryStream stream = new System.IO.MemoryStream(file.Data); 
        System.IO.StreamReader reader = new System.IO.StreamReader(stream); 
        rows = context.Read<T>(reader, fileDescription); 

I’m processing csv files which have an extra semi colon in the header row e.g.

Header1; Header2;Header3;

This triggers errors because the CSVtoLinQ library is expecting 4 header names instead of 3. I can read selected rows into an IEnumerable object, but I can’t directly manipulate the contents of the StreamReader object. I’d prefer to continue with the combination of StreamReader/LinqtoCSV as it provides a lot of validation functionality, but because of the file header issue I need to look at alternatives.

Are there other csv import libraries that provide (i) field based validation when parsing a file (ii) a means of skipping the header row?

You could write a method that opens the file, checks the header/fixes it, then saves to a new file which now ready for the existing process.