AsciiImportExport v0.24

Overview

A .NET library providing fast and easy de/serialization of arbitrary column-based text data.

Changelog

• v0.24 - Now supports export of quoted strings, so column separators in columns are not a problem anymore.
• v0.23 - Bug fix for default string formats
• v0.22 - Now supports creation of columns with PropertyInfo, which enables automatic document format building
• v0.21 - Configurable fill char, line end trim made optional, max column width validation
• v0.20 - Now supports types without default constructor, IFormatProvider for columns, better import performance, small Bugfixes
• v0.12 - Custom export functions now have access to the item to export
• v0.11 - Improved performance and memory usage of import with the use of TextReader
• v0.10 - Added optional praefix for header line when exporting
• v0.9 - Hiding DocumentColumn and DocumentFormatDefinition behind interfaces
• v0.8 - Fixed column default value problems, added flag that indicates if exported lines terminate with the column separator, other small tweaks
• v0.7 - Now supports all built-in value types, Massive performance improvements
• v0.6 - Fixed some smaller bugs, minor restructuring of public surface
• v0.5 - Cleaned up column handling, fixed problems with comments, changed target framework to .NET 3.5 Client Profile
• v0.4 - Resolved some initialization problems in DocumentColumn
• v0.3 - Added comments, cleaned method names
• v0.2 - Added DocumentColumnBuilder for better separation of concerns
• v0.1 - Initial release

Example

POCO

public class Person
{
    public DateTime Birthday { get; set; }
    public Gender Gender { get; set; }
    public double Height { get; set; }
    public string Memo { get; set; }
    public string Name { get; set; }
}

public enum Gender
{
    Unknown,
    Female,
    Male,
}

Data

This is our list:

public List<Person> GetPersonList()
{
    return new List<Person>
                {
                    new Person {Birthday = new DateTime(1983, 1, 29), Gender = Gender.Male, Height = 175.5, Name = "Peter", Memo = "Nice guy!"},
                    new Person {Birthday = new DateTime(1931, 10, 5), Gender = Gender.Male, Height = 173.45, Name = "Paul", Memo = "Sometimes a litte grumpy."},
                    new Person {Birthday = new DateTime(1980, 4, 12), Gender = Gender.Female, Height = 1193, Name = "Mary", Memo = "Tall!"},
                };
}

Definition

Now we define how the data will be exported / imported:

private static IDocumentFormatDefinition<Person> GetDefinition()
{
    return new DocumentFormatDefinitionBuilder<Person>("\t", true)
        .SetCommentString("#")
        .SetExportHeaderLine(true, "# ")
        .AddColumn(x => x.Name)
        .AddColumn(x => x.Gender, b => b.SetImportFunc(StringToGender).SetExportFunc(GenderToString))
        .AddColumn(x => x.Height, "0.00", b => b.SetAlignment(ColumnAlignment.Right))
        .AddColumn(x => x.Birthday, "yyyyMMdd")
        .AddColumn(x => x.Memo, b => b.SetHeader("Description"))
        .Build();
}

private static string GenderToString(Person p, Gender gender)
{
    switch (gender)
    {
        case Gender.Female:
            return "F";
        case Gender.Male:
            return "M";
        default:
            return "U";
    }
}

private static Gender StringToGender(string s)
{
    switch (s)
    {
        case "M":
            return Gender.Male;
        case "F":
            return Gender.Female;
        default:
            return Gender.Unknown;
    }
}

That is what we defined:

• seperate columns by Tabs
• automatically determine the width of a column
• lines starting with '#' are recognized as comments
• generate a header line starting with '# '
• add new column for property Name
• add new column for property Gender and use special import / export methods
• add new column for property Height. Align the data of the column to the right and use a double precision of 2.
• add new column for property Birthday with a custom date format
• add new column for property Memo and set header to "Description"

Export

And now we can export the list:

public string Export(List<Person> list)
{
    IDocumentFormatDefinition<Person> definition = GetDefinition();
    return definition.Export(list);
}

Which will yield:

# Name	Gender	 Height	Birthday	Description
Peter 	M     	 175.50	19830129	Nice guy!
Paul  	M     	 173.45	19311005	Sometimes a litte grumpy.
Mary  	F     	1193.00	19800412	Tall!

Import

No surprises here!

public void ExportAndImport(List<Person> list)
{
    IDocumentFormatDefinition<Person> definition = GetDefinition();
	
	string exportResult = definition.Export(list);
    
	List<Person> importResult = definition.Import(new StringReader(exportResult));
}