File System Types in .Net

Question

I don't get the abstractions and the terminology :-(

For example, DirectoryInfo.FullName is defined as the full path of the directory or file, but it's a string! So is DirectoryInfo.Name, FileInfo.FullName, Path.GetDirectoyName and so on.

This means that in .Net there is no "depth" (or "meat" - my English isn't so good) for the file system objects. There's no protection from a type system. I can't, for example, define two Path objects and ask if one of them is "above" the other - I have to manipulate the strings. I can't differentiate between a Path that identifies a directory and a path that identifies a file. I can't do anything!-( Just manipulate strings.

Is this correct (or am I simply missing something). If correct, are there any alternatives?

Answer 1

Take a look at System.IO.Path. To find out if you have a file, you can do Path.GetFileName("C:\\test\\") == String.Empty. Or if you are sure the file has an extension you can use Path.HasExtension.

Answer 2

You should take a look at the Path class and the Uri class.

MSDN links:

Path Class

Uri Class

For example with the Path class you have methods to get only the directory, only the filenam or even only the filename without the extension for a given path.

With the Uri class you have a property to tell you if it's an absolute Uri (IsAbsoluteUri) and a method to make a relative Uri (MakeRelativeUri). Check the Uri FAQ for further information.

---

Example for traversing a path from top to bottom:

var dir = new DirectoryInfo(Environment.CurrentDirectory);
do
{
    Console.WriteLine(dir.Name);

    dir = dir.Parent;
} while (dir != null);

Answer 3

What is a file path anymore than a string? You can use the Path class to manipulate the strings. The DirectoryInfo class also has a Parent to get it's parent directory, and FileInfo has a Directory property to get it's current directory. What else do you need?

Answer 4

I think the typesafe objects for FS entities are FileInfo and DirectoryInfo themselves.