CODEWITHSUNDEEP
c#path
Pacijent
Pacijent

Reputation: 149

Path GetFileName swap / with \

I have a problem where I am having MySQL DB where I store htmlBody text.

Html body text is text like html. Example:

<img src='https://www.Something.com/Images/SomeImage.jpg'>
<p>This is some image</p>

Now I have winform application where I built custom control which builds up html texts from normal winform elements. Problem comes when I extract elements properties from string loaded from MySQL.

I have string like this: <img src='https://www.Something.com/Images/SomeImage.jpg'>

In my function I detect it is image and I create new custom class for it and inside that class I have 2 elements like this:

public string srcPath { get; set; }
public string srcFileName { get; set; }

To extract path and name from my url (https://www.Something.com/Images/SomeImage.jpg) I use Path but when I do string filename = Path.GetDirectoryName(myUrl); instead of https://www.Something.com/Images I get https:\\www.Something.com\\Images

As you can see on first sight it replaced / with \\ which I could overcome by replacing it again but after https I had 2x // and instead replacing it each with 2x \\ it did it only once so even if I do replace it simple way it wont work since it will create string like https:/www.Something.com/Images.

Solution for this I came with is to create custom function to overcome this which is not problem for me but I guess there must be better way of doing it.

Upvotes: 0

Views: 62

Answers (1)

spender
spender

Reputation: 120440

URIs are not file paths and treating them as such inevitably will lead to problems, as you have discovered. However, the Uri class has a Segments property that can help with this decomposition.

Combined with UriBuilder you can:

var uri = new Uri("http://a/b/c/d");
var ub = new UriBuilder(uri);
ub.Path = string.Concat(uri.Segments.Take(uri.Segments.Length - 1));
Console.WriteLine(ub.Uri.AbsoluteUri);
Console.WriteLine(uri.Segments.Last());

It's worth noting that the last segment of a URI isn't really a filename. The URI as a whole points to a resource.

Upvotes: 1

Related Questions