I came across a curious piece of code as I was working my way through the project I work day-in-day-out on; not that curious pieces of code are rare in this project.

This one made me stop and think. I knew it was wrong, but how could I make it right? The problem is how to determine a unique filename in a directory, say for example if I wanted to allow users to upload images which would be stored in a folder, I don't want to overwrite existing files and I don't want to use the filename of the uploaded file.

Here's the code I came across.

1int count = 1;
2while (count < 1000)
3{
4    string tmp = Path.Combine(this.ImagesPhysicalFolder, modifiedFileName) + count.ToString() + ".jpg";
5    if (!File.Exists(tmp))
6    {
7        modifiedFileName = tmp;
8        break;
9    }
10    ++count;
11}

It's a little verbose, and calling the Path.Combine() inside the loop could be costly. This is the solution I came up with.

1var root = ...;
2
3/* snip */
4
5var count = 0;
6var template = Path.Combine(root, "file_%%.jpg");
7var filename = template.Replace("%%", (count++).ToString());
8while (File.Exists(filename))
9{
10    filename = template.Replace("%%", (count++).ToString());
11}