字符串操作-可以做得更好吗？

Question

Right now we're hand rolling logins and one of the requirements is to have the login page notify the user which module they're going to login to. Now the only thing I have to work with that has the URL they're going to land on is in a query string such as this:

Request.QueryString["ReturnURL"]

With a value like this:

~/moduleFolder/SpecificPage.aspx

Now this is the method that I created that can break that URL up by the forward slash, take that segment seperate the words, capitalize the first word, remove the first slash, and assign it back to the label for display. An example of the code is below:

string incomingName = Request.QueryString["ReturnURL"].ToString();
int first = incomingName.IndexOf(@"/");
int last = incomingName.LastIndexOf(@"/");
string tempName = incomingName.Substring(first, last - first);
string seperatedName = Regex.Replace(tempName, "([a-z])([A-Z])", "$1 $2");
string upperCased = seperatedName.Replace("/", "");
string portalName = char.ToUpper(upperCased[0]) + upperCased.Substring(1);
lblPortalName.Text = portalName;    

Is there a cleaner or better way to write this code with out having so many different instances of a new string?

Answer 1

Yeah, the cleaner way would be like this:

private static string GetMiddleSegment(string URL)
{
    // you should probably use a library function for this kind of thing

    int first = URL.IndexOf(@"/");
    int last = URL.LastIndexOf(@"/");
    return URL.Substring(first + 1, last - first - 1); // this is correct, right?
}

private static string SeparateWords(string camelCase)
{
    return Regex.Replace(camelCase, "([a-z])([A-Z])", "$1 $2");
}

private static string Uppercase(string name)
{
    return char.ToUpper(name[0]) + name.Substring(1);
}

// ...

string incomingURL = Request.QueryString["ReturnURL"].ToString();
string nameSegment = GetMiddleSegment(incomingURL);
string displayName = Uppercase(SeparateWords(nameSegment));
lblPortalName.Text = displayName;

You'll notice that my code doesn't create less string instances. That's because there's absolutely no way that the amount of string instances created here has anything to do with your performance when serving a request.

Answer 2

string url = "~/moduleFolder/SpecificPage.aspx";
string moduleFolder = url.Split('/')[1];
string separatedName = Regex.Replace(moduleFolder, "([a-z])([A-Z])", "$1 $2");
string portalName = CultureInfo.CurrentCulture.TextInfo.ToTitleCase(separatedName);

Answer 3

I wouldn't change it. At best you'll save a few lines of code, but as it is, this code is probably more readable since this allows for descriptive names of the string at each step of manipulation.

Though you might look into library url functions instead of implementing your own.

Answer 4

I would write the code that gets the part between the first and second /s like this:

var tempName = incomingName.Split('/').Last();

This is quite terser and IMO no harder to understand. Plus, it saves you the trouble of doing

string upperCased = seperatedName.Replace("/", "");

so the end result is 3 lines shorter without loss of readability.

Edit:

Simplified the Split call after the OP's comment that there will only be two slashes. The .Last() call can of course be substituted with [1] , although personally I prefer the way .Last() reads. It conveys intent more clearly than a magic number, especially if you are not the author of the code.

Answer 5

Based on RedFilter's regular expression you could created simple extension methods like:

private static string UrlModuleName(this string url)
{
    return Regex.Replace(url.Split('/')[1], "([a-z])([A-Z])",
                         "$1 $2").ToTitleCaseInvariant();
}
private static string ToTitleCaseInvariant(this string input)
{
    return CultureInfo.InvariantCulture.TextInfo.ToTitleCase(input);
}

This is a good idea to get it simple .