Question

Hoping someone can shed some light on this issue we are having because I’m at a loss here.

First, a little background:

I rewrote the URL rewriting for our application and implemented it a couple of weeks ago. I did this using Application_BeginRequest() in the global.asax file and everything was fine with our application except for a small oversight I had made. When I’m rewriting the URLs I’m simply checking for the existence of certain keywords in the path that the user requests and then rewriting the path accordingly. Pretty straight forward stuff, not inventing the wheel here. Dry code, really. However, the text I’m checking for is all lowercase while the path may come in with different cases.

For instance:

string sPath = Request.Url.ToString();
sPath = sPath.Replace(Request.Url.Scheme + "://", "")
            .Replace(Request.Url.Host, "");
if (sPath.TrimStart('/').TrimEnd('/').Split('/')[0].Contains("reports") && sPath.TrimStart('/').TrimEnd('/').Split('/').Length > 2) {
            string[] aVariables = sPath.TrimStart('/').TrimEnd('/').Split('/');
            Context.RewritePath("/app/reports/report-logon.aspx?iLanguageID=" + aVariables[1] + "&sEventCode=" + aVariables[2]);
}

…if someone enters the pages as /Reports/, the rule will not match and they will receive a 404 error as a result.

Simple to fix, though, I thought. One only needs to force the requested path string to lowercase so that anything I attempt to match against it will be looking at a lowercase version of the requested path, and match successfully in cases such as the above. So I adjusted the code to read:

    string sPath = Request.Url.ToString();
    sPath = sPath.Replace(Request.Url.Scheme + "://", "")
                .Replace(Request.Url.Host, "");
    sPath = sPath.ToLower(); // <--- New line
    if (sPath.TrimStart('/').TrimEnd('/').Split('/')[0].Contains("reports") && sPath.TrimStart('/').TrimEnd('/').Split('/').Length > 2) {
                string[] aVariables = sPath.TrimStart('/').TrimEnd('/').Split('/');
                Context.RewritePath("/app/reports/report-logon.aspx?iLanguageID=" + aVariables[1] + "&sEventCode=" + aVariables[2]);
    }

With this fix, when I request any URL that matches against the URL rewriting, however, the CPU on the server spikes to 100% and my entire application crashes. I take out .ToLower(), kill the app pool, and the application is perfectly fine again.

Am I missing something here!?!? What gives? Why does such a simple method cause my application to explode? .ToLower() works everywhere else in our application, and although I’m not using it extensively, I am using it quite successfully in other places around the application.

Answer 1

Not sure exactly why ToLower would cause this (only thing I can think of is that it is modifying request.url, which sends asp.net into a frenzy), but there is an easy fix: use an ignorecase comparison rather than converting everything tolower.

Change:

sPath.TrimStart('/').TrimEnd('/').Split('/')[0].Contains("reports")

to:

sPath.TrimStart('/').TrimEnd('/').Split('/')[0].IndexOf("reports",  StringComparison.InvariantCultureIgnoreCase) != -1

and remove your ToLower logic.