如何为这种方法编写单元测试

Question

I am writing unit test for this method. I have tried lot of times but still can not write any code for it. Please suggest me how to unit test it. I am using C# , nunit framework and rhino mock.

Thanks in advance.

        public FileUploadJsonResult AjaxUploadProfile(int id, string branchName, string filepath, HttpPostedFileBase file)
    {
        // TODO: Add your business logic here and/or save the file
        string statusCode = "1";
        string profilePicture = string.Empty;
        string fileExtension = System.IO.Path.GetExtension(file.FileName.ToLower());
        string fileName = id + "_" + branchName;
        string fileNameWithOriginalExtension = fileName + fileExtension;
        string fileNameWithJPGExtension = fileName + ".jpg";
        string fileServerPath = this.Server.MapPath("~/LO_ProfilePicture/" + fileNameWithJPGExtension);
        string statusMessage = string.Empty;
        if (string.IsNullOrEmpty(fileExtension) || !Utility.isCorrectExtension(fileExtension))
        {
            statusMessage = "Profile picture should be of JPG, BMP, PNG, GIF or JPEG format.";
            return new FileUploadJsonResult { Data = new { message = string.Format(statusMessage, fileNameWithOriginalExtension), filename = string.Empty, profilepic = profilePicture, statusCode = "0" } };
        }
        if (file.ContentLength > PageConstants.PROFILE_PICTURE_FILE_SIZE)
        {
            statusMessage = "Profile picture size should be less than 2MB";
            return new FileUploadJsonResult { Data = new { message = string.Format(statusMessage, fileNameWithOriginalExtension), filename = string.Empty, profilepic = profilePicture, statusCode = "0" } };
        }
        Utility.SaveThumbnailImage(fileServerPath, file.InputStream, PageConstants.BRANCH_PROFILE_PICTURE_FILE_HEIGTH, PageConstants.BRANCH_PROFILE_PICTURE_FILE_WIDTH);
        profilePicture = PageConstants.IMAGE_PATH + "LO_ProfilePicture/" + fileNameWithJPGExtension;
        // Return JSON            
        return new FileUploadJsonResult { Data = new { message = string.Format("Profile Picture is successfully uploaded.", fileNameWithOriginalExtension), filename = fileNameWithJPGExtension, profilepic = profilePicture, statusCode } };
    }

Answer 1

Make it do just the essential part. Split anything that has nothing to do with the operation you're trying to handle to other classes. Put those behind interfaces, so you can mock these in your unittests. This way you'll notice you don't have to test anything with file i/o in this class. In the class below I split up the function in the essential part, some file i/o and retrieving of settings. Even these settings have nothing to do with the current method you're trying to test. The method just needs verification on, for example, the extension, but it doesn't matter on how it does this.

Tip: try to avoid static utility classes. Give them their own class. Also avoid external components such as network communication or file i/o.

As I don't have a lot of context and it may not compile. But I would go with something like:

class Controller {
    public FileUploadJsonResult AjaxUploadProfile(int id, string branchName, string filepath, HttpPostedFileBase file) {
        string fileName = id + "_" + branchName;
        string fileExtension = _fileIO.GetExtensionForFile(file);

        if (!_extensionManager.IsValidExtension(fileExtension)) {
            return CreateAjaxUploadProfileError("Profile picture should be of JPG, BMP, PNG, GIF or JPEG format.");
        }

        if (file.ContentLength > _settingsManager.GetMaximumFileSize()) {
            return CreateAjaxUploadProfileError("Profile picture size should be less than 2MB");
        }

        string fileNameWithJPGExtension = fileName + ".jpg";
        string fileServerPath = _fileIO.GetServerProfilePicture(Server, fileNameWithJPGExtension);
        string fileClientPath = _fileIO.GetClientProfilePicture(fileNameWithJPGExtension);

        var dimensions = _settingsManager.GetThumbnailDimensions();
        _fileIO.SaveThumbnailImage(fileServerPath, file, dimensions.Item1, dimensions.Item2);

        // Return JSON      
        var data = new {
                message = "Profile Picture is successfully uploaded.", 
                filename = fileClientPath,
                profilepic = profilePicture,
                statusCode = "1"
            };
        return new FileUploadJsonResult { Data = data };
    }

    private static CreateAjaxUploadProfileError(string message) {
        var data = new {
                message = message, 
                filename = string.Empty,
                profilepic = string.Empty,
                statusCode = "0"
            };
        return new FileUploadJsonResult { Data = data };
    }
}

class FileIO : IFileIO {
    public string GetExtensionForFile(HttpPostedFileBase file) {
        return System.IO.Path.GetExtension(filePath.FileName.ToLower());
    }

    public string GetServerProfilePicture(T server, string file) {
        return server.MapPath( "~/LO_ProfilePicture/" + file);
    }

    public void SaveThumbnailImage(string path, HttpPostedFileBase file, int height, int width) {
        Utility.SaveThumbnailImage(path, file.InputStream, height, width); // or even inline
    }

    public string GetClientProfilePicture(string fileName) {
        return _settingsManager.GetClientImagePath() + "LO_ProfilePicture/" + fileNameWithJPGExtension;
    }
}

class ExtensionManager : IExtensionManager {
    public bool IsValidExtension(string extension) {
        return Utility.isCorrectExtension(fileExtension); // or even inline
    }
}

class SettingsManager : ISettingsManager {
    public Tuple<int, int> GetThumbnailDimensions() {
        return Tuple.Create<int, int>(PageConstants.BRANCH_PROFILE_PICTURE_FILE_HEIGTH, PageConstants.BRANCH_PROFILE_PICTURE_FILE_WIDTH);
    }

    public int GetMaximumFileSize() {
        return PageConstants.PROFILE_PICTURE_FILE_SIZE;
    }
}

Answer 2

You can look at this function as a combination of multiple functions doing specific work. One function is getting target file path, another is validating extension, another is validating size, another creates thumbnail, etc.

The goal is to breakdown the complex code into small testable functions (units) which you can test independently. So when you put them together you have better confidence that your big function works as expected.