逐级实施回退

Question

我有一个 class ScoreStrategy 描述了如何计算测验分数：

public class ScoreStrategy
{
    public int Id { get; set; }

    public int QuizId { get; set; }

    [Required]
    public Quiz Quiz { get; set; }

    public decimal Correct { get; set; }

    public decimal Incorrect { get; set; }

    public decimal Unattempted { get; set; }
}

三个属性 Correct、Incorrect 和 Unattempted 描述了为响应分配多少点。这些点也可以是负面的。评分策略适用于测验中的所有问题，因此每个测验只能有一个ScoreStrategy。我有两个子classes:

public class DifficultyScoreStrategy : ScoreStrategy
{  
    public QuestionDifficulty Difficulty { get; set; }
}

public class QuestionScoreStrategy : ScoreStrategy
{ 
     [Required]
     public Question Question { get; set; }
}

我的问题有三个难度级别（Easy、Medium、Hard；QuestionDifficulty 是一个枚举）。 DifficultyScoreStrategy 指定是否需要为特定难度的问题分配不同的分数。这将覆盖适用于整个测验的基础 ScoreStrategy。每个难度级别可以有一个实例。

第三，我有一个 QuestionScoreStrategy class 指定特定问题的分数是否必须以不同方式授予。这会覆盖测验范围 ScoreStrategy 和难度范围 DifficultyStrategy。每个问题可以有一个实例。

在评估测验的回答时，我想实施逐级回退机制:

每个问题：

检查问题是否有 QuestionScoreStrategy，如果找到则 return 策略。
如果没有，回退到DifficultyScoreStrategy并检查是否有针对正在评估的问题的难度级别的策略并且 return 如果找到了一个策略，它就会出现。
如果不存在，则回退到测验范围 ScoreStrategy 并检查是否存在，return 如果存在，
如果也没有 ScoreStrategy，则使用默认值 { Correct = 1, Incorrect = 0, Unattempted = 0 }（如果我也可以将其设置为可配置的就好了，就像 .NET 的优雅方式一样：

options => {
    options.UseFallbackStrategy(
        correct: 1, 
        incorrect: 0, 
        unattempted: 0
    );
}

).

总结

^{我已经在 table 中总结了以上信息：}

Strategy Type	Priority	Maximum instances per quiz
`QuestionScoreStrategy`	1st (highest)	As many as there are questions in the quiz
`DifficultyScoreStrategy`	2nd	4, one for each difficulty level
`ScoreStrategy`	3rd	Only one
Fallback strategy (Default `{ Correct = 1, Incorrect = 0, Unattempted = 0}`)	4th (lowest)	Configured for the entire app. Shared by all quizzes

我有一个名为 EvaluationStrategy 的容器 class，其中包含这些评分策略以及其他评估信息：

partial class EvaluationStrategy
{
    public int Id { get; set; }

    public int QuizId { get; set; }

    public decimal MaxScore { get; set; }

    public decimal PassingScore { get; get; }

    public IEnumerable<ScoreStrategy> ScoreStrategies { get; set; }
}

我尝试过的：

我在上面的 EvaluationStrategy class 中添加了一个名为 GetStrategyByQuestion() 的方法（注意它被声明为 partial），它实现了这个回退行为，也是一个依次调用此方法的伴随索引器。我已经声明了两个 HashSet 类型 DifficultyScoreStrategy 和 QuestionScoreStrategy 并且 Initialize() 方法实例化了它们。然后将所有的评分策略按类型切换并添加到相应的HashSet中，每题只能有一个ScoreStrategy，存储在defaultStrategy:

partial class EvaluationStrategy
{
    private ScoreStrategy FallbackStrategy = new() { Correct = 1, Incorrect = 0, Unattempted = 0 }; 
    private ScoreStrategy defaultStrategy;
    HashSet<DifficultyScoreStrategy> dStrategies;
    HashSet<QuestionScoreStrategy> qStrategies;


    public void Initialize()
    {
        qStrategies = new();
        dStrategies = new();
        // Group strategies by type
        foreach (var strategy in strategies)
        {
            switch (strategy)
            {
                case QuestionScoreStrategy qs: qStrategies.Add(qs); break;
                case DifficultyScoreStrategy ds: dStrategies.Add(ds); break;
                case ScoreStrategy s: defaultStrategy = s; break;
            }
        }
    }

    public ScoreStrategy this[Question question] => GetStrategyByQuestion(question);
    

    public ScoreStrategy GetStrategyByQuestion(Question question)
    {
        if (qStrategies is null || dStrategies is null)
        {
            Initialize();
        }
        // Check if question strategy exists
        if (qStrategies.FirstOrDefault(str => str.Question.Id == question.Id) is not null and var qs)
        {
            return qs;
        }
        // Check if difficulty strategy exists
        if (dStrategies.FirstOrDefault(str => str.Question.Difficulty == question.Difficulty) is not null and var ds)
        {
            return ds;
        }
        // Check if default strategy exists
        if (defaultStrategy is not null)
        {
            return defaultStrategy;
        }
        // Fallback
        return FallbackStrategy;
    }
}

这个方法似乎有点笨拙，我觉得不太合适。使用部分 class 并添加到 EvalutationStrategy 似乎也不正确。如何实现这种逐级回退行为？有没有设计pattern/principle我可以在这里使用？如果未配置，我知道 .NET Framework 中的很多事情都会回退到默认约定。我需要类似的东西。或者有人可以简单地推荐一个更清洁、更优雅、可维护性更好的解决方案吗？

NOTE/ADDITIONAL 信息： 所有测验的 ScoreStrategy 和 EvaluationStrategy 都存储在由带有 TPH 映射的 EF Core(.NET 5)：

modelBuilder.Entity<ScoreStrategy>()
                .ToTable("ScoreStrategy")
                .HasDiscriminator<int>("StrategyType")
                .HasValue<ScoreStrategy>(0)
                .HasValue<DifficultyScoreStrategy>(1)
                .HasValue<QuestionScoreStrategy>(2)
                ;
modelBuilder.Entity<EvaluationStrategy>().ToTable("EvaluationStrategy");

我有一个碱基 DbSet<ScoreStrategy> ScoreStrategies 和另一个 DbSet<EvaluationStrategy> EvaluationStrategies。由于 EvaluationStrategy 是一个 EF Core class，我对向其添加逻辑 (GetStrategyByQuestion()) 也有点怀疑。

Answer 1

您可以根据优先级对 ScoringMethods 的顺序进行排序。

首先你按 str is QuestionScoreStrategy 和 str.Question.Id == question.Id 排序。

然后你按str is DifficultyScoreStrategy和str.Question.Difficulty == question.Difficulty排序。

（请注意，由于 false 出现在 true 之前，您必须反转条件）

那你就可以FirstOrDefault() ?? defaultStrategy.

示例：

var defaultStrategy = new() { Correct = 1, Incorrect = 0, Unattempted = 0 };

var selectedStrategy = Strategies.OrderBy(str => 
    !(str is QuestionScoreStrategy questionStrat && questionStrat.Question.Id == question.Id)
).ThenBy(str =>
    !(str is DifficultyScoreStrategy difficultySrat && difficultySrat.Difficulty == question.Difficulty)
).FirstOrDefault() ?? defaultStrategy;

您可以通过添加更多 ThenBy 子句轻松地为此添加更多“级别”。

Answer 2

与波莉

有一个名为 Polly which defines a policy called Fallback 的第 3 方库。

使用这种方法，您可以像这样轻松定义后备链：

public ScoreStrategy GetStrategyByQuestionWithPolly(Question question)
{
    Func<ScoreStrategy, bool> notFound = strategy => strategy is null;

    var lastFallback = Policy<ScoreStrategy>
        .HandleResult(notFound)
        .Fallback(FallbackStrategy);

    var defaultFallback = Policy<ScoreStrategy>
        .HandleResult(notFound)
        .Fallback(defaultStrategy);

    var difficultyFallback = Policy<ScoreStrategy>
        .HandleResult(notFound)
        .Fallback(() => GetApplicableDifficultyScoreStrategy(question));

    var fallbackChain = Policy.Wrap(lastFallback, defaultFallback, difficultyFallback);
    fallbackChain.Execute(() => GetApplicableQuestionScoreStrategy(question));
}

我已经提取了 QuestionScoreStrategy 和 DifficultyScoreStrategy 的策略选择逻辑，如下所示：

private ScoreStrategy GetApplicableQuestionScoreStrategy(Question question)
    => qStrategies.FirstOrDefault(str => str.Question.Id == question.Id);

private ScoreStrategy GetApplicableDifficultyScoreStrategy(Question question)
    => dStrategies.FirstOrDefault(str => str.Difficulty == question.Difficulty);

优点

有一个 return 语句
策略声明与链接分开
每个回退都可以由不同的条件触发
主要选择逻辑与备用逻辑分离

缺点

代码真的很重复
您不能使用流利的 API
您需要使用第 3 方库

没有波莉

如果您不想使用第 3 方库来定义和使用回退链，您可以这样做：

public ScoreStrategy GetStrategyBasedOnQuestion(Question question)
{
    var fallbackChain = new List<Func<ScoreStrategy>>
    {
        () => GetApplicableQuestionScoreStrategy(question),
        () => GetApplicableDifficultyScoreStrategy(question),
        () => defaultStrategy,
        () => FallbackStrategy
    };

    ScoreStrategy selectedStrategy = null;
    foreach (var strategySelector in fallbackChain)
    {
        selectedStrategy = strategySelector();
        if (selectedStrategy is not null)
            break;
    }

    return selectedStrategy;
}

优点

有一个 return 语句
回退链的声明和求值是分开的
简洁明了

缺点

不太灵活：每个回退选择都由相同的条件触发
主要选择不与后备分开

Answer 3

我想象所有数据（问题、策略、测验都存储在数据库中）。然后我会期待这样的方式来获得每个策略：

题攻略

var questionStrategy = dbContext.ScoreStrategies.SingleOrDefault(ss => ss.QuesionId == question.Id);

难度攻略：

var difficultyStrategy = dbContext.ScoreStrategies.SingleOrDefault(ss => ss.Difficulty == question.Difficulty);

测验的默认策略：

var quizStrategy = dbContext.ScoreStrategies.SingleOrDefault(ss => ss.QuizId == question.QuizId)

基于此以及您已经提供的内容，策略只有三个数字：正确答案得分，错误和未尝试答案得分。

因此，这成为抽象 class 的完美候选者，它将为三个实体的基础 class 服务 - 三种类型的策略 - 这将是三个表，因为每个表都有不同的关系：

public abstract class ScoreStrategy
{
    public double Correct { get; set; }
    public double Incorrect { get; set; }
    public double Unattempted { get; set; }
}
// Table with FK relation to Questions table
public class QuestionScoreStrategy : ScoreStrategy
{
    public Question { get; set; }
    public int QuestionId { get; set; }
}
// If you have table with difficulties, there should be FK relation to it.
// If you do not have table - it's worth consideration, you could then 
// easily add more difficulties.
public class DifficultyStrategy : ScoreStrategy
{
    public QuestionDifficulty Difficulty { get; set; }
}
// FK relation to Quizes table
public class QuizScoreStrategy : ScoreStrategy
{
    public Quiz { get; set; }
    public int QuizId { get; set; }
}

通过这种方式，您最终会得到仅存储相关数据的粒度良好的表。

那么，用法将变为：

// Ideally, this method should be in some repoistory (look at repository design pattern) in data access layer
// and should leverage usage of async / await as well.
public ScoreStrategy GetScoreStrategy(Question question)
{
    return dbContext.QuestionScoreStrategies.SingleOrDefault(qs => qs.QuestionId == question.Id)
        ?? dbContext.DifficultyStrategies.SingleOrDefault(ds => ds.Difficulty == question.Difficulty)
        ?? dbContext.QuizScoreStrategies.SingleOrDefault(qs => qs.QuizId == question.QuizId);
}

那么你可以这样使用这个方法：

// This should be outside data access layer. Here you perform logic of getting question.
// This could be some ScoringManager class which should be singleton (one instance only).
// Then you could define fallback in private fields:
private readonly double FALLBACK_CORRECT_SCORE;
private readonly double FALLBACK_INCORRECT_SCORE;
private readonly double FALLBACK_UNATTEMPTED_SCORE;
// private constructor, as this should be singleton
private ScoringManager(double correctScore, double incorrectScore, double unattemptedScore)
    => (FALLBACK_CORRECT_SCORE, FALLBACK_INCORRECT_SCORE, FALLBACK_UNATTEMPTED_SCORE) =
       (correctScore, incorrectScore, unattemptedScore);
 
public double CalcScoreForQuestion(Question question)
{
    var scoreStrategy = GetScoreStrategy(question);
    if (question answered correctly) 
        return scoreStrategy?.Correct ?? FALLBACK_CORRECT_SCORE;
    if (question answered incorrectly) 
        return scoreStrategy?.Incorrect ?? FALLBACK_INCORRECT_SCORE;
    if (question unattempted) 
        return scoreStrategy?.Unattempted ?? FALLBACK_UNATTEMPTED_SCORE;
}

注意

这只是我组织事物的草稿，很可能在编写代码时我会提出改进，但我认为这是前进的方向。例如 ScoringManager 可以有 ConfigureFallbackScore 方法，这将允许动态更改回退分数（这需要使各个字段不是 readonly）。

更新

定义回退策略，以便定义枚举：

public enum FallbackLevel
{
    Difficulty,
    Question,
    Quiz,
}

然后评分管理器可以有配置策略的方法（连同支持字段）：

private FallbackLevel _highPrecedence;
private FallbackLevel _mediumPrecedence;
private FallbackLevel _lowPrecedence;

public void ConfigureFallbackStrategy(FallbackLevel highPrecedence, FallbackLevel mediumPrecedence, FallbackLevel lowPrecedence)
{
    _highPrecedence = highPrecedence;
    _mediumPrecedence = mediumPrecedence;
    _lowPrecedence = lowPrecedence;
}

然后在manager中写获取策略逻辑：

public ScoreStrategy GetScoreStrategy(Question question)
{
   var scoreStrategy = GetScoreStrategy(_highPrecedence, question)
       ?? GetScoreStrategy(_mediumPrecedence, question)
       ?? GetScoreStrategy(_lowPrecedence, question);
}

private ScoreStrategy GetScoreStrategy(FallbackLevel lvl, Question question) => lvl switch
{
    FallbackLevel.Difficulty => dbContext.DifficultyStrategies.SingleOrDefault(ds => ds.Difficulty == question.Difficulty),
    FallbackLevel.Question => dbContext.QuestionScoreStrategies.SingleOrDefault(qs => qs.QuestionId == question.Id),
    FallbackLevel.Quiz => dbContext.QuizScoreStrategies.SingleOrDefault(qs => qs.QuizId == question.QuizId),
}

这样可以非常容易地以任何方式配置回退策略。当然，还有一些注意事项：

确保所有回退策略都是唯一的，例如高、中、低策略不可能相同，
数据库上下文只能通过存储库模式访问
添加更多完整性检查（如空值等）

我省略了那些部分，因为我专注于纯粹的功能。

逐级实施回退

Implementing level by level fallback

.net

c#

oop

fallback

design-patterns

总结

我尝试过的：

与波莉

优点

缺点

没有波莉

优点

缺点

更新