本文为大家带来 Laugh 的“主程必杀技”系列文章第二篇《Unity Analyzer 自动化执行代码规范 -强制加注释（类注释/方法注释）》，系列其他文章将每周更新，或可访问 Laugh 的 Unity 官方开发者社区主页，持续关注学习。
演示代码
获取链接：https://gitee.com/palang/unity-sharp-code-regulator.git
⦁ 检查继承关系
Unity Analyzer 自动化执行代码规范
强制加注释
注释的重要性，主程们应该够能感同深受，这里不再啰嗦。但有多少项目拥有完整的注释，很多人也都心知肚明。今天举个小的例子，来说明如何给类和方法强制添加 xml 文档注释，个性化的需求可以自己添加。
一、环境准备
4. Microsoft.CodeAnalysis.CSharp，安装 3.8.0 版本 基本过程可以参考上一篇文章。
二、实现目标
1. 强制所有类都加 xml 文档注释
2. 强制所有 public 方法都加 xml 文档注释
/// <summary>/// 这是一个测试类，主要用于强制注释的测试[/// </summary>public class TestClass](/// </summary>public class TestClass){/// <summary>/// 这是一个测试方法，用户测试方法注释/// </summary>public void Test(){}}
三、编写 Analyzer 分析器
首先定义错误 ID 在类 DianogsticIDs 中添加成员
public const string FORCE_Comments_ID = "FERR1002";using System;using System.Collections.Generic;using System.Text;namespace Analyzer{public static class DianogsticIDs{public const string FORCE_BRACE_ID = "FERR1001";public const string FORCE_Comments_ID = "FERR1002";}}
然后在 AnalyzerReleases.Unshipped.md 文件添加该规则，如下：
; Unshipped analyzer release; https://github.com/dotnet/roslyn-analyzers/blob/master/src/Microsoft.CodeAnalysis.Analyzers/ReleaseTrackingAnalyzers.Help.md### New RulesRule ID | Category | Severity | Notes--------|----------|----------|-------FERR1001 | Criterion | Error | ForceStatementBraceFERR1002 | Criterion | Error | ForceClassComments
1. 类注释
using Analyzer;using Microsoft.CodeAnalysis;using Microsoft.CodeAnalysis.CSharp;using Microsoft.CodeAnalysis.CSharp.Syntax;using Microsoft.CodeAnalysis.Diagnostics;using System;using System.Collections.Generic;using System.Collections.Immutable;using System.Linq;using System.Text;using System.Text.RegularExpressions;/*** Author: Laugh（笑微）* https://developer.unity.cn/projects/65937455edbc2a001cbd8102*/namespace CdeAnalyzer{[]internal class ForceClassComments : DiagnosticAnalyzer{/// <summary>/// 错误描述/// </summary>private static readonly DiagnosticDescriptor ForceCommentsDescriptor =new DiagnosticDescriptor(DianogsticIDs.FORCE_Comments_ID, // ID"类必须添加XML文档注释", // Title"类必须添加XML文档注释", // Message formatDiagnosticCategories.Criterion, // CategoryDiagnosticSeverity.Error, // SeverityisEnabledByDefault: true // Enabled by default);public override ImmutableArray<DiagnosticDescriptor> SupportedDiagnostics => ImmutableArray.Create(ForceCommentsDescriptor);/// <summary>/// 初始化分析方法/// </summary>/// <param name="context"></param>public override void Initialize(AnalysisContext context){context.RegisterSyntaxTreeAction(AnalyzeSymbol);}/// <summary>/// 检测方法/// </summary>/// <param name="context"></param>private static void AnalyzeSymbol(SyntaxTreeAnalysisContext context){//找到文档的语法根树var root = context.Tree.GetRoot(context.CancellationToken);//找到所有类定义var classNodeList = root.DescendantNodes()?.OfType<ClassDeclarationSyntax>();foreach(var cls in classNodeList){var classDeclaration = (ClassDeclarationSyntax)cls;var tokens = cls.ChildTokens();foreach (var token in tokens){//要列举多种关键字：public, internal, protected, private, no keywordif(token.IsKind(SyntaxKind.PublicKeyword)|| token.IsKind(SyntaxKind.InternalKeyword)|| token.IsKind(SyntaxKind.ClassKeyword)|| token.IsKind(SyntaxKind.ProtectedKeyword)|| token.IsKind(SyntaxKind.PrivateKeyword)){//获取注释文本var comments = token.GetAllTrivia().ToSyntaxTriviaList().ToString();//获取有效注释行List<SyntaxTrivia> liables = new List<SyntaxTrivia>();var triviaList = token.GetAllTrivia();foreach (var trivia in triviaList){if(trivia.IsKind(SyntaxKind.SingleLineCommentTrivia)){if(!trivia.ToString().Contains("<summary>") && !trivia.ToString().Contains("</summary>")){liables.Add(trivia);}}}var liablesStr = string.Join("",liables);liablesStr = liablesStr.Replace("///", "").Trim();if(liablesStr.Length < 4)/*注释长度过短，可以自己添加其他规则*/{//报错var diagnostic = Diagnostic.Create(ForceCommentsDescriptor, cls.GetFirstToken().GetLocation());context.ReportDiagnostic(diagnostic);}}//只判断第一个符合条件的break;}}}}}
2. 方法注释
新建分析器类 ForceFunctionComments.cs，内容如下（具体看代码中的注释）
using Microsoft.CodeAnalysis.Diagnostics;using Microsoft.CodeAnalysis;using System;using System.Collections.Generic;using System.Text;using System.Collections.Immutable;using Analyzer;using Microsoft.CodeAnalysis.CSharp.Syntax;using Microsoft.CodeAnalysis.CSharp;using System.Linq;namespace CdeAnalyzer{[]internal class FroceFunctionComments : DiagnosticAnalyzer{public override ImmutableArray<DiagnosticDescriptor> SupportedDiagnostics => ImmutableArray.Create(ForceCommentsDescriptor);/// <summary>/// 错误描述/// </summary>private static readonly DiagnosticDescriptor ForceCommentsDescriptor =new DiagnosticDescriptor(DianogsticIDs.FORCE_Comments_ID, // ID"类必须添加XML文档注释", // Title"类必须添加XML文档注释", // Message formatDiagnosticCategories.Criterion, // CategoryDiagnosticSeverity.Error, // SeverityisEnabledByDefault: true // Enabled by default);/// <summary>/// 初始化分析方法/// </summary>/// <param name="context"></param>public override void Initialize(AnalysisContext context){context.RegisterSyntaxTreeAction(AnalyzeSymbol);}/// <summary>/// 检测方法/// </summary>/// <param name="context"></param>private static void AnalyzeSymbol(SyntaxTreeAnalysisContext context){//找到文档的语法根树var root = context.Tree.GetRoot(context.CancellationToken);//找到所有类定义var methodNodeList = root.DescendantNodes()?.OfType<MethodDeclarationSyntax>();foreach (var meth in methodNodeList){var classDeclaration = (MethodDeclarationSyntax)meth;var tokens = meth.ChildTokens();foreach (var token in tokens){//关键字：public,其他关键字可以自行处理：internal, protected, private, no keywordif (token.IsKind(SyntaxKind.PublicKeyword)){//获取注释文本var comments = token.GetAllTrivia().ToSyntaxTriviaList().ToString();//获取有效注释行List<SyntaxTrivia> liables = new List<SyntaxTrivia>();var triviaList = token.GetAllTrivia();foreach (var trivia in triviaList){if (trivia.IsKind(SyntaxKind.SingleLineCommentTrivia)){if (!trivia.ToString().Contains("<summary>") && !trivia.ToString().Contains("</summary>")){liables.Add(trivia);}}}var liablesStr = string.Join("", liables);liablesStr = liablesStr.Replace("///", "").Trim();if (liablesStr.Length < 4)/*注释长度过短，可以自己添加其他规则*/{//报错var diagnostic = Diagnostic.Create(ForceCommentsDescriptor, meth.GetFirstToken().GetLocation());context.ReportDiagnostic(diagnostic);}}//只判断第一个符合条件的break;}}}}}
代码可以直接使用。
