@liuhange/dsh-data-quality-scoring 2.0.0

package/README.md

@@ -0,0 +1,34 @@
+# @liuhange/dsh-data-quality-scoring
+
+数据质量评分插件：对数据完整性、准确性、一致性、时效性多维度评分，生成质量报告和改进建议。
+
+## Tool 接口
+
+### `score_data_quality`
+
+对数据文件进行多维度质量评分，生成 JSON + Markdown 双格式报告。
+
+**参数：**
+- `filePath` (string, 必填): 待评分数据文件路径
+- `outputPathDir` (string, 可选): 报告输出目录，默认数据文件所在目录
+
+**输出：**
+- `quality_report.json`: 质量评分 JSON 报告
+- `quality_report.md`: 质量评分 Markdown 报告（首行 `【数据质量评分报告】`）
+
+## 评分维度
+
+| 维度 | 说明 | 计算方式 |
+|------|------|---------|
+| 完整性 | 缺失值比例 | (非缺失值总数 / 字段值总数) × 100 |
+| 准确性 | 格式合规率 + 值域合规率 | 加权平均 × 100 |
+| 一致性 | 跨字段约束满足率 | (满足约束记录数 / 受约束记录数) × 100 |
+| 时效性 | 数据新鲜度 | (1 - 过期比例) × 100 |
+
+## 配置节点
+
+`business-rules.json` → `qualityScoring` 节点。缺失时使用默认配置（权重各 0.25，阈值 60）。
+
+## v1.0 兼容性
+
+本插件为 v2.0 新增，不影响 v1.0 功能。配置缺失时降级到默认评分规则。

package/lib/types/accuracyScorer.d.ts

@@ -0,0 +1,5 @@
+import type { ScoringContext, DimensionScoringResult } from './types.js';
+export declare class AccuracyScorer {
+    score(context: ScoringContext): DimensionScoringResult;
+}
+//# sourceMappingURL=accuracyScorer.d.ts.map

package/lib/types/completenessScorer.d.ts

@@ -0,0 +1,6 @@
+import type { ScoringContext, DimensionScoringResult } from './types.js';
+export declare class CompletenessScorer {
+    score(context: ScoringContext): DimensionScoringResult;
+    private isMissing;
+}
+//# sourceMappingURL=completenessScorer.d.ts.map

package/lib/types/consistencyScorer.d.ts

@@ -0,0 +1,6 @@
+import type { ScoringContext, DimensionScoringResult } from './types.js';
+export declare class ConsistencyScorer {
+    score(context: ScoringContext): DimensionScoringResult;
+    private checkConstraint;
+}
+//# sourceMappingURL=consistencyScorer.d.ts.map

package/lib/types/defaultScoringRules.d.ts

@@ -0,0 +1,3 @@
+import type { QualityScoringConfig } from '@liuhange/dsh-data-asset-shared';
+export declare const defaultScoringRules: QualityScoringConfig;
+//# sourceMappingURL=defaultScoringRules.d.ts.map

package/lib/types/index.d.ts

@@ -0,0 +1,5 @@
+import type { Context } from '@deepseek-ai/cordis';
+export declare const name = "data-quality-scoring";
+export declare const inject: string[];
+export declare function apply(ctx: Context): void;
+//# sourceMappingURL=index.d.ts.map

package/lib/types/invariant.d.ts

@@ -0,0 +1,3 @@
+export declare const invariant = "data-quality-scoring";
+export declare function install(): void;
+//# sourceMappingURL=invariant.d.ts.map

package/lib/types/issueCollector.d.ts

@@ -0,0 +1,5 @@
+import type { DimensionScoringResult, QualityIssue } from './types.js';
+export declare class IssueCollector {
+    collect(dimensionResults: DimensionScoringResult[]): QualityIssue[];
+}
+//# sourceMappingURL=issueCollector.d.ts.map

package/lib/types/qualityReportGenerator.d.ts

@@ -0,0 +1,6 @@
+import type { QualityScoringResult } from './types.js';
+export declare class QualityReportGenerator {
+    writeJson(filePath: string, result: QualityScoringResult): Promise<void>;
+    writeMarkdown(filePath: string, result: QualityScoringResult): Promise<void>;
+}
+//# sourceMappingURL=qualityReportGenerator.d.ts.map

package/lib/types/suggestionGenerator.d.ts

@@ -0,0 +1,6 @@
+import type { QualityDimensions } from './types.js';
+import type { QualityScoringConfig } from '@liuhange/dsh-data-asset-shared';
+export declare class SuggestionGenerator {
+    generate(dimensions: QualityDimensions, thresholds: QualityScoringConfig['thresholds']): string[];
+}
+//# sourceMappingURL=suggestionGenerator.d.ts.map

package/lib/types/timelinessScorer.d.ts

@@ -0,0 +1,5 @@
+import type { ScoringContext, DimensionScoringResult } from './types.js';
+export declare class TimelinessScorer {
+    score(context: ScoringContext): DimensionScoringResult;
+}
+//# sourceMappingURL=timelinessScorer.d.ts.map

package/lib/types/types.d.ts

@@ -0,0 +1,54 @@
+export type QualityDimension = 'completeness' | 'accuracy' | 'consistency' | 'timeliness';
+export type IssueSeverity = 'low' | 'medium' | 'high';
+export interface ScoreDataQualityParams {
+    filePath: string;
+    outputPathDir?: string;
+}
+export interface QualityIssue {
+    dimension: QualityDimension;
+    description: string;
+    location: string;
+    severity: IssueSeverity;
+}
+export interface DimensionScore {
+    score: number;
+    weight: number;
+    weightNormalized: number;
+    subScores?: Record<string, number>;
+    issues: QualityIssue[];
+    suggestion?: string;
+}
+export interface QualityDimensions {
+    completeness: DimensionScore;
+    accuracy: DimensionScore;
+    consistency: DimensionScore;
+    timeliness: DimensionScore;
+}
+export interface QualityScoringResult {
+    totalScore: number;
+    dimensions: QualityDimensions;
+    issues: QualityIssue[];
+    suggestions: string[];
+    weightsNormalized: boolean;
+    scoredAt: string;
+}
+export interface ScoreDataQualityResult {
+    jsonReportPath: string;
+    markdownReportPath: string;
+    report: QualityScoringResult;
+    totalScore: number;
+    dimensions: QualityDimensions;
+    status: 'SUCCESS' | 'FAILED';
+    configStatus: string;
+}
+export interface ScoringContext {
+    records: Record<string, unknown>[];
+    fieldNames: string[];
+    config: unknown;
+}
+export interface DimensionScoringResult {
+    score: number;
+    issues: QualityIssue[];
+    subScores?: Record<string, number>;
+}
+//# sourceMappingURL=types.d.ts.map

package/lib/types/weightedScoreCalculator.d.ts

@@ -0,0 +1,10 @@
+import type { QualityDimensions } from './types.js';
+import type { QualityScoringConfig } from '@liuhange/dsh-data-asset-shared';
+export declare class WeightedScoreCalculator {
+    calculate(dimensions: QualityDimensions, weights: QualityScoringConfig['weights']): {
+        totalScore: number;
+        weightsNormalized: boolean;
+        normalizedWeights: QualityScoringConfig['weights'];
+    };
+}
+//# sourceMappingURL=weightedScoreCalculator.d.ts.map

package/package.json

@@ -0,0 +1,39 @@
+{
+    "name":  "@liuhange/dsh-data-quality-scoring",
+    "description":  "Data quality scoring plugin: multi-dimensional scoring (completeness/accuracy/consistency/timeliness) with weighted total and improvement suggestions",
+    "version":  "2.0.0",
+    "publishConfig":  {
+                          "access":  "public"
+                      },
+    "repository":  {
+                       "type":  "git",
+                       "url":  "git+https://github.com/liuhange789/data-asset-inspector.git",
+                       "directory":  "packages/data-asset/data-quality-scoring"
+                   },
+    "type":  "module",
+    "main":  "lib/index.js",
+    "types":  "lib/types/index.d.ts",
+    "exports":  {
+                    ".":  {
+                              "types":  "./lib/types/index.d.ts",
+                              "default":  "./lib/index.js"
+                          },
+                    "./invariant":  {
+                                        "types":  "./lib/types/invariant.d.ts",
+                                        "default":  "./lib/invariant.js"
+                                    },
+                    "./src/*":  "./src/*",
+                    "./package.json":  "./package.json"
+                },
+    "files":  [
+                  "lib/index.js",
+                  "lib/invariant.js",
+                  "lib/types/**/*.d.ts"
+              ],
+    "license":  "MIT",
+    "peerDependencies":  {
+                             "@deepseek-ai/cordis":  "^0.1.0",
+                             "@deepseek-ai/dsh-tools":  "^0.1.0",
+                             "@liuhange/dsh-data-asset-shared":  "^2.0.0"
+                         }
+}