miaoguo_system_server/.vscode/秒过知识库更新/audit_miaoguo_literacy_2026_fall.mjs

#!/usr/bin/env node

import assert from "node:assert/strict";
import fs from "node:fs";
import path from "node:path";
import { fileURLToPath } from "node:url";
import mysql from "mysql2/promise";
import config from "../../src/config/dev.js";

const taskDir = path.dirname(fileURLToPath(import.meta.url));
const sourcePath = path.join(taskDir, "source_chinese_2026_fall.json");
const applyReportPath = path.join(taskDir, "apply_report_2026_fall.json");
const jsonReportPath = path.join(
  taskDir,
  "miaoguo_literacy_dependency_audit_current_2026_fall.json",
);
const markdownReportPath = path.join(taskDir, "MiaoguoLiteracy依赖检查_2026年秋季.md");

const sources = JSON.parse(fs.readFileSync(sourcePath, "utf8")).filter(
  (source) => source.unitType === 2,
);
const applyReport = JSON.parse(fs.readFileSync(applyReportPath, "utf8"));
const backup = JSON.parse(fs.readFileSync(applyReport.backupPath, "utf8"));

function splitWords(example) {
  return example.split(",").filter(Boolean);
}

function inspectJson(row) {
  if (!row) return { status: "NO_ROW", detail: "MiaoguoLiteracy 中没有可用记录" };
  if (!row.JSONString) return { status: "EMPTY_JSON", detail: "JSONString 为空" };
  try {
    const json = JSON.parse(row.JSONString);
    if (!json.CHN) return { status: "MISSING_CHN", detail: "JSONString 中不存在 CHN" };
    if (
      !Array.isArray(json.CHN.PinYin) ||
      !json.CHN.PinYin[0]?.pinyin ||
      !json.CHN.PinYin[0]?.explain
    ) {
      return {
        status: "INVALID_CHN_STRUCTURE",
        detail: "CHN.PinYin[0].pinyin/explain 不完整，现有接口无法使用",
      };
    }
    return { status: "OK", detail: "存在可用 CHN" };
  } catch (error) {
    return { status: "INVALID_JSON", detail: error.message };
  }
}

function writeReports(report) {
  fs.writeFileSync(jsonReportPath, `${JSON.stringify(report, null, 2)}\n`, "utf8");
  const lines = [
    "# MiaoguoLiteracy 缺失清单（2026 年秋季）",
    "",
    `检查时间：${report.generatedAt}`,
    "",
    "检查口径：检查三套新版词语表的全部最终词语；按线上接口过滤掉 `SearchType=shici/eng`，并按 `Word,SearchType DESC` 选择每个词的首条记录。除检查 `CHN` 外，还检查了接口实际使用的 `CHN.PinYin[0].pinyin/explain`。",
    "",
    `- 新版词语总数：${report.summary.finalOccurrences}`,
    `- 不重复词语数：${report.summary.uniqueFinalWords}`,
    `- 相比更新前新出现的不重复词语：${report.summary.uniqueNewlyIntroducedWords}`,
    `- 已有可用 CHN：${report.summary.usableWords}`,
    `- 缺少可用 CHN：${report.summary.problemWords}`,
    "",
    "## 需要新增的词语",
    "",
    "| 词语 | 数据集 | 单元 | 状态 | 相对旧版 |",
    "|---|---|---|---|---|",
  ];
  for (const problem of report.problems) {
    lines.push(
      `| ${problem.word} | ${problem.locations.map((item) => item.label).join("、")} | ${problem.locations.map((item) => item.unit).join("、")} | ${problem.status}: ${problem.detail} | ${problem.isNewlyIntroduced ? "新出现" : "旧版已有"} |`,
    );
  }
  lines.push(
    "",
    "## 补充说明",
    "",
    `另发现 ${report.nonBlockingLowerPriorityIssues.length} 条低优先级 \`SearchType=list\` 记录自身没有 CHN；这些词同时存在排序更靠前且具有完整 CHN 的 \`zici\` 记录，线上接口实际会使用 \`zici\`，所以没有列入待新增清单。`,
    "",
    "本报告只做查询和分析，没有向 `MiaoguoLiteracy` 写入任何数据。",
    "",
  );
  fs.writeFileSync(markdownReportPath, `${lines.join("\n")}\n`, "utf8");
}

async function main() {
  assert.equal(sources.length, 3, "本批次应有三套语文词语表");
  const oldWordsByBook = new Map(
    sources.map((source) => [
      source.bookIdOld,
      new Set(
        backup.hanziRows
          .filter((row) => row.HanziBookID === source.bookIdOld)
          .flatMap((row) => splitWords(row.Example ?? "")),
      ),
    ]),
  );

  const references = new Map();
  for (const source of sources) {
    const oldWords = oldWordsByBook.get(source.bookIdOld);
    for (const unit of source.units) {
      for (const word of splitWords(unit.example)) {
        const items = references.get(word) ?? [];
        items.push({
          label: source.label,
          bookIdOld: source.bookIdOld,
          unit: unit.name,
          isNewlyIntroduced: !oldWords.has(word),
        });
        references.set(word, items);
      }
    }
  }

  const words = [...references.keys()];
  const connection = await mysql.createConnection(config.database);
  try {
    const [rows] = await connection.query(
      `SELECT ID,Word,SearchType,JSONString
         FROM MiaoguoLiteracy
        WHERE Word IN (?)
          AND SearchType<>?
          AND SearchType<>?
        ORDER BY Word,SearchType DESC,ID`,
      [words, "shici", "eng"],
    );

    const rowsByWord = new Map();
    for (const row of rows) {
      const items = rowsByWord.get(row.Word) ?? [];
      items.push(row);
      rowsByWord.set(row.Word, items);
    }

    const problems = [];
    const nonBlockingLowerPriorityIssues = [];
    for (const word of words) {
      const wordRows = rowsByWord.get(word) ?? [];
      const selectedRow = wordRows[0];
      const selectedCheck = inspectJson(selectedRow);
      const locations = references.get(word);
      if (selectedCheck.status !== "OK") {
        problems.push({
          word,
          status: selectedCheck.status,
          detail: selectedCheck.detail,
          selectedRow: selectedRow
            ? { id: selectedRow.ID, searchType: selectedRow.SearchType }
            : null,
          rowCount: wordRows.length,
          isNewlyIntroduced: locations.some((item) => item.isNewlyIntroduced),
          locations,
        });
      }
      for (const row of wordRows.slice(1)) {
        const check = inspectJson(row);
        if (check.status !== "OK") {
          nonBlockingLowerPriorityIssues.push({
            word,
            id: row.ID,
            searchType: row.SearchType,
            status: check.status,
            detail: check.detail,
            selectedRowId: selectedRow?.ID ?? null,
          });
        }
      }
    }

    const uniqueNewlyIntroducedWords = words.filter((word) =>
      references.get(word).some((item) => item.isNewlyIntroduced),
    ).length;
    const report = {
      generatedAt: new Date().toISOString(),
      readOnly: true,
      querySemantics:
        "与 GetMiaoguoLiteracyWords 一致：排除 SearchType=shici/eng，按 Word,SearchType DESC 取词语首条记录",
      summary: {
        finalOccurrences: sources.reduce((total, source) => total + source.wordNum, 0),
        uniqueFinalWords: words.length,
        uniqueNewlyIntroducedWords,
        eligibleDatabaseRows: rows.length,
        usableWords: words.length - problems.length,
        problemWords: problems.length,
      },
      datasets: sources.map((source) => ({
        label: source.label,
        bookIdOld: source.bookIdOld,
        finalWords: source.wordNum,
        oldUniqueWords: oldWordsByBook.get(source.bookIdOld).size,
        newlyIntroducedOccurrences: source.units
          .flatMap((unit) => splitWords(unit.example))
          .filter((word) => !oldWordsByBook.get(source.bookIdOld).has(word)).length,
      })),
      problems,
      nonBlockingLowerPriorityIssues,
    };
    writeReports(report);
    console.log(
      JSON.stringify(
        {
          summary: report.summary,
          missingWords: problems.map((problem) => problem.word),
          jsonReportPath,
          markdownReportPath,
        },
        null,
        2,
      ),
    );
    if (problems.length > 0) process.exitCode = 2;
  } finally {
    await connection.end();
  }
}

main().catch((error) => {
  console.error(error);
  process.exitCode = 1;
});