SkyJourney
64445322eb
新增能力：将 DOCX 发布验收拆分为四套独立 140，支持真实语料冻结、指纹复用、失败与基础设施错误独立统计，并为表格换行、全 JSON 围栏、代码连续性和长文档分页建立通用门禁。 问题修复：冻结 Paged.js 分片前的逻辑表格列轨并传递打印几何，统一 Markdown 表格换行、代码、段落与 OOXML 翻译；改进 PDF 文本流排序、语义块映射、颜色与栅格比较，消除窄字符重叠和跨行范围符号误报。 兼容与部署：版本统一为 0.6.2；正式 Docker 镜像内置固定 Chromium、Pandoc 3.9.0.2 和 Serif/Sans/Mono 字体；Desktop NSIS 与 ZIP 继续直接内置字体，无需系统字体安装。 验证结果：合成基线与长庆严格 280/280，M4N 140/140；健康数据残余误报 6/115（5.22%），均核查为重复表头自动对齐/取样误报且基础设施错误为 0。全项目测试、类型检查、生产构建和 git diff --check 通过；正式 Docker、NSIS、ZIP、离线镜像、部署包、清单及 SHA-256 均已生成并校验。
101 lines
3.1 KiB
JavaScript
101 lines
3.1 KiB
JavaScript
import fs from "node:fs";
|
||
import path from "node:path";
|
||
|
||
export const DOCX_REAL_WORLD_CORPUS = Object.freeze([
|
||
{
|
||
id: "changqing-tender-requirements",
|
||
label: "长庆油田智能健康技术服务招标技术要求",
|
||
markdownPath: "tmp/长庆油田智能健康技术服务招标技术要求.md",
|
||
requiredFeatures: ["inline-code", "table-math-alignment"]
|
||
},
|
||
{
|
||
id: "health-data-schema",
|
||
label: "运动健康类和营养饮食数据结构升级",
|
||
markdownPath: "tmp/运动健康类和营养饮食数据数据结构升级.md",
|
||
requiredFeatures: ["json-code-indentation"],
|
||
codeIndentationProbe: {
|
||
rootLine: "{",
|
||
nestedLineIncludes: '"heart_rate_daily"',
|
||
minimumIndentPt: 4
|
||
}
|
||
},
|
||
{
|
||
id: "m4n-table-design",
|
||
label: "M4N 全量表设计模块化讲解",
|
||
markdownPath: "tmp/数据中台项目周报_2026_W30_附件_M4N全量表设计模块化讲解.md",
|
||
requiredFeatures: ["mermaid", "long-document"]
|
||
}
|
||
]);
|
||
|
||
export function resolveDocxRealWorldCorpus(repositoryDirectory) {
|
||
return DOCX_REAL_WORLD_CORPUS.map((entry) => ({
|
||
...entry,
|
||
absoluteMarkdownPath: path.resolve(
|
||
repositoryDirectory,
|
||
entry.markdownPath
|
||
),
|
||
byteLength: fs.existsSync(path.resolve(repositoryDirectory, entry.markdownPath))
|
||
? fs.statSync(path.resolve(repositoryDirectory, entry.markdownPath)).size
|
||
: 0
|
||
}));
|
||
}
|
||
|
||
export function sortDocxRealWorldCorpusBySize(repositoryDirectory) {
|
||
return resolveDocxRealWorldCorpus(repositoryDirectory).sort(
|
||
(first, second) =>
|
||
first.byteLength - second.byteLength ||
|
||
first.id.localeCompare(second.id, "en")
|
||
);
|
||
}
|
||
|
||
export function validateDocxRealWorldCorpus(repositoryDirectory) {
|
||
const corpus = sortDocxRealWorldCorpusBySize(repositoryDirectory);
|
||
const failures = [];
|
||
for (const entry of corpus) {
|
||
if (!fs.existsSync(entry.absoluteMarkdownPath)) {
|
||
failures.push(`${entry.id} 缺少 Markdown：${entry.markdownPath}`);
|
||
continue;
|
||
}
|
||
const source = fs.readFileSync(entry.absoluteMarkdownPath, "utf8");
|
||
if (!source.trim()) {
|
||
failures.push(`${entry.id} Markdown 为空`);
|
||
}
|
||
if (
|
||
entry.requiredFeatures.includes("json-code-indentation") &&
|
||
!/```json\s*[\s\S]*?\n[\t ]+\S/gu.test(source)
|
||
) {
|
||
failures.push(`${entry.id} 缺少带缩进的 JSON 围栏探针`);
|
||
}
|
||
if (
|
||
entry.requiredFeatures.includes("inline-code") &&
|
||
!/(^|[^`])`[^`\n]+`([^`]|$)/mu.test(source)
|
||
) {
|
||
failures.push(`${entry.id} 缺少行内代码探针`);
|
||
}
|
||
if (
|
||
entry.requiredFeatures.includes("table-math-alignment") &&
|
||
!/^\|.*\$[^$]+\$.*\|/mu.test(source)
|
||
) {
|
||
failures.push(`${entry.id} 缺少表格公式对齐探针`);
|
||
}
|
||
}
|
||
return { corpus, failures };
|
||
}
|
||
|
||
export function summarizeDocxReleaseGateMatrix({
|
||
baselineCaseCount,
|
||
corpusCaseCounts
|
||
}) {
|
||
const corpusCaseCount = Object.values(corpusCaseCounts).reduce(
|
||
(sum, value) => sum + value,
|
||
0
|
||
);
|
||
return {
|
||
baselineCaseCount,
|
||
corpusDocumentCount: Object.keys(corpusCaseCounts).length,
|
||
corpusCaseCounts,
|
||
corpusCaseCount,
|
||
totalCaseCount: baselineCaseCount + corpusCaseCount
|
||
};
|
||
}
|