68 lines
4.2 KiB
Python
68 lines
4.2 KiB
Python
"""Reproducible human-facing catalog, Mermaid and coverage views."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import html
|
|
import json
|
|
|
|
|
|
def generated_outputs(kb):
|
|
nodes = kb.catalog["nodes"]
|
|
children = {n["id"]: [c for c in nodes if c["parent_id"] == n["id"]] for n in nodes}
|
|
|
|
def subtree(node):
|
|
return [node] + [n for child in children[node["id"]] for n in subtree(child)]
|
|
|
|
def diagram(selected):
|
|
included = {n["id"] for n in selected}
|
|
name = lambda nid: "root" if nid == "mcb-root" else "n" + nid.split("-")[1]
|
|
lines = ["flowchart LR"]
|
|
for n in selected:
|
|
lines.append(f' {name(n["id"])}["{html.escape(n["label_zh"], quote=True)}"]')
|
|
for n in selected:
|
|
if n["parent_id"] in included:
|
|
lines.append(f' {name(n["parent_id"])} --> {name(n["id"])}')
|
|
return "\n".join(lines)
|
|
|
|
full = diagram(nodes)
|
|
mermaid = ["# MCB 中文分类树:Mermaid", "",
|
|
"本文件由分类 JSON 生成。203 个类别,另加展示根节点;包含一级总览、完整图及 15 个完整分支图。",
|
|
"中文为工作译名,暂译/待核标记不是原始类别名称。编号与英文原名见[完整分类目录](mcb-taxonomy.zh-CN.md),来源见[来源说明](sources.md)。", "",
|
|
"## 一级总览", "", "```mermaid", diagram([n for n in nodes if n["depth"] <= 1]), "```", "",
|
|
"## 完整树", "", "```mermaid", full, "```", ""]
|
|
for branch in children["mcb-root"]:
|
|
mermaid += [f'## {branch["label_zh"]}', "", "```mermaid", diagram(subtree(branch)), "```", ""]
|
|
catalog = ["# MCB 中文完整分类目录", "",
|
|
"203 个分类节点:一级 15、二级 109、三级 79,173 个叶子。项目展示根节点不计入 203。",
|
|
"此完整性仅针对公开论文附录图 7,不代表所有机械零件。原文重复名称按路径区分,原文拼写保留。",
|
|
"[Mermaid 树](mcb-taxonomy-mermaid.zh-CN.md) · [来源与疑点](sources.md)", "",
|
|
"| 本地 ID | 中文工作译名 | 英文原名 | 父节点 | 层级 |", "| --- | --- | --- | --- | ---: |"]
|
|
catalog += [f'| {n["id"]} | {n["label_zh"]} | {n["label_en"]} | {n["parent_id"]} | {n["depth"]} |'
|
|
for n in nodes if n["depth"]]
|
|
coverage = ["# 知识覆盖清单", "",
|
|
"分类目录有 203 类,当前有 6 个项目零件族的工程知识。关联某个宽泛来源节点,不代表该节点下全部变体均已覆盖。",
|
|
"状态 `retrieval_verified` 仅表示固定离线检索和参数用例通过,`geometry_validated=false`。", "",
|
|
"| 零件族 | ID | MCB 关系 | 来源节点 | 状态 |", "| --- | --- | --- | --- | --- |"]
|
|
index = []
|
|
for fid, doc in sorted(kb.families.items()):
|
|
m = doc["meta"]
|
|
coverage.append(f'| [{m["name_zh"]}](families/{fid}.md) | `{fid}` | {m["classification"]["relation"]} | {", ".join(m["classification"]["node_ids"]) or "项目扩充"} | {m["status"]} |')
|
|
index.append({k: m[k] for k in ("doc_id", "name_zh", "name_en", "aliases", "classification", "feature_ids", "knowledge_version", "status")}
|
|
| {"path": doc["path"]})
|
|
coverage += ["", "每族范围、未覆盖变体、参数和精度要求以主文档为准。六个完整示例可从各主文档进入。",
|
|
"", "## 扩展状态", "", "其余目录节点仅提供检索线索,不生成空白 Skill 或空白工程文档。新增族的流程见[维护说明](maintenance.md)。"]
|
|
return {
|
|
"references/classification/mcb-taxonomy.zh-CN.md": "\n".join(catalog) + "\n",
|
|
"references/classification/mcb-taxonomy-mermaid.zh-CN.md": "\n".join(mermaid) + "\n",
|
|
"references/classification/mcb-taxonomy.zh-CN.mmd": full + "\n",
|
|
"references/coverage.md": "\n".join(coverage) + "\n",
|
|
"index.json": json.dumps({"version": "0.1.0", "generated_from": "family frontmatter", "families": index}, ensure_ascii=False, indent=2) + "\n",
|
|
}
|
|
|
|
|
|
def build(kb):
|
|
outputs = generated_outputs(kb)
|
|
for relative, content in outputs.items():
|
|
(kb.root / relative).write_text(content, encoding="utf-8")
|
|
return {"generated": list(outputs)}
|