@topmindspace/tms-skills 0.1.19 → 0.2.0

package/topmind-x-article/scripts/md2x.py

@@ -0,0 +1,121 @@
+#!/usr/bin/env python3
+"""md → X 长文可粘贴纯文本。
+
+把 Markdown 原稿转成可直接粘贴进 X Article 编辑器的纯文本：
+标题/正文保留，排版标记剥离，图片转为 [图N] 占位。
+
+用法：
+    python3 md2x.py <输入.md> --out <输出/X发布稿.txt>
+
+规则见 references/x-format.md。
+"""
+import argparse
+import re
+import sys
+from pathlib import Path
+
+
+def convert(md: str):
+    images = []
+    out_lines = []
+    in_code = False
+
+    def img_repl(m):
+        alt = m.group(1).strip()
+        images.append(alt)
+        return f"[图{len(images)}]"
+
+    for raw in md.splitlines():
+        line = raw.rstrip()
+
+        # frontmatter 直接跳过
+        if line.strip() == "---":
+            continue
+
+        # 代码块：去围栏，内容缩进保留
+        if line.strip().startswith("```"):
+            in_code = not in_code
+            continue
+        if in_code:
+            out_lines.append("    " + line)
+            continue
+
+        # 图片占位
+        line = re.sub(r"!\[([^\]]*)\]\([^)]*\)", img_repl, line)
+
+        # 标题 → 纯文本行
+        m = re.match(r"^(#{1,6})\s+(.*)$", line)
+        if m:
+            out_lines.append("")
+            out_lines.append(m.group(2).strip())
+            out_lines.append("")
+            continue
+
+        # 引用 → 纯文本
+        line = re.sub(r"^>\s?", "", line)
+
+        # 分割线 → 空行
+        if re.match(r"^(\*{3,}|-{3,}|_{3,})\s*$", line):
+            out_lines.append("")
+            continue
+
+        # 表格行 → "项：值" 列表
+        if "|" in line and line.strip().startswith("|"):
+            cells = [c.strip() for c in line.strip().strip("|").split("|")]
+            if all(re.match(r"^:?-+:?$", c) for c in cells):
+                continue
+            label = cells[0]
+            rest = "，".join(c for c in cells[1:] if c)
+            out_lines.append(f"{label}：{rest}" if rest else label)
+            continue
+
+        # 行内标记剥离
+        line = re.sub(r"\*\*(.+?)\*\*", r"\1", line)      # 加粗
+        line = re.sub(r"__(.+?)__", r"\1", line)
+        line = re.sub(r"\*(.+?)\*", r"\1", line)          # 斜体
+        line = re.sub(r"`(.+?)`", r"\1", line)            # 行内代码
+        line = re.sub(r"~~(.+?)~~", r"\1", line)          # 删除线
+        line = re.sub(r"\[([^\]]+)\]\(([^)]+)\)", r"\1（\2）", line)  # 链接
+        line = re.sub(r"^(\s*)- \[ \]\s+", r"\1- ", line)  # 任务列表
+        line = re.sub(r"^(\s*)- \[x\]\s+", r"\1- ", line, flags=re.I)
+
+        out_lines.append(line)
+
+    # 压缩 3+ 空行为 2 个
+    text = "\n".join(out_lines)
+    text = re.sub(r"\n{3,}", "\n\n", text).strip() + "\n"
+
+    # 文末附图注清单
+    if images:
+        text += "\n—— 配图清单 ——\n"
+        for i, alt in enumerate(images, 1):
+            text += f"[图{i}] {alt or '（未写图注）'}\n"
+
+    return text, images
+
+
+def main() -> None:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("input", help="输入 Markdown")
+    ap.add_argument("--out", required=True, help="输出 X发布稿.txt")
+    args = ap.parse_args()
+
+    inp = Path(args.input)
+    if not inp.is_file():
+        sys.exit(f"错误：找不到输入文件 {inp}")
+    md = inp.read_text(encoding="utf-8")
+    # 去掉首尾 frontmatter 块
+    md = re.sub(r"^---\n.*?\n---\n", "", md, flags=re.S)
+    text, images = convert(md)
+
+    out = Path(args.out)
+    out.parent.mkdir(parents=True, exist_ok=True)
+    out.write_text(text, encoding="utf-8")
+
+    chars = len(text.replace("\n", ""))
+    print(f"OK {out}")
+    print(f"正文约 {chars} 字，配图 {len(images)} 张")
+
+
+if __name__ == "__main__":
+    main()

package/topmind-x-article/scripts/negative_tests.py

@@ -0,0 +1,74 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""topmind-x-article 异常输入测试：md2x.py 在各种坏输入下不崩溃（任一失败 → 退出码 1）。"""
+from __future__ import annotations
+
+import subprocess
+import sys
+import tempfile
+from pathlib import Path
+
+try:
+    sys.stdout.reconfigure(encoding="utf-8", errors="replace")
+except Exception:
+    pass
+
+ROOT = Path(__file__).resolve().parent.parent
+MD2X = ROOT / "scripts" / "md2x.py"
+PY = sys.executable
+fails: list[str] = []
+
+
+def run(args: list[str]) -> subprocess.CompletedProcess:
+    return subprocess.run([PY, str(MD2X), *args], capture_output=True, text=True, timeout=30)
+
+
+def check(name: str, cond: bool, detail: str = "") -> None:
+    print(("  ✓ " if cond else "  ✗ ") + name + (f" — {detail}" if detail and not cond else ""))
+    if not cond:
+        fails.append(name)
+
+
+def main() -> None:
+    print("[negative_tests] topmind-x-article")
+    with tempfile.TemporaryDirectory() as td:
+        td = Path(td)
+
+        # 1. 空文件
+        (td / "empty.md").write_text("", encoding="utf-8")
+        r = run([str(td / "empty.md"), "--out", str(td / "o1.txt")])
+        check("空文件不崩溃", r.returncode == 0 and "Traceback" not in r.stderr)
+
+        # 2. 只有 frontmatter
+        (td / "fm.md").write_text("---\ntitle: t\n---\n", encoding="utf-8")
+        r = run([str(td / "fm.md"), "--out", str(td / "o2.txt")])
+        check("纯 frontmatter 不崩溃", r.returncode == 0 and "Traceback" not in r.stderr)
+
+        # 3. 未闭合代码围栏
+        (td / "fence.md").write_text("# t\n\n```python\nprint(1)\n", encoding="utf-8")
+        r = run([str(td / "fence.md"), "--out", str(td / "o3.txt")])
+        check("未闭合围栏不崩溃", r.returncode == 0 and "Traceback" not in r.stderr)
+
+        # 4. 无分隔行的表格
+        (td / "tbl.md").write_text("# t\n\n| a | b |\n| 1 | 2 |\n", encoding="utf-8")
+        r = run([str(td / "tbl.md"), "--out", str(td / "o4.txt")])
+        check("异常表格不崩溃", r.returncode == 0 and "Traceback" not in r.stderr)
+
+        # 5. 不存在的输入 → 非零退出且无 Traceback
+        r = run([str(td / "nope.md"), "--out", str(td / "o5.txt")])
+        check("缺失输入干净报错", r.returncode != 0 and "Traceback" not in r.stderr)
+
+        # 6. 正常输入产出配图清单
+        (td / "ok.md").write_text("# t\n\n![a](x.png)\n", encoding="utf-8")
+        r = run([str(td / "ok.md"), "--out", str(td / "o6.txt")])
+        out = (td / "o6.txt").read_text(encoding="utf-8") if (td / "o6.txt").exists() else ""
+        check("正常输入含配图清单", r.returncode == 0 and "[图1]" in out)
+
+    if fails:
+        print(f"\n[negative_tests] {len(fails)} 项失败")
+        sys.exit(1)
+    print("\n[negative_tests] 全部通过")
+
+
+if __name__ == "__main__":
+    main()

package/topmind-x-article/scripts/package_skill.py

@@ -0,0 +1,188 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""技能分发包打包器（发布前完整性校验 + 发布清单）。
+
+用法:
+    python scripts/package_skill.py            # 校验 → 打包 → 写发布清单
+    python scripts/package_skill.py --check    # 只校验不打包（CI / 提交前自检）
+
+产物:
+    dist/<skill>.zip            分发包（解压目录名 = 技能名）
+    dist/<skill>.manifest.json  发布清单（版本 / 条目 / 大小 / SHA-256）
+
+校验（任一不过 → 退出码 1）:
+    · SKILL.md frontmatter 含 name / description；name 与目录名一致；
+      description ≤ 1024 字符（平台截断阈值）
+    · README.md / package.json 存在；package.json name 与技能名一致
+    · scripts/*.py 全部可编译；references/*.md 非空
+    · SKILL.md 引用的 references/<file> 全部存在
+    · 无 scripts/_* / __pycache__ / .env 等不应进包的文件
+"""
+from __future__ import annotations
+
+import hashlib
+import json
+import re
+import sys
+import zipfile
+from datetime import datetime, timezone
+from pathlib import Path
+
+try:
+    sys.stdout.reconfigure(encoding="utf-8", errors="replace")
+except Exception:
+    pass
+
+ROOT = Path(__file__).resolve().parent.parent
+NAME = ROOT.name
+OUT = ROOT / "dist" / f"{NAME}.zip"
+MANIFEST = ROOT / "dist" / f"{NAME}.manifest.json"
+DESC_LIMIT = 1024
+
+INCLUDE = ["SKILL.md", "README.md", "package.json", "assets/", "references/", "scripts/", "evals/", "agents/"]
+EXCLUDE_RES = [
+    re.compile(r"(^|/)scripts/_"),
+    re.compile(r"(^|/)\.env($|\.)"),
+    re.compile(r"(^|/)__pycache__(/|$)"),
+    re.compile(r"(^|/)node_modules(/|$)"),
+    re.compile(r"(^|/)dist(/|$)"),
+]
+
+errors: list[str] = []
+
+
+def fail(msg: str) -> None:
+    errors.append(msg)
+    print(f"  ✗ {msg}")
+
+
+def ok(msg: str) -> None:
+    print(f"  ✓ {msg}")
+
+
+def read_frontmatter() -> dict:
+    text = (ROOT / "SKILL.md").read_text(encoding="utf-8")
+    m = re.match(r"^---\r?\n([\s\S]*?)\r?\n---", text)
+    if not m:
+        fail("SKILL.md 缺 frontmatter")
+        return {}
+    fm = m.group(1)
+    name = re.search(r"^name:\s*(\S+)\s*$", fm, re.M)
+    desc = ""
+    dm = re.search(r'^description:\s*"([^"]*)"\s*$', fm, re.M)
+    if dm:
+        desc = dm.group(1).strip()
+    else:
+        dm = re.search(r"^description:\s*>\-\s*\n((?:[ \t]+.*\n?)+)", fm, re.M)
+        if dm:
+            desc = " ".join(line.strip() for line in dm.group(1).splitlines() if line.strip())
+    return {"name": name.group(1) if name else "", "description": desc}
+
+
+def check() -> bool:
+    print(f"[check] {NAME}")
+    fm = read_frontmatter()
+    if fm.get("name") != NAME:
+        fail(f"frontmatter name={fm.get('name')!r} 与目录名 {NAME!r} 不一致")
+    else:
+        ok("name 与目录名一致")
+    desc = fm.get("description", "")
+    if not desc:
+        fail("description 为空")
+    elif len(desc) > DESC_LIMIT:
+        fail(f"description {len(desc)} 字符，超过 {DESC_LIMIT}")
+    else:
+        ok(f"description {len(desc)} 字符")
+
+    for f in ("README.md", "package.json"):
+        if not (ROOT / f).is_file():
+            fail(f"缺 {f}")
+        else:
+            ok(f"有 {f}")
+    try:
+        pkg = json.loads((ROOT / "package.json").read_text(encoding="utf-8"))
+        if pkg.get("name") != NAME:
+            fail(f"package.json name={pkg.get('name')!r} 与技能名不一致")
+        if not re.match(r"^\d+\.\d+\.\d+", str(pkg.get("version", ""))):
+            fail("package.json version 非 semver")
+    except (OSError, json.JSONDecodeError) as e:
+        fail(f"package.json 解析失败: {e}")
+
+    for p in ROOT.rglob("*"):
+        if not p.is_file() or ".git" in p.parts:
+            continue
+        rel = p.relative_to(ROOT).as_posix()
+        for rx in EXCLUDE_RES:
+            if rx.search(rel):
+                fail(f"不应进包的文件: {rel}")
+                break
+    ok("无禁用文件")
+
+    for py in sorted((ROOT / "scripts").glob("*.py")):
+        try:
+            compile(py.read_text(encoding="utf-8"), str(py), "exec")
+        except SyntaxError as e:
+            fail(f"{py.name} 编译失败: {e}")
+    ok("scripts/*.py 编译通过")
+
+    refdir = ROOT / "references"
+    if refdir.is_dir():
+        for md in sorted(refdir.glob("*.md")):
+            if md.stat().st_size == 0:
+                fail(f"references/{md.name} 为空")
+        skill_text = (ROOT / "SKILL.md").read_text(encoding="utf-8")
+        for ref in sorted(set(re.findall(r"references/([\w\-.]+\.md)", skill_text))):
+            if not (refdir / ref).is_file():
+                fail(f"SKILL.md 引用的 references/{ref} 不存在")
+        ok("references 完整性通过")
+
+    if errors:
+        print(f"\n[check] {NAME}: {len(errors)} 项失败")
+        return False
+    print(f"\n[check] {NAME}: 全部通过")
+    return True
+
+
+def package() -> None:
+    if not check():
+        sys.exit(1)
+    (ROOT / "dist").mkdir(exist_ok=True)
+    entries: list[str] = []
+    with zipfile.ZipFile(OUT, "w", zipfile.ZIP_DEFLATED) as z:
+        for p in sorted(ROOT.rglob("*")):
+            if not p.is_file() or ".git" in p.parts:
+                continue
+            rel = p.relative_to(ROOT)
+            rel_posix = rel.as_posix()
+            if any(rx.search(rel_posix) for rx in EXCLUDE_RES):
+                continue
+            if not any(
+                rel_posix == inc.rstrip("/") or rel_posix.startswith(inc)
+                for inc in INCLUDE
+                if inc.endswith("/")
+            ) and rel_posix not in INCLUDE:
+                continue
+            z.write(p, arcname=f"{NAME}/{rel_posix}")
+            entries.append(rel_posix)
+    sha = hashlib.sha256(OUT.read_bytes()).hexdigest()
+    pkg = json.loads((ROOT / "package.json").read_text(encoding="utf-8"))
+    manifest = {
+        "name": NAME,
+        "version": pkg.get("version"),
+        "built_at": datetime.now(timezone.utc).isoformat(),
+        "files": len(entries),
+        "sha256": sha,
+        "entries": entries,
+    }
+    MANIFEST.write_text(json.dumps(manifest, ensure_ascii=False, indent=2), encoding="utf-8")
+    print(f"\n[package] {OUT} ({len(entries)} 文件, sha256={sha[:12]}…)")
+
+
+def main() -> None:
+    if "--check" in sys.argv[1:]:
+        sys.exit(0 if check() else 1)
+    package()
+
+
+if __name__ == "__main__":
+    main()