ai-dockpack 0.1.0__py3-none-any.whl

ai_dockpack/build_impl.py

@@ -0,0 +1,423 @@
+from __future__ import annotations
+
+import json
+import hashlib
+from dataclasses import dataclass
+from pathlib import Path
+from typing import List, Dict, Any, Optional
+import zipfile
+import re
+from importlib import metadata as importlib_metadata
+from markdown_it import MarkdownIt
+
+
+@dataclass
+class DocFile:
+    relpath: str
+    abspath: Path
+    text: str
+    links: List[str]
+
+
+def _read_text(path: Path) -> str:
+    return path.read_text(encoding="utf-8", errors="ignore")
+
+
+def _sha256_text(text: str) -> str:
+    return hashlib.sha256(text.encode("utf-8", errors="ignore")).hexdigest()
+
+
+def _get_attr(token, key: str) -> Optional[str]:
+    """兼容 markdown-it-py 的 attrs 结构：可能是 dict，也可能是 [(k,v), ...]"""
+    attrs = getattr(token, "attrs", None)
+    if not attrs:
+        return None
+
+    if isinstance(attrs, dict):
+        v = attrs.get(key)
+        return v if isinstance(v, str) and v else None
+
+    # 兼容 list/tuple of pairs
+    try:
+        for item in attrs:
+            if isinstance(item, (list, tuple)) and len(item) >= 2:
+                k, v = item[0], item[1]
+                if k == key and isinstance(v, str) and v:
+                    return v
+    except TypeError:
+        return None
+
+    return None
+
+
+def _extract_links_md(text: str) -> List[str]:
+    """从 Markdown 里抽取链接（MVP：抽出 href，不做判断）"""
+    md = MarkdownIt()
+    tokens = md.parse(text)
+    links: List[str] = []
+
+    def walk(ts):
+        for t in ts:
+            if t.type == "link_open":
+                href = _get_attr(t, "href")
+                if href:
+                    links.append(href)
+
+            # 关键：Markdown 链接常在 inline token 的 children 里
+            children = getattr(t, "children", None)
+            if children:
+                walk(children)
+
+    walk(tokens)
+    return links
+
+
+def _collect_doc_files(repo: Path) -> List[Path]:
+    """MVP：只收 README*.md + docs/**/*.md"""
+    files: List[Path] = []
+
+    for name in ["README.md", "README.MD", "readme.md", "Readme.md"]:
+        p = repo / name
+        if p.exists() and p.is_file():
+            files.append(p)
+
+    docs_dir = repo / "docs"
+    if docs_dir.exists() and docs_dir.is_dir():
+        files.extend(sorted(docs_dir.rglob("*.md")))
+
+    uniq: List[Path] = []
+    seen = set()
+    for f in files:
+        rp = str(f.resolve())
+        if rp not in seen:
+            uniq.append(f)
+            seen.add(rp)
+    return uniq
+
+
+def _make_tree_text(relpaths: List[str]) -> str:
+    tree: Dict[str, Any] = {}
+
+    def add_path(path: str):
+        parts = path.split("/")
+        node = tree
+        for part in parts[:-1]:
+            node = node.setdefault(part, {})
+        node.setdefault("__files__", []).append(parts[-1])
+
+    def render(node: Dict[str, Any], indent: int = 0) -> List[str]:
+        lines: List[str] = []
+        dirs = sorted([k for k in node.keys() if k != "__files__"])
+        files = sorted(node.get("__files__", []))
+        for d in dirs:
+            lines.append("  " * indent + f"- {d}/")
+            lines.extend(render(node[d], indent + 1))
+        for f in files:
+            lines.append("  " * indent + f"- {f}")
+        return lines
+
+    for rp in relpaths:
+        add_path(rp)
+
+    lines = render(tree, 0)
+    return ("\n".join(lines) + "\n") if lines else ""
+
+def _anchors_in_markdown(text: str) -> set[str]:
+    """非常粗糙的锚点提取：把 '# Title' 变成 'title' 这种 slug（够 MVP 用）"""
+    anchors: set[str] = set()
+    for line in text.splitlines():
+        s = line.strip()
+        if s.startswith("#"):
+            # 去掉前面的 ### 和空格
+            title = s.lstrip("#").strip()
+            if not title:
+                continue
+            slug = (
+                title.lower()
+                .replace(" ", "-")
+                .replace("/", "-")
+            )
+            anchors.add(slug)
+    return anchors
+
+def _resolve_relative_link(doc_rel: str, href: str) -> str:
+    """
+    把 './x.md' 这种相对链接解析成 repo 内的相对路径（POSIX 风格）
+    doc_rel: 当前文档相对路径，如 'docs/setup.md'
+    """
+    base_dir = str(Path(doc_rel).parent).replace("\\", "/")
+    # base_dir 可能是 '.'，处理成空
+    if base_dir == ".":
+        base_dir = ""
+    joined = (Path(base_dir) / href).as_posix()
+    # 规范化 ./ ../
+    norm = Path(joined).as_posix()
+    return norm
+
+def _health_check(repo: Path, docfiles: List[DocFile], links_map: Dict[str, List[str]]) -> Dict[str, Any]:
+    """MVP：只检查两类：缺文件、缺锚点（只处理 md 内部链接）"""
+    # 方便查文件是否存在
+    existing_paths = {d.relpath for d in docfiles}
+    # 方便查某个文档有哪些 anchors
+    anchors_map = {d.relpath: _anchors_in_markdown(d.text) for d in docfiles}
+
+    issues: List[Dict[str, Any]] = []
+
+    for doc_rel, hrefs in links_map.items():
+        for href in hrefs:
+            if href.startswith("#"):
+                # 锚点检查
+                anchor = href[1:]
+                if anchor and anchor not in anchors_map.get(doc_rel, set()):
+                    issues.append(
+                        {"type": "missing_anchor", "doc": doc_rel, "target": href}
+                    )
+                continue
+            # 只处理 repo 内相对 md 链接（MVP：以 .md 结尾或以 ./ ../ 开头）
+            if href.startswith("http://") or href.startswith("https://"):
+                continue
+            if not (href.endswith(".md") or href.startswith("./") or href.startswith("../")):
+                continue
+
+            resolved = _resolve_relative_link(doc_rel, href)
+            # 统一成 posix
+            resolved = resolved.replace("\\", "/")
+
+            if resolved not in existing_paths:
+                issues.append(
+                    {
+                        "type": "missing_file",
+                        "doc": doc_rel,
+                        "target": href,
+                        "resolved_path": resolved,
+                    }
+                )
+
+    return {"schema": "ai-dockpack/health@0.1", "issues": issues}                            
+
+def _read_tool_version() -> str:
+    """
+    读取“工具自身”的版本（ai-dockpack 工具项目版本）
+    优先：工具项目根目录的 pyproject.toml
+    兜底：importlib.metadata 读取已安装包版本
+    """
+    tool_root = Path(__file__).resolve().parents[2]  # .../ai-dockpack/
+    pyproject = tool_root / "pyproject.toml"
+    if pyproject.exists():
+        text = pyproject.read_text(encoding="utf-8", errors="ignore")
+        m = re.search(r'^\s*version\s*=\s*"([^"]+)"\s*$', text, flags=re.M)
+        if m:
+            return m.group(1).strip()
+
+    try:
+        return importlib_metadata.version("ai-dockpack")
+    except Exception:
+        return ""
+
+def _parse_semver(v: str):
+    """把 '1.2.3' -> (1,2,3)，解析失败返回 None（MVP）"""
+    m = re.match(r"^\s*(\d+)\.(\d+)\.(\d+)\s*$", v)
+    if not m:
+        return None
+    return tuple(int(x) for x in m.groups())
+
+def _freshness_check(repo: Path, docfiles: List[DocFile]) -> Dict[str, Any]:
+    """
+    MVP：只做一种规则
+    - 在文档中找 `ai-dockpack==X.Y.Z`
+    - 若 X.Y.Z < 当前项目版本（repo/pyproject.toml 的 version），报告 outdated_pinned_version
+    """
+    current = _read_tool_version()
+    current_sem = _parse_semver(current) if current else None
+
+    issues: List[Dict[str, Any]] = []
+    if not current_sem:
+        return {"schema": "ai-dockpack/freshness@0.1", "current_version": current, "issues": issues}
+
+    pattern = re.compile(r"ai-dockpack==(\d+\.\d+\.\d+)")
+
+    for d in docfiles:
+        for m in pattern.finditer(d.text):
+            pinned = m.group(1)
+            pinned_sem = _parse_semver(pinned)
+            if pinned_sem and pinned_sem < current_sem:
+                issues.append(
+                    {
+                        "type": "outdated_pinned_version",
+                        "doc": d.relpath,
+                        "found": pinned,
+                        "current": current,
+                    }
+                )
+
+    return {"schema": "ai-dockpack/freshness@0.1", "current_version": current, "issues": issues}        
+
+def _slugify(title: str) -> str:
+    s = title.strip().lower()
+    s = s.replace("/", "-")
+    s = "-".join(s.split())
+    return s
+
+def _split_markdown_by_headings(text: str) -> List[Dict[str, str]]:
+    """
+    极简切段：
+    - 以 '#', '##', '###' 开头的行作为新段落起点
+    - 每段返回：section(title)、anchor(slug)、content
+    """
+    lines = text.splitlines()
+    chunks: List[Dict[str, str]] = []
+
+    current_title = "Document"
+    current_anchor = "document"
+    buf: List[str] = []
+
+    def flush():
+        nonlocal buf
+        content = "\n".join(buf).strip()
+        if content:
+            chunks.append(
+                {"section": current_title, "anchor": current_anchor, "content": content}
+            )
+        buf = []
+
+    for line in lines:
+        s = line.strip()
+        if s.startswith("#"):
+            # 标题行：先把上一段 flush
+            flush()
+            title = s.lstrip("#").strip()
+            current_title = title if title else "Untitled"
+            current_anchor = _slugify(current_title)
+            # 标题行本身也写进 chunk 内容（AI 读更自然）
+            buf.append(line)
+        else:
+            buf.append(line)
+
+    flush()
+    return chunks    
+
+def _render_report(manifest: Dict[str, Any], health: Dict[str, Any], freshness: Dict[str, Any]) -> str:
+    doc_count = manifest.get("doc_count", 0)
+    chunk_count = manifest.get("chunk_count", 0)
+
+    health_issues = health.get("issues", []) or []
+    fresh_issues = freshness.get("issues", []) or []
+
+    lines: List[str] = []
+    lines.append("# AI DocPack Report")
+    lines.append("")
+    lines.append("## Summary")
+    lines.append(f"- docs: **{doc_count}**")
+    lines.append(f"- chunks: **{chunk_count}**")
+    lines.append(f"- health issues: **{len(health_issues)}**")
+    lines.append(f"- freshness issues: **{len(fresh_issues)}**")
+    lines.append("")
+
+    lines.append("## Doc Health (broken / missing)")
+    if not health_issues:
+        lines.append("- ✅ no issues")
+    else:
+        for it in health_issues:
+            t = it.get("type", "")
+            doc = it.get("doc", "")
+            target = it.get("target", "")
+            if t == "missing_file":
+                rp = it.get("resolved_path", "")
+                lines.append(f"- ❌ **missing_file** in `{doc}` → `{target}` (resolved: `{rp}`)")
+            elif t == "missing_anchor":
+                lines.append(f"- ❌ **missing_anchor** in `{doc}` → `{target}`")
+            else:
+                lines.append(f"- ❌ **{t}** in `{doc}` → `{target}`")
+    lines.append("")
+
+    lines.append("## Freshness (possibly outdated)")
+    if not fresh_issues:
+        lines.append("- ✅ no issues")
+    else:
+        for it in fresh_issues:
+            t = it.get("type", "")
+            doc = it.get("doc", "")
+            found = it.get("found", "")
+            current = it.get("current", freshness.get("current_version", ""))
+            if t == "outdated_pinned_version":
+                lines.append(f"- ⚠️ **outdated_pinned_version** in `{doc}`: found `{found}`, current `{current}`")
+            else:
+                lines.append(f"- ⚠️ **{t}** in `{doc}`")
+    lines.append("")
+    return "\n".join(lines)
+
+def build_docpack(repo_path: str, out_zip: str) -> Path:
+    repo = Path(repo_path).resolve()
+    if not repo.exists():
+        raise FileNotFoundError(f"repo not found: {repo}")
+
+    doc_paths = _collect_doc_files(repo)
+
+    docfiles: List[DocFile] = []
+    all_links: Dict[str, List[str]] = {}
+
+    for p in doc_paths:
+        rel = str(p.relative_to(repo)).replace("\\", "/")
+        text = _read_text(p)
+        links = _extract_links_md(text)
+
+        docfiles.append(DocFile(relpath=rel, abspath=p, text=text, links=links))
+        if links:
+            all_links[rel] = links
+    chunk_total = sum(len(_split_markdown_by_headings(d.text)) for d in docfiles)
+    tree_txt = _make_tree_text([d.relpath for d in docfiles])
+
+    manifest: Dict[str, Any] = {
+        "schema": "ai-dockpack/manifest@0.1",
+        "repo": str(repo),
+        "doc_count": len(docfiles),
+        "chunk_count": chunk_total,
+        "docs": [
+            {
+                "path": d.relpath,
+                "bytes": len(d.text.encode("utf-8", errors="ignore")),
+                "sha256": _sha256_text(d.text),
+                "mtime": int(d.abspath.stat().st_mtime),
+                "links_count": len(d.links),
+            }
+            for d in docfiles
+        ],
+        "generated_at": __import__("datetime").datetime.utcnow().isoformat() + "Z",
+    }
+
+    out = Path(out_zip).resolve()
+
+    with zipfile.ZipFile(out, "w", compression=zipfile.ZIP_DEFLATED) as z:
+        z.writestr("manifest.json", json.dumps(manifest, ensure_ascii=False, indent=2))
+        z.writestr("tree.txt", tree_txt)
+        z.writestr("links.json", json.dumps(all_links, ensure_ascii=False, indent=2))
+        health = _health_check(repo, docfiles, all_links)
+        z.writestr("health.json", json.dumps(health, ensure_ascii=False, indent=2))
+        freshness = _freshness_check(repo, docfiles)
+        z.writestr("freshness.json", json.dumps(freshness, ensure_ascii=False, indent=2))
+        report_md = _render_report(manifest, health, freshness)
+        z.writestr("report.md", report_md)
+
+        chunk_total = 0
+        for d in docfiles:
+            parts = _split_markdown_by_headings(d.text)
+            for idx, part in enumerate(parts, start=1):
+                
+
+                chunk_name = f"{d.relpath.replace('/', '__')}__{idx:03d}.txt"
+                chunk_text = part["content"]
+                chunk_sha = _sha256_text(chunk_text)
+
+                header = (
+                    "---\n"
+                    f"source_path: {d.relpath}\n"
+                    f"section: {part['section']}\n"
+                    f"anchor: {part['anchor']}\n"
+                    f"bytes: {len(chunk_text.encode('utf-8', errors='ignore'))}\n"
+                    f"sha256: {chunk_sha}\n"
+                    "---\n\n"
+                )
+
+                z.writestr(f"chunks/{chunk_name}", header + chunk_text + "\n")
+
+    return out

ai_dockpack/cli.py

@@ -0,0 +1,35 @@
+import typer
+from rich import print
+
+app = typer.Typer(help="ai-dockpack: AI-readable doc pack generator (CLI)")
+
+@app.command()
+def build(
+    repo: str = typer.Option(".", help="要扫描的仓库路径（默认当前目录）"),
+    out: str = typer.Option("docpack.zip", help="输出压缩包文件名（先占位）"),
+):
+    """生成 docpack.zip（Step1 先把命令跑通，后续再实现真正打包逻辑）"""
+    from .build_impl import build_docpack
+    out_path = build_docpack(repo, out)
+    print(f"[green]OK[/green] 已生成：{out_path}")
+
+
+@app.command()
+def about():
+    """显示工具信息（用于强制进入“多命令模式”）"""
+    print("[cyan]ai-dockpack[/cyan] - AI-readable doc pack generator (CLI)")
+
+if __name__ == "__main__":
+    app()
+
+@app.command("init")
+def init_cmd(
+    repo: str = typer.Option(".", help="要写入 workflow 的仓库路径（默认当前目录）"),
+    release: bool = typer.Option(True, "--release/--no-release", help="是否发布到 GitHub Releases"),
+    force: bool = typer.Option(False, "--force", help="覆盖已存在的 docpack.yml"),
+):
+    """一键生成 .github/workflows/docpack.yml（让仓库自动产出 docpack.zip）"""
+    from .init_impl import init_workflow
+
+    wf = init_workflow(repo, publish_release=release, force=force)
+    print(f"[green]OK[/green] 已生成 workflow：{wf}")

ai_dockpack/init_impl.py

@@ -0,0 +1,84 @@
+from __future__ import annotations
+
+from pathlib import Path
+from textwrap import dedent,indent
+
+
+def init_workflow(repo_path: str, *, publish_release: bool = True, force: bool = False) -> Path:
+    """
+    在目标仓库生成 .github/workflows/docpack.yml
+    - publish_release=True: 生成 Release（Assets 里可直接下载 docpack.zip）
+    - publish_release=False: 只上传 Actions Artifact
+    """
+    repo = Path(repo_path).resolve()
+    if not repo.exists():
+        raise FileNotFoundError(f"repo not found: {repo}")
+
+    wf_dir = repo / ".github" / "workflows"
+    wf_dir.mkdir(parents=True, exist_ok=True)
+
+    wf_file = wf_dir / "docpack.yml"
+    if wf_file.exists() and not force:
+        raise FileExistsError(f"{wf_file} already exists (use --force to overwrite)")
+
+    # permissions（只有发布 release 时才需要）
+    permissions_block = ""
+    if publish_release:
+        permissions_block = "permissions:\n  contents: write\n\n"
+
+    # Release step：先写成“顶格”，再统一缩进到 steps 的层级（6 个空格）
+    release_step = ""
+    if publish_release:
+        release_step = indent(
+            dedent(
+                """\
+                - name: Publish Release (docpack.zip)
+                  uses: softprops/action-gh-release@v2
+                  with:
+                    tag_name: docpack-${{ github.run_number }}
+                    name: DocPack ${{ github.run_number }}
+                    prerelease: true
+                    files: docpack.zip
+                  env:
+                    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+                """
+            ).rstrip(),
+            "      ",
+        )
+
+    yml = (
+        "name: Build DocPack\n\n"
+        "on:\n"
+        "  push:\n"
+        '    branches: ["main"]\n'
+        "  workflow_dispatch:\n\n"
+        f"{permissions_block}"
+        "jobs:\n"
+        "  build-docpack:\n"
+        "    runs-on: ubuntu-latest\n"
+        "    steps:\n"
+        "      - name: Checkout\n"
+        "        uses: actions/checkout@v4\n\n"
+        "      - name: Setup Python\n"
+        "        uses: actions/setup-python@v5\n"
+        "        with:\n"
+        '          python-version: "3.11"\n\n'
+        "      - name: Install ai-dockpack (from GitHub)\n"
+        "        run: |\n"
+        "          python -m pip install --upgrade pip\n"
+        '          pip install --no-cache-dir --force-reinstall "git+https://github.com/Kuan-Peng/ai-dockpack.git@main"\n\n'
+        "      - name: Build docpack.zip\n"
+        "        run: |\n"
+        "          ai-dockpack build --repo . --out docpack.zip\n\n"
+        "      - name: Upload artifact\n"
+        "        uses: actions/upload-artifact@v4\n"
+        "        with:\n"
+        "          name: docpack\n"
+        "          path: docpack.zip\n"
+    )
+
+    if publish_release:
+        yml = yml + "\n" + release_step + "\n"
+
+    wf_file.write_text(yml, encoding="utf-8")
+    return wf_file

ai_dockpack-0.1.0.dist-info/METADATA

@@ -0,0 +1,14 @@
+Metadata-Version: 2.4
+Name: ai-dockpack
+Version: 0.1.0
+Summary: AI-readable doc pack generator (CLI)
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: typer>=0.12
+Requires-Dist: rich>=13
+Requires-Dist: markdown-it-py>=3.0.0
+Dynamic: license-file
+
+# ai-dockpack
+AI-readable doc pack generator (CLI)

ai_dockpack-0.1.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+ai_dockpack/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ai_dockpack/build_impl.py,sha256=m7kmKbSqWANwLsZp5lLb0Bx9E8J9kb42Tzd2lQBsHIs,14394
+ai_dockpack/cli.py,sha256=slarLqMHV-cuAUEe4rj02-TXoeoHiNJdiaCyU0Ebg5I,1373
+ai_dockpack/init_impl.py,sha256=BFSI7hrdVNaofz8MoYfABAmQKDt2WmjUtQCWWIFI0Go,2957
+ai_dockpack-0.1.0.dist-info/licenses/LICENSE,sha256=k6iVkamcaiKJ63W8XHdUJEg1VbOMrb8s7a0tM9NzffQ,1066
+ai_dockpack-0.1.0.dist-info/METADATA,sha256=B-J-VFJ9B5L0OV4v7ExMMdstL4uOGlN9lUQgmmxwL88,348
+ai_dockpack-0.1.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+ai_dockpack-0.1.0.dist-info/entry_points.txt,sha256=SGl4EPmqn0V8M4m3A-JkYmEYYptf0Zin9JcGxKR5Jfg,52
+ai_dockpack-0.1.0.dist-info/top_level.txt,sha256=cJHOzSmSfahNTXKS7HGlzmoRZSbD1tc6PWAOiC7vWd0,12
+ai_dockpack-0.1.0.dist-info/RECORD,,

ai_dockpack-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.10.2)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

ai_dockpack-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+ai-dockpack = ai_dockpack.cli:app

ai_dockpack-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Kuan-Peng
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

ai_dockpack-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+ai_dockpack