docs2epub 0.1.2__tar.gz → 0.1.6__tar.gz

docs2epub-0.1.6/AGENTS.md

@@ -0,0 +1,41 @@
+# docs2epub AGENTS
+
+This file documents local conventions for working on `docs2epub`.
+
+**Focus**
+- Keep the scraper generalistic across doc-site frontends (GitBook, Docusaurus, similar).
+- Prefer resilient HTML heuristics over site-specific hacks.
+- Fail gracefully when content is missing; surface actionable errors.
+
+**Development Workflow**
+- Use TDD for bug fixes and new behaviors. Add a failing test first.
+- Prefer unit tests with `monkeypatch` and deterministic HTML fixtures.
+- Keep tests fast and offline. Only do real network checks for manual validation.
+- Run tests with `uv run pytest -q`.
+
+**Scraping Heuristics**
+- Primary crawl: sidebar/index extraction.
+- Expand index/category pages by collecting in-page content links.
+- Fallback crawl: “Next” navigation when no sidebar is found.
+- Normalize URLs: strip fragments and queries; lower-case scheme/host.
+- Filter non-doc links by extension; avoid cross-origin URLs by default.
+- Resolve relative URLs against the page URL, not the site root.
+
+**Code Layout**
+- Core crawler logic: `src/docs2epub/docusaurus_next.py`.
+- EPUB generation: `src/docs2epub/epub.py` and `src/docs2epub/pandoc_epub2.py`.
+- HTML cleanup: `src/docs2epub/kindle_html.py`.
+- Tests live in `tests/`.
+
+**Release Discipline**
+- Bump version in `pyproject.toml` for user-visible changes.
+- Run `uv lock` after bumping the version.
+- Build artifacts with `uv build` before publishing.
+- Publish with `uv publish` when explicitly requested.
+- Do not commit generated EPUBs or other artifacts.
+
+**Validation**
+- Quick manual checks (optional):
+  - `uvx --from . docs2epub https://midl.gitbook.io/midl out.epub`
+  - `uvx --from . docs2epub https://tutorial.docusaurus.io/docs/intro out.epub`
+- Clean up any generated files after validation.

{docs2epub-0.1.2 → docs2epub-0.1.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docs2epub
-Version: 0.1.2
+Version: 0.1.6
 Summary: Turn documentation sites into an EPUB (Kindle-friendly).
 Author: Breno Brito
 License: MIT

{docs2epub-0.1.2 → docs2epub-0.1.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "docs2epub"
-version = "0.1.2"
+version = "0.1.6"
 description = "Turn documentation sites into an EPUB (Kindle-friendly)."
 readme = "README.md"
 requires-python = ">=3.12"

docs2epub-0.1.6/src/docs2epub/docusaurus_next.py

@@ -0,0 +1,412 @@
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass
+from urllib.parse import urljoin, urlparse
+
+import requests
+from bs4 import BeautifulSoup, Tag
+
+from .model import Chapter
+
+
+DEFAULT_USER_AGENT = "docs2epub/0.1 (+https://github.com/brenorb/docs2epub)"
+
+_SIDEBAR_SELECTORS = [
+  'aside[data-testid="table-of-contents"]',
+  "aside#table-of-contents",
+  'nav[aria-label="Table of contents"]',
+  'nav[aria-label="Table of Contents"]',
+  'nav[aria-label="Docs sidebar"]',
+  'nav[aria-label="Docs navigation"]',
+  'nav[aria-label="Documentation"]',
+  'nav[aria-label="Docs"]',
+  "aside.theme-doc-sidebar-container",
+  "div.theme-doc-sidebar-container",
+  "nav.theme-doc-sidebar-menu",
+  "nav.menu",
+  'nav[class*="menu"]',
+  'aside[class*="sidebar"]',
+  'nav[class*="sidebar"]',
+]
+
+_NON_DOC_EXTENSIONS = {
+  ".png",
+  ".jpg",
+  ".jpeg",
+  ".gif",
+  ".svg",
+  ".webp",
+  ".css",
+  ".js",
+  ".map",
+  ".json",
+  ".xml",
+  ".rss",
+  ".pdf",
+  ".zip",
+  ".tar",
+  ".gz",
+  ".tgz",
+  ".epub",
+  ".mp4",
+  ".mp3",
+  ".wav",
+}
+
+
+@dataclass(frozen=True)
+class DocusaurusNextOptions:
+  start_url: str
+  base_url: str | None = None
+  max_pages: int | None = None
+  sleep_s: float = 0.5
+  user_agent: str = DEFAULT_USER_AGENT
+
+
+def _slugify_filename(text: str) -> str:
+  value = text.strip().lower()
+  value = re.sub(r"[^\w\s-]", "", value)
+  value = re.sub(r"[\s_-]+", "-", value)
+  value = value.strip("-")
+  return value or "chapter"
+
+
+def _extract_article(soup: BeautifulSoup) -> Tag:
+  article = soup.find("article")
+  if article:
+    return article
+  main = soup.find("main")
+  if main:
+    article = main.find("article")
+    if article:
+      return article
+    return main
+  role_main = soup.find(attrs={"role": "main"})
+  if role_main:
+    return role_main
+  raise RuntimeError("Could not find <article> in page HTML")
+
+
+def _canonicalize_url(url: str) -> str:
+  parsed = urlparse(url)
+  path = parsed.path or "/"
+  if path != "/" and path.endswith("/"):
+    path = path.rstrip("/")
+  return parsed._replace(
+    scheme=parsed.scheme.lower(),
+    netloc=parsed.netloc.lower(),
+    path=path,
+    query="",
+    fragment="",
+  ).geturl()
+
+
+def _infer_root_path(start_url: str) -> str:
+  parsed = urlparse(start_url)
+  path = (parsed.path or "").rstrip("/")
+  if not path:
+    return ""
+  parts = path.split("/")
+  if len(parts) <= 2:
+    return path
+  return "/".join(parts[:-1])
+
+
+def _path_within_root(path: str, root_path: str) -> bool:
+  if not root_path or root_path == "/":
+    return True
+  if path == root_path:
+    return True
+  root = root_path if root_path.endswith("/") else f"{root_path}/"
+  return path.startswith(root)
+
+
+def _is_probable_doc_link(url: str) -> bool:
+  parsed = urlparse(url)
+  path = (parsed.path or "").lower()
+  for ext in _NON_DOC_EXTENSIONS:
+    if path.endswith(ext):
+      return False
+  return True
+
+
+def _sidebar_candidates(soup: BeautifulSoup) -> list[Tag]:
+  seen: set[int] = set()
+  candidates: list[Tag] = []
+
+  for selector in _SIDEBAR_SELECTORS:
+    for el in soup.select(selector):
+      key = id(el)
+      if key in seen:
+        continue
+      seen.add(key)
+      candidates.append(el)
+
+  keywords = ["sidebar", "toc", "table of contents", "table-of-contents", "docs", "documentation"]
+  for el in soup.find_all(["nav", "aside", "div"]):
+    key = id(el)
+    if key in seen:
+      continue
+    label = str(el.get("aria-label") or "").lower()
+    elem_id = str(el.get("id") or "").lower()
+    data_testid = str(el.get("data-testid") or "").lower()
+    classes = " ".join(el.get("class", [])).lower()
+    haystack = " ".join([label, elem_id, data_testid, classes])
+    if any(k in haystack for k in keywords):
+      seen.add(key)
+      candidates.append(el)
+
+  return candidates
+
+
+def _looks_like_pager(container: Tag, links: list[Tag]) -> bool:
+  label = str(container.get("aria-label") or "").lower()
+  if "docs pages" in label or "breadcrumb" in label:
+    return True
+  if not links:
+    return True
+  texts = []
+  for a in links:
+    text = " ".join(a.get_text(" ", strip=True).split()).lower()
+    if text:
+      texts.append(text)
+  if not texts:
+    return False
+  pager_words = {"next", "previous", "prev", "back"}
+  return all(text in pager_words for text in texts)
+
+
+def _extract_sidebar_urls(
+  soup: BeautifulSoup,
+  *,
+  base_url: str,
+  start_url: str,
+) -> list[str]:
+  candidates = _sidebar_candidates(soup)
+  if not candidates:
+    return []
+
+  origin = urlparse(start_url).netloc.lower()
+  root_path = _infer_root_path(start_url)
+  best: list[str] = []
+  for container in candidates:
+    anchors = list(container.find_all("a", href=True))
+    if _looks_like_pager(container, anchors):
+      continue
+
+    urls: list[str] = []
+    seen: set[str] = set()
+    for a in anchors:
+      href = str(a.get("href") or "").strip()
+      if not href or href.startswith("#"):
+        continue
+      if href.startswith(("mailto:", "tel:", "javascript:")):
+        continue
+      abs_url = urljoin(base_url, href)
+      parsed = urlparse(abs_url)
+      if parsed.scheme not in ("http", "https"):
+        continue
+      if origin and parsed.netloc.lower() != origin:
+        continue
+      if not _is_probable_doc_link(abs_url):
+        continue
+      if not _path_within_root(parsed.path or "", root_path):
+        continue
+      canonical = _canonicalize_url(abs_url)
+      if canonical in seen:
+        continue
+      seen.add(canonical)
+      urls.append(canonical)
+
+    if len(urls) > len(best):
+      best = urls
+
+  return best
+
+
+def _extract_content_urls(
+  container: Tag,
+  *,
+  base_url: str,
+  start_url: str,
+) -> list[str]:
+  origin = urlparse(start_url).netloc.lower()
+  root_path = _infer_root_path(start_url)
+  urls: list[str] = []
+  seen: set[str] = set()
+
+  for a in container.find_all("a", href=True):
+    href = str(a.get("href") or "").strip()
+    if not href or href.startswith("#"):
+      continue
+    if href.startswith(("mailto:", "tel:", "javascript:")):
+      continue
+    abs_url = urljoin(base_url, href)
+    parsed = urlparse(abs_url)
+    if parsed.scheme not in ("http", "https"):
+      continue
+    if origin and parsed.netloc.lower() != origin:
+      continue
+    if not _is_probable_doc_link(abs_url):
+      continue
+    if not _path_within_root(parsed.path or "", root_path):
+      continue
+    canonical = _canonicalize_url(abs_url)
+    if canonical in seen:
+      continue
+    seen.add(canonical)
+    urls.append(canonical)
+
+  return urls
+
+
+def _remove_unwanted(article: Tag) -> None:
+  for selector in [
+    'nav[aria-label="Breadcrumbs"]',
+    'nav[aria-label="Breadcrumb"]',
+    'nav[aria-label="Docs pages"]',
+    "div.theme-doc-footer",
+    "div.theme-doc-footer-edit-meta-row",
+    "div.theme-doc-version-badge",
+    "script",
+    "style",
+    "noscript",
+    "iframe",
+    "svg",
+    "button",
+  ]:
+    for el in list(article.select(selector)):
+      el.decompose()
+
+
+def _absolutize_urls(container: Tag, base_url: str) -> None:
+  for el in container.find_all(True):
+    if el.has_attr("href"):
+      href = str(el.get("href") or "")
+      if href.startswith("/"):
+        el["href"] = urljoin(base_url, href)
+    if el.has_attr("src"):
+      src = str(el.get("src") or "")
+      if src.startswith("/"):
+        el["src"] = urljoin(base_url, src)
+
+
+def _extract_next_url(soup: BeautifulSoup, base_url: str) -> str | None:
+  nav = soup.select_one('nav[aria-label="Docs pages"]')
+  if not nav:
+    return None
+
+  for a in nav.find_all("a", href=True):
+    text = " ".join(a.get_text(" ", strip=True).split())
+    if text.lower().startswith("next"):
+      return urljoin(base_url, a["href"])
+
+  return None
+
+
+def iter_docusaurus_next(options: DocusaurusNextOptions) -> list[Chapter]:
+  session = requests.Session()
+  session.headers.update({"User-Agent": options.user_agent})
+
+  url = options.start_url
+  base_url = options.base_url or options.start_url
+
+  visited: set[str] = set()
+  chapters: list[Chapter] = []
+
+  def fetch_soup(target_url: str) -> BeautifulSoup:
+    resp = session.get(target_url, timeout=30)
+    resp.raise_for_status()
+    return BeautifulSoup(resp.text, "lxml")
+
+  initial_soup = fetch_soup(url)
+  sidebar_urls = _extract_sidebar_urls(initial_soup, base_url=base_url, start_url=url)
+  initial_key = _canonicalize_url(url)
+
+  def consume_page(target_url: str, *, soup: BeautifulSoup | None = None) -> Tag | None:
+    if options.max_pages is not None and len(chapters) >= options.max_pages:
+      return None
+    key = _canonicalize_url(target_url)
+    if key in visited:
+      return None
+    visited.add(key)
+
+    page_soup = soup
+    if page_soup is None:
+      try:
+        page_soup = fetch_soup(target_url)
+      except requests.HTTPError as exc:
+        status = exc.response.status_code if exc.response is not None else None
+        if status in {404, 410} and key != initial_key:
+          return None
+        raise
+
+    article = _extract_article(page_soup)
+    title_el = article.find(["h1", "h2"])
+    title = (
+      " ".join(title_el.get_text(" ", strip=True).split())
+      if title_el
+      else f"Chapter {len(chapters) + 1}"
+    )
+
+    _remove_unwanted(article)
+    _absolutize_urls(article, base_url=target_url)
+
+    for a in list(article.select('a.hash-link[href^="#"]')):
+      a.decompose()
+
+    html = article.decode_contents()
+    chapters.append(Chapter(index=len(chapters) + 1, title=title, url=target_url, html=html))
+
+    if options.sleep_s > 0 and (options.max_pages is None or len(chapters) < options.max_pages):
+      import time
+
+      time.sleep(options.sleep_s)
+
+    return article
+
+  if sidebar_urls:
+    if initial_key not in {_canonicalize_url(u) for u in sidebar_urls}:
+      sidebar_urls.insert(0, url)
+    queue = list(sidebar_urls)
+    discovered = {_canonicalize_url(u) for u in queue}
+    idx = 0
+    while idx < len(queue):
+      if options.max_pages is not None and len(chapters) >= options.max_pages:
+        break
+      target_url = queue[idx]
+      use_soup = initial_soup if _canonicalize_url(target_url) == initial_key else None
+      article = consume_page(target_url, soup=use_soup)
+      if article is None:
+        idx += 1
+        continue
+      extra = _extract_content_urls(article, base_url=target_url, start_url=url)
+      for link in extra:
+        key = _canonicalize_url(link)
+        if key in discovered:
+          continue
+        discovered.add(key)
+        queue.append(link)
+      idx += 1
+    return chapters
+
+  # Fallback: follow next/previous navigation.
+  current_url = url
+  soup = initial_soup
+  while True:
+    if options.max_pages is not None and len(chapters) >= options.max_pages:
+      break
+
+    article = consume_page(current_url, soup=soup)
+    if article is None:
+      break
+
+    next_url = _extract_next_url(soup, base_url=base_url)
+    if not next_url:
+      break
+
+    current_url = next_url
+    soup = fetch_soup(current_url)
+
+  return chapters

docs2epub-0.1.6/tests/test_docusaurus_next.py

@@ -0,0 +1,42 @@
+from docs2epub.docusaurus_next import DocusaurusNextOptions, iter_docusaurus_next
+
+
+def test_iter_docusaurus_next_falls_back_to_main_when_no_article(monkeypatch):
+  html = """
+  <!doctype html>
+  <html>
+    <body>
+      <main>
+        <div>
+          <h1>Overview</h1>
+          <p>Hello world</p>
+        </div>
+      </main>
+    </body>
+  </html>
+  """
+
+  class DummyResponse:
+    text = html
+
+    def raise_for_status(self) -> None:
+      return None
+
+  class DummySession:
+    def __init__(self) -> None:
+      self.headers = {}
+
+    def get(self, url: str, timeout: int = 30) -> DummyResponse:
+      return DummyResponse()
+
+  monkeypatch.setattr(
+    "docs2epub.docusaurus_next.requests.Session",
+    lambda: DummySession(),
+  )
+
+  options = DocusaurusNextOptions(start_url="https://example.com/docs", sleep_s=0)
+  chapters = iter_docusaurus_next(options)
+
+  assert len(chapters) == 1
+  assert chapters[0].title == "Overview"
+  assert "Hello world" in chapters[0].html

docs2epub-0.1.6/tests/test_sidebar_scrape.py

@@ -0,0 +1,163 @@
+import requests
+
+from docs2epub.docusaurus_next import DocusaurusNextOptions, iter_docusaurus_next
+
+
+def _make_session(pages: dict[str, str]):
+  class DummyResponse:
+    def __init__(self, text: str) -> None:
+      self.text = text
+
+    def raise_for_status(self) -> None:
+      return None
+
+  class DummySession:
+    def __init__(self) -> None:
+      self.headers = {}
+
+    def get(self, url: str, timeout: int = 30) -> DummyResponse:
+      if url not in pages:
+        raise AssertionError(f"unexpected url fetch: {url}")
+      return DummyResponse(pages[url])
+
+  return DummySession
+
+
+def _make_session_with_status(pages: dict[str, tuple[int, str]]):
+  class DummyResponse:
+    def __init__(self, url: str, status_code: int, text: str) -> None:
+      self.url = url
+      self.status_code = status_code
+      self.text = text
+
+    def raise_for_status(self) -> None:
+      if self.status_code >= 400:
+        raise requests.HTTPError(
+          f"{self.status_code} Client Error",
+          response=self,
+        )
+
+  class DummySession:
+    def __init__(self) -> None:
+      self.headers = {}
+
+    def get(self, url: str, timeout: int = 30) -> DummyResponse:
+      if url not in pages:
+        raise AssertionError(f"unexpected url fetch: {url}")
+      status_code, text = pages[url]
+      return DummyResponse(url, status_code, text)
+
+  return DummySession
+
+
+def test_iter_uses_gitbook_sidebar_links(monkeypatch):
+  start_url = "https://example.com/book/intro"
+  sidebar = """
+  <aside data-testid="table-of-contents">
+    <a href="/book/intro">Intro</a>
+    <a href="/book/chapter-1">Chapter 1</a>
+  </aside>
+  """
+  pages = {
+    start_url: f"<html><body>{sidebar}<main><h1>Intro</h1><p>Intro text</p></main></body></html>",
+    "https://example.com/book/chapter-1": f"<html><body>{sidebar}<main><h1>Chapter 1</h1><p>Ch1</p></main></body></html>",
+  }
+
+  monkeypatch.setattr(
+    "docs2epub.docusaurus_next.requests.Session",
+    lambda: _make_session(pages)(),
+  )
+
+  options = DocusaurusNextOptions(start_url=start_url, sleep_s=0)
+  chapters = iter_docusaurus_next(options)
+
+  assert [c.title for c in chapters] == ["Intro", "Chapter 1"]
+
+
+def test_iter_uses_docusaurus_menu_sidebar(monkeypatch):
+  start_url = "https://example.com/docs/intro"
+  sidebar = """
+  <nav class="menu">
+    <a class="menu__link" href="/docs/intro">Intro</a>
+    <a class="menu__link" href="/docs/install">Install</a>
+  </nav>
+  """
+  pages = {
+    start_url: f"<html><body>{sidebar}<article><h1>Intro</h1><p>Intro text</p></article></body></html>",
+    "https://example.com/docs/install": f"<html><body>{sidebar}<article><h1>Install</h1><p>Install text</p></article></body></html>",
+  }
+
+  monkeypatch.setattr(
+    "docs2epub.docusaurus_next.requests.Session",
+    lambda: _make_session(pages)(),
+  )
+
+  options = DocusaurusNextOptions(start_url=start_url, sleep_s=0)
+  chapters = iter_docusaurus_next(options)
+
+  assert [c.title for c in chapters] == ["Intro", "Install"]
+
+
+def test_iter_expands_links_from_index_pages(monkeypatch):
+  start_url = "https://example.com/docs/intro"
+  sidebar = """
+  <nav class="menu">
+    <a class="menu__link" href="/docs/intro">Intro</a>
+    <a class="menu__link" href="/docs/category/getting-started">Getting Started</a>
+  </nav>
+  """
+  pages = {
+    start_url: f"<html><body>{sidebar}<article><h1>Intro</h1><p>Intro text</p></article></body></html>",
+    "https://example.com/docs/category/getting-started": (
+      "<html><body>"
+      f"{sidebar}"
+      '<article><h1>Getting Started</h1>'
+      '<a href="/docs/one">One</a>'
+      '<a href="/docs/two">Two</a>'
+      "</article></body></html>"
+    ),
+    "https://example.com/docs/one": f"<html><body>{sidebar}<article><h1>One</h1><p>One text</p></article></body></html>",
+    "https://example.com/docs/two": f"<html><body>{sidebar}<article><h1>Two</h1><p>Two text</p></article></body></html>",
+  }
+
+  monkeypatch.setattr(
+    "docs2epub.docusaurus_next.requests.Session",
+    lambda: _make_session(pages)(),
+  )
+
+  options = DocusaurusNextOptions(start_url=start_url, sleep_s=0)
+  chapters = iter_docusaurus_next(options)
+
+  assert [c.title for c in chapters] == ["Intro", "Getting Started", "One", "Two"]
+
+
+def test_iter_skips_sidebar_links_that_404(monkeypatch):
+  start_url = "https://example.com/docs/intro"
+  sidebar = """
+  <nav class="menu">
+    <a class="menu__link" href="/docs/intro">Intro</a>
+    <a class="menu__link" href="/docs/missing">Missing</a>
+    <a class="menu__link" href="/docs/other">Other</a>
+  </nav>
+  """
+  pages = {
+    start_url: (
+      200,
+      f"<html><body>{sidebar}<article><h1>Intro</h1><p>Intro text</p></article></body></html>",
+    ),
+    "https://example.com/docs/missing": (404, "Not found"),
+    "https://example.com/docs/other": (
+      200,
+      f"<html><body>{sidebar}<article><h1>Other</h1><p>Other text</p></article></body></html>",
+    ),
+  }
+
+  monkeypatch.setattr(
+    "docs2epub.docusaurus_next.requests.Session",
+    lambda: _make_session_with_status(pages)(),
+  )
+
+  options = DocusaurusNextOptions(start_url=start_url, sleep_s=0)
+  chapters = iter_docusaurus_next(options)
+
+  assert [c.title for c in chapters] == ["Intro", "Other"]

{docs2epub-0.1.2 → docs2epub-0.1.6}/uv.lock

@@ -166,7 +166,7 @@ wheels = [
 
 [[package]]
 name = "docs2epub"
-version = "0.1.1"
+version = "0.1.6"
 source = { editable = "." }
 dependencies = [
     { name = "beautifulsoup4" },

docs2epub-0.1.2/src/docs2epub/docusaurus_next.py

@@ -1,140 +0,0 @@
-from __future__ import annotations
-
-import re
-from dataclasses import dataclass
-from urllib.parse import urljoin
-
-import requests
-from bs4 import BeautifulSoup, Tag
-
-from .model import Chapter
-
-
-DEFAULT_USER_AGENT = "docs2epub/0.1 (+https://github.com/brenorb/docs2epub)"
-
-
-@dataclass(frozen=True)
-class DocusaurusNextOptions:
-  start_url: str
-  base_url: str | None = None
-  max_pages: int | None = None
-  sleep_s: float = 0.5
-  user_agent: str = DEFAULT_USER_AGENT
-
-
-def _slugify_filename(text: str) -> str:
-  value = text.strip().lower()
-  value = re.sub(r"[^\w\s-]", "", value)
-  value = re.sub(r"[\s_-]+", "-", value)
-  value = value.strip("-")
-  return value or "chapter"
-
-
-def _extract_article(soup: BeautifulSoup) -> Tag:
-  article = soup.find("article")
-  if article:
-    return article
-  main = soup.find("main")
-  if main:
-    article = main.find("article")
-    if article:
-      return article
-  raise RuntimeError("Could not find <article> in page HTML")
-
-
-def _remove_unwanted(article: Tag) -> None:
-  for selector in [
-    'nav[aria-label="Breadcrumbs"]',
-    'nav[aria-label="Docs pages"]',
-    "div.theme-doc-footer",
-    "div.theme-doc-footer-edit-meta-row",
-    "div.theme-doc-version-badge",
-    "script",
-    "style",
-    "noscript",
-    "iframe",
-    "svg",
-    "button",
-  ]:
-    for el in list(article.select(selector)):
-      el.decompose()
-
-
-def _absolutize_urls(container: Tag, base_url: str) -> None:
-  for el in container.find_all(True):
-    if el.has_attr("href"):
-      href = str(el.get("href") or "")
-      if href.startswith("/"):
-        el["href"] = urljoin(base_url, href)
-    if el.has_attr("src"):
-      src = str(el.get("src") or "")
-      if src.startswith("/"):
-        el["src"] = urljoin(base_url, src)
-
-
-def _extract_next_url(soup: BeautifulSoup, base_url: str) -> str | None:
-  nav = soup.select_one('nav[aria-label="Docs pages"]')
-  if not nav:
-    return None
-
-  for a in nav.find_all("a", href=True):
-    text = " ".join(a.get_text(" ", strip=True).split())
-    if text.lower().startswith("next"):
-      return urljoin(base_url, a["href"])
-
-  return None
-
-
-def iter_docusaurus_next(options: DocusaurusNextOptions) -> list[Chapter]:
-  session = requests.Session()
-  session.headers.update({"User-Agent": options.user_agent})
-
-  url = options.start_url
-  base_url = options.base_url or options.start_url
-
-  visited: set[str] = set()
-  chapters: list[Chapter] = []
-
-  idx = 1
-  while True:
-    if options.max_pages is not None and idx > options.max_pages:
-      break
-
-    if url in visited:
-      break
-    visited.add(url)
-
-    resp = session.get(url, timeout=30)
-    resp.raise_for_status()
-
-    soup = BeautifulSoup(resp.text, "lxml")
-    article = _extract_article(soup)
-
-    title_el = article.find(["h1", "h2"])
-    title = (
-      " ".join(title_el.get_text(" ", strip=True).split()) if title_el else f"Chapter {idx}"
-    )
-
-    _remove_unwanted(article)
-    _absolutize_urls(article, base_url=base_url)
-
-    for a in list(article.select('a.hash-link[href^="#"]')):
-      a.decompose()
-
-    html = article.decode_contents()
-
-    chapters.append(Chapter(index=idx, title=title, url=url, html=html))
-
-    next_url = _extract_next_url(soup, base_url=base_url)
-    if not next_url:
-      break
-
-    url = next_url
-    idx += 1
-
-    if options.sleep_s > 0:
-      import time
-
-      time.sleep(options.sleep_s)
-
-  return chapters