Rework category labels, slug validation, and task-list checkboxes

- Remove the section-header ➕ "add landing page" button; landing pages are
  now created via the page editor pen.
- Treat saving a page with empty (after stripping) Markdown as deleting it:
  childless nodes are removed, nodes with children become content-less labels.
- Restrict server-side slugs to [a-z0-9_-] with no leading underscore or dot;
  drop the reserved-filename blacklist.
- Render placeholder pages for content-less category labels and link them to
  their first published child in the navigation.
- Make task-list checkboxes live, non-disabled inputs: toggling updates the
  Markdown source, updates the open CodeMirror document when the page editor is
  open, and persists to the server when no editor is open. Errors revert the UI.
- Update docs and AGENTS.md conventions accordingly.

AGENTS.md

@@ -39,9 +39,9 @@ not for the public pages. See `docs/design-principles.md` for the design.
     main level pages, not their parent); it cannot have children, and
     renaming its slug away leaves no front page ("/" redirects to the
     first nav item). `Node.content` is
-    the Markdown page, or None for a pure category label whose URL
-    redirects to its first child; every label's title and slug are
-    editable. Siblings order by the fractional `Node.order` key: a moved
+    the Markdown page, or None for a pure category label whose URL renders
+    a placeholder page (while nav links to it point at its first child);
+    every label's title and slug are editable. Siblings order by the fractional `Node.order` key: a moved
     item gets a fresh key relative to its new siblings, all others keep
     theirs. `resolve`/`find_slot` walk the tree by path; moves are slot
     detach/attach carrying the whole subtree. Legacy flat `Data.pages`
@@ -65,8 +65,8 @@ not for the public pages. See `docs/design-principles.md` for the design.
   - `views.py` — the shared page layout as an html5tagger `Template` with
     placeholders (`Title`, `Brand`, `Banner`, `Nav`, `Sidebar`, `Main`), nav
     rendering straight from the `Data.menu` tree (siblings sorted by
-    `Node.order`; content-less labels redirect to their first child via
-    `first_leaf`), and page/404 rendering. If the markdown contains its own h1, the page title
+    `Node.order`; nav links to content-less labels point at their first
+    child via `first_leaf`), and page/404 rendering. If the markdown contains its own h1, the page title
     is NOT rendered as an additional h1 (it still supplies <title> and nav
     labels). The navbar holds
     top-level items only; the current section's subitems go to a left
@@ -171,15 +171,12 @@ not for the public pages. See `docs/design-principles.md` for the design.
 - Keep dependencies minimal; add via `uv add` and mention it.
 - The public URL space belongs to content (pretty slugs at root). Reserve
   only `/_/` for the machinery (files, API, built assets, admin), plus
-  `/favicon.ico` from the build. Slugs are lowercase ASCII `[a-z0-9-]`
-  (the site editor filters input live via `slugify.js`, built on the
-  `transliteration` npm package — unicode folds to ASCII, spaces become
-  hyphens; an empty slug on a new page is derived from its title), may
-  not begin with `_` or `.`, and may not be a reserved file name
-  (`robots.txt`, `ads.txt`, `sitemap.xml`, `openapi.json`, `favicon.ico`,
-  `site.webmanifest`). The API rejects such paths with a human-readable
-  reason shown in the editor, and such URLs are never looked up as
-  content when serving.
+  `/favicon.ico` from the build. Slugs are lowercase ASCII letters, digits,
+  hyphens and underscores `[a-z0-9_-]` (the site editor filters input live
+  via `slugify.js`, built on the `transliteration` npm package — unicode
+  folds to ASCII, spaces become hyphens; an empty slug on a new page is
+  derived from its title), may not begin with `_` or `.`, and such URLs are
+  never looked up as content.
 - No auth in core code; trusted single author. Never add output
   sanitization "for safety" against the author — embedded HTML/scripts in
   Markdown are passed through deliberately.

docs/design-principles.md

@@ -32,11 +32,10 @@ evolves.
   lives under `/_/` (the API at `/_/api/`, uploaded files at `/_/f/`, built
   assets at `/_/assets/`, and the admin shell at `/_/admin`). The only
   other reserved root path is `/favicon.ico`, served from the build.
-  Slugs are lowercase ASCII (`[a-z0-9-]`; input is transliterated and
-  filtered as you type, and a new page's empty slug is derived from its
-  title), may not begin with `_` or `.`, and may not occupy a reserved
-  root file name (`robots.txt`, `sitemap.xml`, `favicon.ico`, …) — such
-  URLs are never looked up as content.
+  Slugs are lowercase ASCII letters, digits, hyphens and underscores
+  (`[a-z0-9_-]`; input is transliterated and filtered as you type, and a
+  new page's empty slug is derived from its title), may not begin with
+  `_` or `.`, and such URLs are never looked up as content.
 - **Single user, trusted author.** No auth concerns in the core design.
   Everything published is public; only editing tools will later sit behind
   access control (external SSO when that time comes). The author is trusted
@@ -105,9 +104,11 @@ evolves.
   the sidebar is empty (and hidden) elsewhere. Other sections' subitems
   are never shown without navigating into them first.
 - **Landing pages are optional.** Every label can either have content
-  (`Node.content`, a Markdown page) or none — a content-less label
-  redirects to its first child instead of 404ing, so categories need no
-  filler content. Title and slug of every label are editable; renaming a
+  (`Node.content`, a Markdown page) or none — a content-less label renders
+  a placeholder page (404 with a pen to create it) instead of redirecting,
+  while nav links to it point straight at its first child, so categories
+  need no filler content and normal navigation never sees the placeholder.
+  Title and slug of every label are editable; renaming a
   slug moves the whole subtree. The sidebar never lists the section
   itself, avoiding title duplication with the navbar.
 - **Menu order is manual.** Each node has a fractional `order` key among

frontend/src/PageEditor.vue

@@ -120,8 +120,10 @@ function openPath(p) {
   send({ type: 'open', path: p })
 }
 
-function setDocument(text) {
-  view.dispatch({ changes: { from: 0, to: view.state.doc.length, insert: text } })
+function setDocument(text, preserveSelection = false) {
+  const tr = { changes: { from: 0, to: view.state.doc.length, insert: text } }
+  if (preserveSelection) tr.selection = view.state.selection
+  view.dispatch(tr)
 }
 
 function onHashChange() {
@@ -272,6 +274,11 @@ onMounted(() => {
     parent: editorEl.value,
   })
   view.scrollDOM.addEventListener('scroll', syncScroll)
+  window.__pageritePageEditor = {
+    getMarkdown: () => view.state.doc.toString(),
+    setMarkdown: (text) => setDocument(text, true),
+    path: () => path.value,
+  }
   if (props.standalone) addEventListener('hashchange', onHashChange)
   addEventListener('keydown', onKeydown)
 })
@@ -283,6 +290,7 @@ onUnmounted(() => {
     ws.close()
   }
   view?.destroy()
+  delete window.__pageritePageEditor
   if (props.standalone) removeEventListener('hashchange', onHashChange)
   removeEventListener('keydown', onKeydown)
 })

frontend/src/SiteEditor.vue

@@ -8,8 +8,8 @@
 //
 // The tree comes from the server nested (GET /_/api/pages); every node is
 // real — a label with a title and slug, with content (landing page) or
-// without (category redirecting to its first child). The front page is a
-// top-level row with an empty slug, not the parent of the others.
+// without (category whose URL renders a placeholder page). The front page
+// is a top-level row with an empty slug, not the parent of the others.
 import { computed, onMounted, onUnmounted, provide, ref } from 'vue'
 import StructureTree from './StructureTree.vue'
 import { EditorView, basicSetup } from 'codemirror'
@@ -125,8 +125,8 @@ async function loadPlain(p) {
     const res = await fetch(finalUrl)
     const type = res.headers.get('content-type') || ''
     if (!type.includes('text/html')) return
-    // Category URLs redirect to their first child; reflect that. A 404
-    // layout is fine too (new pages are created by editing them).
+    // Category and missing URLs render a placeholder 404 page — fine to
+    // swap in (new pages are created by editing them).
     if (res.redirected) finalUrl = res.url
     doc = new DOMParser().parseFromString(await res.text(), 'text/html')
   } catch { return }
@@ -239,22 +239,6 @@ async function commitPending() {
   navigate(newPath)
 }
 
-// Give a content-less category a landing page (empty page at its path).
-async function addContent(node) {
-  const res = await fetch(`/_/api/pages/${node.path}`, {
-    method: 'PUT',
-    headers: { 'content-type': 'application/json' },
-    body: JSON.stringify({ title: node.title, markdown: '', published: node.published }),
-  })
-  if (res.ok) {
-    saveError.value = ''
-    await refreshPages()
-    navigate(node.path)
-  } else {
-    saveError.value = '⚠️ changes could not be saved'
-  }
-}
-
 // --- Site-wide brand (header link + <title> suffix) ----------------------
 // Edits apply to the live page immediately and save while typing. An
 // empty brand removes the header link and the title suffix entirely.
@@ -330,8 +314,8 @@ async function removePage(node) {
     refreshPages()
     const p = node.path
     if (p === path.value || (p && path.value.startsWith(`${p}/`))) {
-      // The current page was deleted — or reduced to a category that now
-      // redirects to its first child. Either way, re-render from the server.
+      // The current page was deleted — or reduced to a category, which now
+      // renders a placeholder page. Either way, re-render from the server.
       if (node.children.length) loadPlain(path.value)
       else navigate('')
     } else {
@@ -429,7 +413,6 @@ provide('structureHandlers', {
   reorder: onReorder,
   titleInput: onTitleInput,
   commitSlug,
-  addContent,
   commitPending,
   discardPending,
   newPage,

frontend/src/StructureTree.vue

@@ -5,8 +5,8 @@
 // Every node is real: a label whose title and slug are always editable
 // inline — the title saves while typing (and focusing it opens the page),
 // the slug commits on blur/Enter since it renames the path, moving the
-// whole subtree. Nodes without content are category labels that redirect
-// to their first child; the ➕ on their row gives them a landing page.
+// whole subtree. Nodes without content are category labels whose URL
+// renders a placeholder page; the ➕ on their row gives them a landing page.
 // Every non-empty list (and the root list) ends with a ➕ footer row:
 // clicking it adds a *pending* row (a local-only item persisted to the
 // server only on commit, ✓/Enter, Esc discards) at the end of that list,
@@ -140,13 +140,6 @@ function onEnd() {
             />
             <span class="acts">
               <span v-if="!element.published" class="draft">draft</span>
-              <button
-                v-if="!element.has_content"
-                type="button"
-                class="act"
-                title="add a landing page (currently redirects to the first child)"
-                @click="handlers.addContent(element)"
-              >➕</button>
               <button
                 type="button"
                 class="act del"

frontend/src/assets/style.css

@@ -316,11 +316,17 @@ article ul ul ul li::before {
   content: "🔹";
 }
 
-/* Task lists render emoji checkmarks (see markdown.py), no diamond. */
+/* Task lists: real checkboxes (not emoji), clickable in the public page. */
 article .task-list-item::before {
   content: none;
 }
 
+article .task-list-item-checkbox {
+  cursor: pointer;
+  margin-inline-end: 0.35em;
+  vertical-align: middle;
+}
+
 article {
   position: relative;
 }

frontend/src/pagerite.js

@@ -169,7 +169,7 @@ import "./assets/style.css";
       const res = await fetch(url);
       const type = res.headers.get("content-type") || "";
       if (!res.ok || !type.includes("text/html")) throw new Error("not a page");
-      // Section URLs redirect to their first child; reflect that.
+      // Reflect any redirect the server issued.
       if (res.redirected) finalUrl = res.url;
       doc = new DOMParser().parseFromString(await res.text(), "text/html");
     } catch {
@@ -255,5 +255,41 @@ import "./assets/style.css";
 
   addEventListener("popstate", () => load(location.href, false, true));
 
+  // --- Task-list checkboxes ------------------------------------------------
+  // Checkboxes in the rendered article are live: toggling them edits the
+  // Markdown source. If the page editor is open, its CodeMirror document is
+  // updated directly; otherwise the server copy is toggled and saved.
+  async function toggleTask(checkbox, index) {
+    const editor = window.__pageritePageEditor;
+    const pagePath = editor ? editor.path() : currentPath;
+    const path = pagePath.replace(/^\/+|\/+$/g, "");
+    const originalChecked = !checkbox.checked;
+    try {
+      const body = { path, index };
+      if (editor) body.markdown = editor.getMarkdown();
+      const res = await fetch("/_/api/toggle-task", {
+        method: "POST",
+        headers: { "content-type": "application/json" },
+        body: JSON.stringify(body),
+      });
+      if (!res.ok) {
+        const detail = await res.json().catch(() => ({}));
+        throw new Error(detail.detail || res.statusText);
+      }
+      const { markdown } = await res.json();
+      if (editor) editor.setMarkdown(markdown);
+    } catch {
+      checkbox.checked = originalChecked;
+    }
+  }
+
+  addEventListener("change", (ev) => {
+    const checkbox = ev.target.closest(".task-list-item-checkbox");
+    if (!checkbox) return;
+    const index = Number(checkbox.dataset.taskIndex);
+    if (Number.isNaN(index)) return;
+    toggleTask(checkbox, index);
+  });
+
   applyEffects();
 })();

pagerite/app.py

@@ -14,6 +14,7 @@ walking the tree (``resolve``), moves are slot detach/attach
 
 import mimetypes
 import os
+import re
 from collections.abc import AsyncIterator
 from contextlib import asynccontextmanager
 from datetime import UTC, datetime
@@ -37,7 +38,7 @@ from pagerite.data import (
     resolve,
     sorted_nodes,
 )
-from pagerite.markdown import has_h1, render
+from pagerite.markdown import has_h1, render, toggle_task
 
 DB_PATH = os.getenv("PAGERITE_DB", "pagerite.kanta")
 
@@ -82,6 +83,25 @@ def _ensure(menu: dict[str, Node], path: str) -> Node:
     return node
 
 
+def _remove_page_content(menu: dict[str, Node], path: str) -> None:
+    """Delete a page's markdown content.
+
+    A node with children becomes a content-less category label; a childless
+    node is removed entirely. Does nothing if the path does not exist.
+    """
+    slot = find_slot(menu, path)
+    if slot is None:
+        return
+    node = slot[0].get(slot[1])
+    if node is None:
+        return
+    if node.children:
+        node.content = None
+        node.modified = datetime.now(UTC)
+    else:
+        del slot[0][slot[1]]
+
+
 def _migrate_legacy() -> None:
     """Rebuild the legacy flat page store as a tree (one-time migration)."""
     if not data.pages:
@@ -179,18 +199,22 @@ async def save_page(path: str, page: PageIn) -> None:
     """Create or replace the page at a slug path ("" or "/" = front page).
 
     Missing ancestors are created as content-less category labels. Giving
-    a category markdown turns it into a landing page.
+    a category markdown turns it into a landing page. An empty markdown
+    string (after stripping) deletes the page instead.
     """
     path = path.strip("/")
     _check_reserved(path)
     with kanta.transaction("save page", extra=path):
-        node = _ensure(data.menu, path)
-        node.title = page.title
-        node.content = page.markdown
-        node.published = page.published
-        if page.banner is not None:
-            node.banner = page.banner
-        node.modified = datetime.now(UTC)
+        if page.markdown.strip() == "":
+            _remove_page_content(data.menu, path)
+        else:
+            node = _ensure(data.menu, path)
+            node.title = page.title
+            node.content = page.markdown
+            node.published = page.published
+            if page.banner is not None:
+                node.banner = page.banner
+            node.modified = datetime.now(UTC)
         data.version += 1
 
 
@@ -275,6 +299,44 @@ async def put_settings(settings: SettingsIn) -> None:
         data.version += 1
 
 
+class ToggleTaskIn(BaseModel):
+    """Payload for toggling one task-list checkbox."""
+
+    path: str
+    index: int
+    markdown: str | None = None
+
+
+@app.post("/_/api/toggle-task")
+async def toggle_task_endpoint(body: ToggleTaskIn) -> dict[str, str]:
+    """Toggle the Nth task-list checkbox in a page's Markdown source.
+
+    If ``markdown`` is provided the source is left untouched and the toggled
+    Markdown is returned (used while the page editor is open, so the live
+    CodeMirror document can be updated). Otherwise the stored page at
+    ``path`` is read, toggled, and saved.
+    """
+    path = body.path.strip("/")
+    _check_reserved(path)
+    if body.markdown is not None:
+        new_markdown = toggle_task(body.markdown, body.index)
+        if new_markdown is None:
+            raise HTTPException(400, "invalid task index")
+        return {"markdown": new_markdown}
+    chain = resolve(data.menu, path)
+    node = chain[-1] if chain else None
+    if node is None or node.content is None:
+        raise HTTPException(404, "no such page")
+    new_markdown = toggle_task(node.content, body.index)
+    if new_markdown is None:
+        raise HTTPException(400, "invalid task index")
+    with kanta.transaction("toggle task", extra=path):
+        node.content = new_markdown
+        node.modified = datetime.now(UTC)
+        data.version += 1
+    return {"markdown": new_markdown}
+
+
 @app.put("/_/api/files/{name}")
 async def upload_file(name: str, request: Request) -> dict[str, str]:
     """Store an upload (image, video...) in the content-addressed store.
@@ -342,32 +404,26 @@ async def delete_page(path: str) -> None:
         data.version += 1
 
 
+_SLUG_RE = re.compile(r"^[a-z0-9][a-z0-9_-]*$")
+
+
 def _is_reserved(path: str) -> bool:
-    """Slug shape that content may never use: any segment beginning with
-    "_" or "." ("/_/" is the machinery prefix; dot-segments look like
-    hidden or filesystem paths)."""
-    return any(seg.startswith(("_", ".")) for seg in path.split("/"))
-
-
-# Well-known root file names that content may not occupy (they would
-# shadow real files or carry special meaning to crawlers/clients).
-RESERVED_FILES = {
-    "robots.txt",
-    "ads.txt",
-    "sitemap.xml",
-    "openapi.json",
-    "favicon.ico",
-    "site.webmanifest",
-}
+    """Slug shape that content may never use: each segment must be lower-case
+    ASCII letters, digits, hyphens and underscores (underscores may not be
+    the first character), and dots are never allowed.
+    """
+    if path == "":
+        return False
+    return any(not _SLUG_RE.match(seg) for seg in path.split("/"))
 
 
 def _check_reserved(path: str) -> None:
-    """Reject configured slugs beginning with "_" or ".", and reserved
-    file names at the root."""
+    """Reject paths that do not follow the slug charset."""
     if _is_reserved(path):
-        raise HTTPException(400, 'slugs cannot begin with "_" or "."')
-    if path in RESERVED_FILES:
-        raise HTTPException(400, f'"{path}" is a reserved file name')
+        raise HTTPException(
+            400,
+            'slugs may only use a-z, 0-9, "-" and "_" (not as the first character), and no dots',
+        )
 
 
 @app.websocket("/_/api/ws/editor")
@@ -472,15 +528,21 @@ async def editor_ws(ws: WebSocket) -> None:
                             tnodes[tslug] = node
                         else:
                             node = old if old is not None else _ensure(data.menu, path)
-                        if "title" in msg:
-                            node.title = msg["title"]
+                        removed = False
                         if "markdown" in msg:
-                            node.content = msg["markdown"]
-                        if "published" in msg:
-                            node.published = bool(msg["published"])
-                        if "banner" in msg:
-                            node.banner = msg["banner"]
-                        node.modified = datetime.now(UTC)
+                            if msg["markdown"].strip() == "":
+                                _remove_page_content(data.menu, path)
+                                removed = True
+                            else:
+                                node.content = msg["markdown"]
+                        if not removed:
+                            if "title" in msg:
+                                node.title = msg["title"]
+                            if "published" in msg:
+                                node.published = bool(msg["published"])
+                            if "banner" in msg:
+                                node.banner = msg["banner"]
+                            node.modified = datetime.now(UTC)
                         data.version += 1
                     await ws.send_json({"type": "saved", "path": path})
     except WebSocketDisconnect:
@@ -508,12 +570,12 @@ frontend.route(app, "/")
 async def show_page(request: Request, path: str) -> HTMLResponse | Response:
     """Render the content page at a slug path, or 404.
 
-    A node without content is a category label: its URL redirects to the
-    first child page in menu order.
+    A node without content is a category label: its URL renders a
+    placeholder page (nav links point straight at its first child).
     """
     path = path.strip("/")
-    if path and (_is_reserved(path) or path in RESERVED_FILES):
-        # Reserved slug shape or file name: never content — no tree lookup.
+    if path and _is_reserved(path):
+        # Reserved slug shape: never content — no tree lookup.
         return HTMLResponse(views.render_not_found(data.menu, path, data.brand), 404)
     chain = resolve(data.menu, path)
     node = chain[-1] if chain else None
@@ -528,9 +590,9 @@ async def show_page(request: Request, path: str) -> HTMLResponse | Response:
             headers={"etag": etag},
         )
     if node is not None and node.published and node.content is None:
-        # Category label without a landing page: open its first child.
-        if (leaf := views.first_leaf(data.menu, path)) is not None:
-            return RedirectResponse(f"/{leaf}")
+        # Category label without a landing page: placeholder with the pen
+        # to create it (404 — no page here, but the node is real).
+        return HTMLResponse(views.render_category(data.menu, path, data.brand), 404)
     if node is None and not path:
         # No front page (no top-level node with slug ""): "/" opens the
         # first item of the navigation instead.

pagerite/data.py

@@ -3,8 +3,8 @@
 The site structure is a tree of Nodes. Every node is a menu label with a
 configurable title and slug (its key in the parent's ``children``); the
 URL path is the chain of slugs from the top level. ``content`` is the
-node's Markdown page, or None for a pure category label, whose URL
-redirects to the first child page.
+node's Markdown page, or None for a pure category label, whose URL renders
+a placeholder page while nav links point at its first child.
 """
 
 from datetime import UTC, datetime
@@ -29,7 +29,7 @@ class Node(msgspec.Struct, omit_defaults=True):
     title: str = ""
     order: float = 0
     #: Markdown source of the node's page; None = pure category label
-    #: (redirects to the first child page).
+    #: (its URL renders a placeholder page).
     content: str | None = None
     #: Raw HTML for the header banner (img, styled div, canvas+script...).
     #: Empty inherits the nearest ancestor's banner, front page last.

pagerite/markdown.py

@@ -13,6 +13,8 @@ page's own path (so `![alt](photo.avif)` in `/docs/design` is served from
 classes, e.g. `![alt](photo.avif "Caption"){.right}`.
 """
 
+import re
+
 from markdown_it import MarkdownIt
 from markdown_it.common.utils import escapeHtml
 from markdown_it.renderer import RendererHTML
@@ -30,6 +32,9 @@ from pygments.util import ClassNotFound
 _formatter = HtmlFormatter(style="github-dark", nowrap=True)
 
 
+_TASK_MARKER_RE = re.compile(r"^(\s*(?:>\s*)*(?:[-*+]|\d+\.)\s+)\[( |x|X)\](\s+|$)")
+
+
 def _highlight(text: str, lang: str, _attrs: str) -> str:
     """Syntax-highlight a fenced code block with Pygments.
 
@@ -83,31 +88,37 @@ def _unwrap_lone_figures(state) -> None:
                 tokens[i + 1].hidden = True
 
 
-md = (
-    MarkdownIt("default", {"html": True, "highlight": _highlight})
-    .use(attrs_plugin)
-    .use(footnote_plugin)
-    .use(deflist_plugin)
-    .use(tasklists_plugin)
-)
-def _checkbox_emojis(state) -> None:
-    """Render task-list checkboxes as emoji instead of disabled inputs.
+def _tag_task_checkboxes(state) -> None:
+    """Tag rendered task-list checkboxes with a stable index.
 
-    A disabled <input> renders grey and washed out; a colored emoji
-    shows the state without any styling.
+    The public page and the editor preview use the index to identify which
+    `[ ]`/`[x]` marker in the Markdown source to toggle when a visitor
+    clicks the checkbox.
     """
+    index = 0
     for token in state.tokens:
         if token.type != "inline" or not token.children:
             continue
         for child in token.children:
             if child.type == "html_inline" and 'type="checkbox"' in child.content:
-                child.type = "text"
-                child.content = "✅" if "checked" in child.content else "⬜"
+                child.content = child.content.replace(
+                    'type="checkbox"',
+                    f'data-task-index="{index}" type="checkbox"',
+                    1,
+                )
+                index += 1
 
 
+md = (
+    MarkdownIt("default", {"html": True, "highlight": _highlight})
+    .use(attrs_plugin)
+    .use(footnote_plugin)
+    .use(deflist_plugin)
+    .use(tasklists_plugin, enabled=True)
+)
 md.add_render_rule("image", _image_rule)
 md.core.ruler.push("unwrap_lone_figures", _unwrap_lone_figures)
-md.core.ruler.push("checkbox_emojis", _checkbox_emojis)
+md.core.ruler.push("tag_task_checkboxes", _tag_task_checkboxes)
 
 
 def render(text: str, page_path: str = "") -> str:
@@ -123,3 +134,43 @@ def has_h1(text: str) -> bool:
     <title> and navigation labels).
     """
     return any(t.type == "heading_open" and t.tag == "h1" for t in md.parse(text))
+
+
+def toggle_task(text: str, index: int) -> str | None:
+    """Toggle the Nth task-list checkbox marker in ``text``.
+
+    Returns the modified Markdown source, or ``None`` if the index is out
+    of range or the marker could not be found.
+    """
+    tokens = md.parse(text, {"page_path": ""})
+    checkbox_lines: list[int | None] = []
+    for token in tokens:
+        if token.type == "inline" and token.children:
+            for child in token.children:
+                if child.type == "html_inline" and 'type="checkbox"' in child.content:
+                    checkbox_lines.append(token.map[0] if token.map else None)
+                    break
+
+    if not (0 <= index < len(checkbox_lines)):
+        return None
+    line_idx = checkbox_lines[index]
+    if line_idx is None or line_idx < 0:
+        return None
+
+    lines = text.splitlines(keepends=True)
+    if line_idx >= len(lines):
+        return None
+    line = lines[line_idx]
+
+    def repl(m: re.Match[str]) -> str:
+        prefix = m.group(1)
+        marker = m.group(2)
+        new_marker = "x" if marker.strip() == "" else " "
+        return f"{prefix}[{new_marker}]{m.group(3)}"
+
+    new_line = _TASK_MARKER_RE.sub(repl, line, count=1)
+    if new_line == line:
+        return None
+
+    lines[line_idx] = new_line
+    return "".join(lines)

pagerite/seed.py

@@ -424,8 +424,8 @@ DUNES_SVG = """\
 
 #: path -> (title, markdown, {filename: bytes}, banner HTML, menu order).
 #: Note there are deliberately no "docs" or "blog" landing pages: those
-#: labels are created without content, so entering them redirects to the
-#: first child (see views.first_leaf).
+#: labels are created without content, so they render a placeholder page
+#: and their nav links point at the first child (see views.first_leaf).
 PAGES: dict[str, tuple[str, str, dict[str, bytes], str, float]] = {
     "": ("Welcome", WELCOME, {"waves.svg": WAVES_SVG.encode()}, FRONT_BANNER, 1),
     "about": ("About", ABOUT, {}, "", 2),

pagerite/views.py

@@ -8,8 +8,9 @@ can swap them without reloading the page chrome.
 Navigation walks the Node tree directly (see data.py): nav_html lists the
 top level — the front page (slug "") is an ordinary top-level item, not
 the parent of the others — and sidebar_html the children of the current
-top-level section. Nodes without content are category labels; their URLs
-redirect to the first child page (first_leaf).
+top-level section. Nodes without content are category labels; nav links
+to them point straight at their first child page (first_leaf), and their
+own URL renders a placeholder page (render_category).
 """
 
 from pathlib import Path
@@ -58,14 +59,18 @@ def _title(slug: str, node: Node) -> str:
     return node.title or prettify(slug) or "Home"
 
 
-def _nav_link(doc, node: Node, path: str, current: str) -> None:
-    """Render one <li> linking the node (category labels redirect to their
-    first child server-side, so linking them is always fine)."""
+def _nav_link(doc, menu: dict[str, Node], node: Node, path: str, current: str) -> None:
+    """Render one <li> linking the node. Category labels (no content of
+    their own) link straight to their first child page, so normal
+    navigation bypasses the placeholder page at their own URL."""
     # A top-level item is current also when viewing any of its subpages.
     is_current = current == path or (path and current.startswith(f"{path}/"))
+    href = f"/{path}"
+    if node.content is None and (leaf := first_leaf(menu, path)) is not None:
+        href = f"/{leaf}"
     doc.li.a(
         _title(path.rpartition("/")[2], node),
-        href=f"/{path}",
+        href=href,
         **{"class": "current"} if is_current else {},
     )
 
@@ -81,7 +86,7 @@ def nav_html(menu: dict[str, Node], current: str) -> HTML:
     with nav:
         for slug, node in sorted_nodes(menu):
             if node.published:
-                _nav_link(nav, node, slug, current)
+                _nav_link(nav, menu, node, slug, current)
     return HTML(str(nav))
 
 
@@ -101,14 +106,14 @@ def sidebar_html(menu: dict[str, Node], current: str) -> HTML:
     with nav:
         for slug, child in sorted_nodes(node.children):
             if child.published:
-                _nav_link(nav, child, f"{section}/{slug}", current)
+                _nav_link(nav, menu, child, f"{section}/{slug}", current)
     return HTML(str(nav))
 
 
 def first_leaf(menu: dict[str, Node], path: str) -> str | None:
     """First published descendant page (content set) in menu order.
 
-    This is the redirect target for content-less category labels.
+    This is the nav-link target for content-less category labels.
     """
     chain = resolve(menu, path)
     if chain is None:
@@ -208,6 +213,37 @@ def render_page(menu: dict[str, Node], path: str, brand: str = SITE_NAME) -> str
     )
 
 
+def render_category(menu: dict[str, Node], path: str, brand: str = SITE_NAME) -> str:
+    """Render the placeholder for a content-less category label (404).
+
+    The node exists in the tree but has no page of its own. Nav links
+    point straight at its first child, so this is mainly seen in the site
+    editor, where the pen creates the landing page.
+    """
+    node = resolve(menu, path)[-1]
+    title = _title(path.rpartition("/")[2], node)
+    doc = E.article
+    with doc:
+        doc.h1(title)
+        # Editing works here too: the pen creates this category's page.
+        doc.button("🖊️", **_edit_attrs(path))
+        doc.p(
+            "Pages in this section are listed in the menu on the left."
+        )
+    scripts, styles = _page_assets()
+    return str(
+        _layout(styles, scripts)(
+            Title=f"{title} – {brand}" if brand else title,
+            Brand=_brand_link(brand),
+            Nav=nav_html(menu, path),
+            Sidebar=sidebar_html(menu, path),
+            Banner=banner_html(menu, path),
+            BannerEdit=HTML(str(E.button("🖊️", **_edit_attrs(path, "site")))),
+            Main=HTML(str(doc)),
+        ),
+    )
+
+
 def render_not_found(menu: dict[str, Node], path: str, brand: str = SITE_NAME) -> str:
     """Render a 404 page within the normal layout."""
     doc = E.article