docs: v0.1.129 sizes from the real artifacts; fix IndexNow step

.github/workflows/indexnow.yml

@@ -45,13 +45,8 @@ jobs:
 
       - name: Note on Google (its ping endpoint no longer exists)
         run: |
-          # Google deprecated this in 2023 — they recommend Search Console.
-          # We try anyway because it's a free no-op if it fails.
-          # Google retired the sitemap ping endpoint in June 2023; it now
-          # returns 404, so calling it was doing nothing. Google discovers
-          # the sitemap from robots.txt and re-crawls on its own schedule —
-          # the only manual lever is Search Console > Sitemaps / URL
-          # Inspection, which needs an authenticated human. IndexNow above
-          # still covers Bing, Yandex, Seznam and Naver.
-            && echo "✓ Google sitemap ping accepted" \
-            || echo "::warning::Google sitemap ping failed (expected since 2023; use Search Console)"
+          # Google retired the sitemap ping endpoint in June 2023 (it now
+          # returns 404). Google discovers the sitemap from robots.txt; the
+          # only manual lever is Search Console. IndexNow above still covers
+          # Bing, Yandex, Seznam and Naver.
+          echo "Google: no ping endpoint — use Search Console > Sitemaps"

docs/docs.html

@@ -30,12 +30,12 @@
 <meta property="og:image:width" content="1200">
 <meta property="og:image:height" content="630">
 <meta property="og:image:type" content="image/png">
-<meta property="og:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~12 MB native, zero Electron, 49-tool MCP server.">
+<meta property="og:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~13 MB native, zero Electron, 49-tool MCP server.">
 <meta name="twitter:card" content="summary_large_image">
 <meta name="twitter:title" content="Notepatra Documentation — Install, Features, Plugins, AI, MCP">
 <meta name="twitter:description" content="Complete Notepatra documentation: installation, the Password Generator and other plugins, AI integration, the MCP server, keyboard shortcuts, compare/diff, configuration, building from source, verifying releases, troubleshooting.">
 <meta name="twitter:image" content="https://notepatra.org/og-card.png">
-<meta name="twitter:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~12 MB native, zero Electron, 49-tool MCP server.">
+<meta name="twitter:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~13 MB native, zero Electron, 49-tool MCP server.">
 
 <meta http-equiv="Content-Security-Policy" content="default-src 'self'; img-src 'self' data: https://raw.githubusercontent.com; style-src 'self' 'unsafe-inline'; script-src 'self' 'unsafe-inline'; font-src 'self' data:; connect-src 'self' https://api.github.com; frame-ancestors 'none'; base-uri 'self'; form-action 'self'; upgrade-insecure-requests">
 <meta http-equiv="X-Content-Type-Options" content="nosniff">
@@ -429,7 +429,7 @@ <h1>Notepatra Documentation</h1>
   <!-- ══════════════ Getting started ══════════════ -->
 
   <h2 id="introduction" class="anchor-target">Introduction</h2>
-  <p>Notepatra is a <strong>native</strong> code editor written in C++17 with a Rust core for the heavy-lifting (memory-mapped file I/O, Aho-Corasick search, Myers diff, formatters). It targets <strong>Linux x64 / ARM64, macOS Apple Silicon, and Windows x64</strong>. The bare executable is ~12 MB (12.6 MB on Linux x64) on each platform; downloads range 4.1 MB (Linux ARM64) to ~43.0 MB (Windows MSI with Qt DLLs bundled).</p>
+  <p>Notepatra is a <strong>native</strong> code editor written in C++17 with a Rust core for the heavy-lifting (memory-mapped file I/O, Aho-Corasick search, Myers diff, formatters). It targets <strong>Linux x64 / ARM64, macOS Apple Silicon, and Windows x64</strong>. The bare executable is ~13 MB (13.4 MB on Linux x64) on each platform; downloads range 4.4 MB (Linux ARM64) to ~43.9 MB (Windows MSI with Qt DLLs bundled).</p>
   <p>Unlike Electron-based editors, Notepatra does not ship a browser runtime. It uses Qt5 + QScintilla — a battle-tested Scintilla-based editor engine — so 238 file types, 82 language lexers, brace matching, folding, and code completion are built in.</p>
   <p>Unlike cloud-first editors, Notepatra's AI is <strong>local-first</strong>. The AI dock dropdown ships <strong>six backends</strong>: <a href="https://ollama.com">Ollama</a> (default · <code>localhost:11434</code>), <a href="https://github.com/ggml-org/llama.cpp">llama.cpp</a>'s llama-server (loads any GGUF directly · <code>localhost:8080</code>), <a href="https://openrouter.ai">OpenRouter</a> (cloud · 100+ models — Claude, GPT, Gemini, Grok, Kimi, …), <a href="https://ollama.com/cloud">Ollama Cloud</a> (cloud · gpt-oss:120b, qwen3-coder:480b, deepseek-v3.1:671b), <a href="https://openai.com">OpenAI</a> direct, and <a href="https://azure.microsoft.com/en-us/products/ai-services/openai-service">Azure OpenAI</a> (enterprise). The <code>llama.cpp</code> entry also accepts a user-configured base URL via Settings → Preferences → AI, so it can reach any OpenAI-compatible HTTP server you've installed. Nothing leaves your computer unless you pick a cloud backend. No mandatory API keys, no subscription, no telemetry.</p>
   <p><strong>The AI runs as an agent.</strong> Tick <strong>Coding Mode</strong> in the AI dock (with a workspace open and a tool-capable model) and the model gets up to 13 native tools — file tools <code>read_file</code>, <code>list_dir</code>, <code>write_file</code>, <code>search</code>, <code>apply_diff</code>; data tools <code>query_sql</code>, <code>csv_query</code>, <code>generate_chart</code>; and read-only git tools <code>git_status</code>, <code>git_diff</code>, <code>git_log</code>, <code>git_branch_list</code>, <code>git_show</code> — plus three-layer path safety, a 25-call hard cap per turn, and persistent per-workspace chat history. See <a href="#ai-coding-mode">Coding Mode (agentic)</a> below for the full surface.</p>
@@ -459,12 +459,12 @@ <h3>Manual download</h3>
   <table>
     <thead><tr><th>Platform</th><th>Asset</th><th>Size</th></tr></thead>
     <tbody>
-      <tr><td>🐧 Linux x64</td><td><code>notepatra-linux-x64.tar.gz</code></td><td>~4.4 MB</td></tr>
-      <tr><td>🐧 Linux ARM64</td><td><code>notepatra-linux-arm64.tar.gz</code></td><td>~4.1 MB</td></tr>
-      <tr><td>🍎 macOS Apple Silicon</td><td><code>notepatra-macos-arm64.dmg</code></td><td>~27.7 MB</td></tr>
-      <tr><td>🪟 Windows x64 (MSI)</td><td><code>notepatra-&lt;version&gt;.msi</code></td><td>~43.0 MB</td></tr>
-      <tr><td>🪟 Windows x64 (installer)</td><td><code>notepatra-setup-&lt;version&gt;.exe</code></td><td>~32.8 MB</td></tr>
-      <tr><td>🪟 Windows x64 (portable)</td><td><code>notepatra-windows-x64.zip</code></td><td>~37.4 MB</td></tr>
+      <tr><td>🐧 Linux x64</td><td><code>notepatra-linux-x64.tar.gz</code></td><td>~4.7 MB</td></tr>
+      <tr><td>🐧 Linux ARM64</td><td><code>notepatra-linux-arm64.tar.gz</code></td><td>~4.4 MB</td></tr>
+      <tr><td>🍎 macOS Apple Silicon</td><td><code>notepatra-macos-arm64.dmg</code></td><td>~28.0 MB</td></tr>
+      <tr><td>🪟 Windows x64 (MSI)</td><td><code>notepatra-&lt;version&gt;.msi</code></td><td>~43.9 MB</td></tr>
+      <tr><td>🪟 Windows x64 (installer)</td><td><code>notepatra-setup-&lt;version&gt;.exe</code></td><td>~33.0 MB</td></tr>
+      <tr><td>🪟 Windows x64 (portable)</td><td><code>notepatra-windows-x64.zip</code></td><td>~37.8 MB</td></tr>
     </tbody>
   </table>
 
@@ -618,7 +618,7 @@ <h2 id="first-launch" class="anchor-target">First launch</h2>
   <!-- ══════════════ Lite vs Full + packs (v0.1.64+) ══════════════ -->
 
   <h2 id="lite-vs-full" class="anchor-target">Lite mode vs Full mode (v0.1.64+)</h2>
-  <p>Starting in v0.1.64, Notepatra ships two build flavors per release. <strong>Lite</strong> is the default download — a bare ~12.6 MB binary with no heavy dependencies bundled. <strong>Full</strong> is the opt-in flavor. On every platform it bundles the <strong>DuckDB</strong> query engine (v1.1.3) next to the binary; on <strong>Linux and Windows</strong> it additionally bundles QtWebEngine for inline Vega-Lite chart rendering. <strong>macOS Full is DuckDB-only</strong> — Homebrew's <code>qt@5</code> no longer ships QtWebEngine (and there is no Apple-Silicon Qt5 WebEngine), so inline Vega charts stay a Linux/Windows Full feature; the native fenced <code>```chart</code> renderer (QtCharts) still works everywhere. On Linux the <code>-full</code> tarball links <code>libQt5WebEngineWidgets</code> from your distro (install <code>qtwebengine5-dev</code> / <code>qt5-qtwebengine</code>) and ships <code>libduckdb.so</code> alongside the binary. Heavy features go through <strong>packs</strong> that you install on demand, the way DuckDB ships a tiny core plus optional extensions (<code>httpfs</code>, <code>parquet</code>, <code>spatial</code>, etc.).</p>
+  <p>Starting in v0.1.64, Notepatra ships two build flavors per release. <strong>Lite</strong> is the default download — a bare ~13.4 MB binary with no heavy dependencies bundled. <strong>Full</strong> is the opt-in flavor. On every platform it bundles the <strong>DuckDB</strong> query engine (v1.1.3) next to the binary; on <strong>Linux and Windows</strong> it additionally bundles QtWebEngine for inline Vega-Lite chart rendering. <strong>macOS Full is DuckDB-only</strong> — Homebrew's <code>qt@5</code> no longer ships QtWebEngine (and there is no Apple-Silicon Qt5 WebEngine), so inline Vega charts stay a Linux/Windows Full feature; the native fenced <code>```chart</code> renderer (QtCharts) still works everywhere. On Linux the <code>-full</code> tarball links <code>libQt5WebEngineWidgets</code> from your distro (install <code>qtwebengine5-dev</code> / <code>qt5-qtwebengine</code>) and ships <code>libduckdb.so</code> alongside the binary. Heavy features go through <strong>packs</strong> that you install on demand, the way DuckDB ships a tiny core plus optional extensions (<code>httpfs</code>, <code>parquet</code>, <code>spatial</code>, etc.).</p>
 
   <h3>Why this split?</h3>
   <p>v0.1.63 bundled QtWebEngine into every release for one specific feature — the AI Data Analyst's inline chart card. Most users will never trigger a chart, but every install paid the disk + download cost. v0.1.64 flipped the default: lite by default, charts move to an optional pack. Same architecture mirrors how Cursor / VS Code / Sublime add language servers and renderers — small core, install what you need.</p>
@@ -630,7 +630,7 @@ <h3>Why this split?</h3>
     <tbody>
       <tr>
         <td><strong>Lite</strong> (default)</td>
-        <td>~12 MB</td>
+        <td>~13 MB</td>
         <td>Everything: editor, lexers, AI Assistant, Coding Mode, Data Analyst tool calls, Git, REST client, hex editor, plugins, themes.</td>
         <td>Inline Vega-Lite chart rendering. The Data Analyst can still generate chart specs — they show as a "Chart rendering requires the Charts Pack" card with [View JSON instead] to copy the spec out.</td>
       </tr>
@@ -2156,10 +2156,10 @@ <h4>Is Notepatra a port of another editor?</h4>
   <p>No. Notepatra is its own application — a native Qt5 + QScintilla editor with an original C++17/Rust core for the hot paths (memory-mapped file I/O, Aho-Corasick search, Myers diff, the formatters). The editing surface is QScintilla — the mature Scintilla engine, the same one Notepad++ is built on — so folding, brace matching and lexing are battle-tested rather than reinvented; everything around it (the AI dock, the tools, the Rust core, the UI) is Notepatra's own. It builds from a single codebase for Linux x64 / ARM64, macOS Apple Silicon, and Windows x64.</p>
 
   <h4>What's the difference between the Lite and Full builds, and which do I get?</h4>
-  <p>The default download is <strong>Lite</strong>: the complete editor — all 82 lexers, the AI dock, diagrams, formatters, Git, terminal — with no browser engine bundled, which is why the bare binary stays ~12 MB. <strong>Full</strong> bundles the DuckDB engine on every platform, and — on <strong>Linux and Windows</strong> — adds QtWebEngine so the <code>generate_chart</code> AI tool can render interactive Vega-Lite charts inline. (<strong>macOS Full is DuckDB-only</strong>: Homebrew <code>qt@5</code> no longer ships QtWebEngine and there is no Apple-Silicon Qt5 WebEngine, so inline Vega charts are a Linux/Windows feature.) Everything else is identical. Most people want Lite; reach for Full only if you need the DuckDB engine or WebEngine-rendered Vega charts — the native QtCharts inline charts and the <code>.npd</code> diagram renderer work in <em>both</em> flavors on <em>every</em> platform. See <a href="#lite-vs-full">Lite mode vs Full mode</a>.</p>
+  <p>The default download is <strong>Lite</strong>: the complete editor — all 82 lexers, the AI dock, diagrams, formatters, Git, terminal — with no browser engine bundled, which is why the bare binary stays ~13 MB. <strong>Full</strong> bundles the DuckDB engine on every platform, and — on <strong>Linux and Windows</strong> — adds QtWebEngine so the <code>generate_chart</code> AI tool can render interactive Vega-Lite charts inline. (<strong>macOS Full is DuckDB-only</strong>: Homebrew <code>qt@5</code> no longer ships QtWebEngine and there is no Apple-Silicon Qt5 WebEngine, so inline Vega charts are a Linux/Windows feature.) Everything else is identical. Most people want Lite; reach for Full only if you need the DuckDB engine or WebEngine-rendered Vega charts — the native QtCharts inline charts and the <code>.npd</code> diagram renderer work in <em>both</em> flavors on <em>every</em> platform. See <a href="#lite-vs-full">Lite mode vs Full mode</a>.</p>
 
   <h4>How big is it, really?</h4>
-  <p>The bare executable is ~12 MB (~12.6 MB on Linux x64) on every platform — no Electron, no bundled browser runtime. Download sizes differ because Linux links Qt from your distro while macOS and Windows bundle it: roughly 4.4 MB (Linux x64 tar.gz), 27.7 MB (macOS DMG with Qt), 43.0 MB (Windows MSI with the Qt DLLs). The GitHub release page is always canonical for exact bytes.</p>
+  <p>The bare executable is ~13 MB (~13.4 MB on Linux x64) on every platform — no Electron, no bundled browser runtime. Download sizes differ because Linux links Qt from your distro while macOS and Windows bundle it: roughly 4.7 MB (Linux x64 tar.gz), 28.0 MB (macOS DMG with Qt), 43.9 MB (Windows MSI with the Qt DLLs). The GitHub release page is always canonical for exact bytes.</p>
 
   <h4>Does my code go to the cloud? Is there any telemetry?</h4>
   <p>No telemetry, no analytics, no account — the editor starts fully offline. Every outbound connection is one you trigger: (1) your selected AI backend — a <em>local</em> Ollama (<code>localhost:11434</code>) or llama-server (<code>localhost:8080</code>) by default, or whichever URL you set on the llama.cpp entry; (2) <code>git push/pull</code> from the Git panel; (3) the REST client when you send a request; (4) one GitHub-API call at launch if the update check is enabled. Cloud AI backends (Ollama Cloud / OpenRouter / OpenAI / Azure OpenAI) are <strong>opt-in</strong> from the dropdown and use <em>your</em> key — your code leaves the machine only when you pick one and send it. All of this is verifiable with <code>strace</code>, and the built-in AI Interaction Log records exactly what was sent so you never have to take our word for it.</p>

docs/enterprise.html

@@ -29,12 +29,12 @@
     <meta property="og:image:width" content="1200">
     <meta property="og:image:height" content="630">
     <meta property="og:image:type" content="image/png">
-    <meta property="og:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~12 MB native, zero Electron, 49-tool MCP server.">
+    <meta property="og:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~13 MB native, zero Electron, 49-tool MCP server.">
     <meta name="twitter:card" content="summary_large_image">
     <meta name="twitter:title" content="Notepatra for Enterprise — License, Supply Chain, IT/Legal Review">
     <meta name="twitter:description" content="Plain-English explainer for IT, legal, and procurement teams reviewing Notepatra. GPL-3.0 standalone app, no telemetry, no mandatory network calls, signed releases.">
     <meta name="twitter:image" content="https://notepatra.org/og-card.png">
-    <meta name="twitter:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~12 MB native, zero Electron, 49-tool MCP server.">
+    <meta name="twitter:image:alt" content="Notepatra — the code editor built for the AI era. C++ and Rust, ~13 MB native, zero Electron, 49-tool MCP server.">
     <style>
         * { margin: 0; padding: 0; box-sizing: border-box; }
 
@@ -295,7 +295,7 @@ <h2>Snapshot for your software-acceptance form</h2>
                 <tr><td>Supply chain</td><td>Each release page on GitHub publishes SHA-256 checksums, cosign signatures (recorded in the Sigstore Rekor transparency log), and SLSA build-provenance attestations. The releases are built from public source on GitHub Actions; bit-for-bit reproducible builds are planned but not yet implemented.</td></tr>
                 <tr><td>Code signing</td><td>SHA-256 + cosign on every release artifact at the time of writing. Operating-system code signing (Authenticode on Windows, Apple notarization on macOS) is on the roadmap; current builds are unsigned at the OS level but are cryptographically verifiable via cosign.</td></tr>
                 <tr><td>Platforms</td><td>Linux x64 / ARM64, macOS Apple Silicon, Windows x64</td></tr>
-                <tr><td>Footprint</td><td>Bare executable under 12 MB (~11.5 MB on Linux x64). Latest Lite (default) downloads: 4.1 MB (Linux x64 tar.gz), 27.5 MB (macOS DMG), 42.4 MB (Windows MSI); the opt-in Full download is larger (bundles DuckDB, plus WebEngine on Linux/Windows). Exact sizes vary release-to-release; the GitHub release page is canonical.</td></tr>
+                <tr><td>Footprint</td><td>Bare executable under 12 MB (~11.5 MB on Linux x64). Latest Lite (default) downloads: 4.4 MB (Linux x64 tar.gz), 27.5 MB (macOS DMG), 42.4 MB (Windows MSI); the opt-in Full download is larger (bundles DuckDB, plus WebEngine on Linux/Windows). Exact sizes vary release-to-release; the GitHub release page is canonical.</td></tr>
                 <tr><td>Updates</td><td>Manual. No auto-update daemon. No background service.</td></tr>
                 <tr><td>Plugins</td><td>Drop a <code>.so</code>/<code>.dylib</code>/<code>.dll</code> in <code>~/.config/notepatra/plugins/</code>. No remote plugin registry. No auto-install.</td></tr>
             </table>