MediaWiki:Gadget-LabelScanIndexer.js: Unterschied zwischen den Versionen

Version vom 9. November 2025, 16:26 Uhr

/* Gadget: LabelScanIndexer (Auto-Save)
 * Läuft nur auf der Seite "Hilfe:LabelScan-Indexer" (Namespace Help = 12)
 * Erzeugt Embeddings lokal (CLIP) und speichert in MediaWiki:Gadget-LabelScan-index.json
 */

/* global mw */
(function () {
  // ---------- Seitenerkennung (robust) ----------
  const NS = mw.config.get('wgNamespaceNumber'); // 12 = Help/Hilfe
  const TITLE = mw.config.get('wgTitle');        // nur der Titel ohne Namespace

  if (!(NS === 12 && TITLE === 'LabelScan-Indexer')) {
    // Debug-Hinweis, falls du auf der falschen Seite testest
    // console.debug('[LabelScanIndexer] nicht aktiv auf', NS, TITLE);
    return;
  }

  const INDEX_TITLE = 'MediaWiki:Gadget-LabelScan-index.json';

  // ---------- Modell / Pfade ----------
  const TRANSFORMERS_URL = 'https://cdn.jsdelivr.net/npm/@xenova/transformers@2.15.0';
  const MODEL_ID = 'Xenova/clip-vit-base-patch32';
  const LOCAL_MODEL_PATH = '/models';

  // ---------- UI helpers ----------
  const $ = (id) => document.getElementById(id);
  const status = (t) => { const el = $('idx-status'); if (el) el.textContent = t || ''; };

  function hasInterfaceRight() {
    const groups = mw.config.get('wgUserGroups') || [];
    return groups.includes('interface-admin') || groups.includes('sysop');
  }

  function float32ToBase64(vec) {
    const bytes = new Uint8Array(vec.buffer);
    let bin = '', chunk = 0x8000;
    for (let i = 0; i < bytes.length; i += chunk) {
      bin += String.fromCharCode.apply(null, bytes.subarray(i, i + chunk));
    }
    return btoa(bin);
  }

  // EXIF-korrekte Canvas-Erzeugung
  async function fileToCanvasExif(file) {
    if ('createImageBitmap' in window) {
      const bmp = await createImageBitmap(file, { imageOrientation: 'from-image' });
      if ('OffscreenCanvas' in window) {
        const c = new OffscreenCanvas(bmp.width, bmp.height);
        c.getContext('2d').drawImage(bmp, 0, 0);
        return c;
      } else {
        const c = document.createElement('canvas');
        c.width = bmp.width; c.height = bmp.height;
        c.getContext('2d').drawImage(bmp, 0, 0);
        return c;
      }
    }
    // Fallback klassisch
    const url = URL.createObjectURL(file);
    try {
      const img = await new Promise((res, rej) => {
        const im = new Image();
        im.onload = () => res(im);
        im.onerror = rej;
        im.src = url;
      });
      const c = document.createElement('canvas');
      c.width = img.width; c.height = img.height;
      c.getContext('2d').drawImage(img, 0, 0);
      return c;
    } finally {
      URL.revokeObjectURL(url);
    }
  }

  // ---------- Transformers laden (einmalig) ----------
  let _modelPromise;
  async function ensureModel() {
    if (_modelPromise) return _modelPromise;
    _modelPromise = (async () => {
      const mod = await import(/* webpackIgnore: true */ TRANSFORMERS_URL);

      mod.env.allowLocalModels = true;
      mod.env.allowRemoteModels = false;
      mod.env.localModelPath = LOCAL_MODEL_PATH;

      // Optional WebGPU bevorzugen:
      // mod.env.backends = mod.env.backends || {};
      // mod.env.backends.onnx = mod.env.backends.onnx || {};
      // mod.env.backends.onnx.preferredBackend = 'webgpu';

      // WASM-Runtime-Pfade
      mod.env.backends = mod.env.backends || {};
      mod.env.backends.onnx = mod.env.backends.onnx || {};
      mod.env.backends.onnx.wasm = mod.env.backends.onnx.wasm || {};
      mod.env.backends.onnx.wasm.wasmPaths =
        'https://cdn.jsdelivr.net/npm/@xenova/transformers@2.15.0/dist/';

      const [processor, model] = await Promise.all([
        mod.AutoProcessor.from_pretrained(MODEL_ID),
        mod.CLIPVisionModelWithProjection.from_pretrained(MODEL_ID, { quantized: true }),
      ]);

      console.log('[LabelScanIndexer] Modell geladen');
      return { mod, processor, model };
    })();
    return _modelPromise;
  }

  async function buildEmbeddingFromFile(file) {
    const { mod, processor, model } = await ensureModel();
    const canvas = await fileToCanvasExif(file);
    const blob = (canvas.convertToBlob)
      ? await canvas.convertToBlob({ type: 'image/jpeg', quality: 0.95 })
      : await new Promise((r) => canvas.toBlob(r, 'image/jpeg', 0.95));
    const raw = await mod.RawImage.fromBlob(blob);
    const inputs = await processor(raw, { return_tensors: 'pt' });
    const out = await model.forward({ pixel_values: inputs.pixel_values });
    const vec = out?.image_embeds?.data || out?.image_embeds;
    if (!(vec instanceof Float32Array)) throw new Error('Embedding-Format unerwartet');

    // Normieren
    let n = 0; for (let i = 0; i < vec.length; i++) n += vec[i] * vec[i];
    const norm = Math.sqrt(n) || 1;
    const v = new Float32Array(vec.length);
    for (let i = 0; i < vec.length; i++) v[i] = vec[i] / norm;
    return v;
  }

  // ---------- Index laden/speichern ----------
  async function fetchIndexJSON() {
    const url = mw.util.getUrl(INDEX_TITLE, { action: 'raw', ctype: 'application/json' });
    const res = await fetch(url, { cache: 'no-store' });
    if (!res.ok) throw new Error('Index nicht ladbar: ' + res.status);
    try { return JSON.parse(await res.text()) || []; }
    catch (_) { return []; }
  }

  async function saveIndexJSON(newArray, summary) {
    await mw.loader.using(['mediawiki.api']);
    const api = new mw.Api();
    const text = JSON.stringify(newArray, null, 2) + '\n';
    return api.postWithToken('csrf', {
      action: 'edit',
      title: INDEX_TITLE,
      text,
      summary: summary || 'LabelScan: +1 embedding (Auto-Indexer)',
      nocreate: 0,
      bot: 1
    });
  }

  // ---------- Click-Handler ----------
  const runBtn = document.getElementById('idx-run');
  if (!runBtn) {
    console.warn('[LabelScanIndexer] Button #idx-run nicht gefunden – ist das HTML auf der Seite eingebunden?');
    return;
  }

  runBtn.addEventListener('click', async () => {
    try {
      if (!hasInterfaceRight()) {
        alert('⚠️ Du brauchst Admin/Interface-Rechte (editinterface).');
        return;
      }

      const title = $('idx-title')?.value.trim();
      const thumb = $('idx-thumb')?.value.trim();
      const file = $('idx-file')?.files?.[0];

      if (!title) return alert('Titel fehlt.');
      if (!file) return alert('Bitte eine Bilddatei wählen.');

      runBtn.disabled = true;

      status('Embedding berechnen …');
      const vec = await buildEmbeddingFromFile(file);
      const b64 = float32ToBase64(vec);

      $('idx-out').value = JSON.stringify({ title, thumb, embed: b64 }, null, 2);

      status('Index laden …');
      const arr = await fetchIndexJSON();
      arr.push({ title, thumb, embed: b64 });

      status('Speichern …');
      await saveIndexJSON(arr, `LabelScan: +1 embedding für "${title}"`);

      status('Gespeichert ✅');
    } catch (e) {
      console.error(e);
      status('Fehler ❌ ' + (e?.message || e));
      alert('Fehler: ' + (e?.message || e));
    } finally {
      runBtn.disabled = false;
    }
  });

  console.log('[LabelScanIndexer] bereit');
})();