secure-redact 1.0.0 → 1.0.2

package/dist/assets/nlp.worker-u7Lr_A3c.js

@@ -1,521 +0,0 @@
-const f = /* @__PURE__ */ new Set([
-  "aarav",
-  "aditi",
-  "aditya",
-  "akash",
-  "amit",
-  "amita",
-  "ananya",
-  "anil",
-  "anita",
-  "anjali",
-  "ankita",
-  "arjun",
-  "arun",
-  "aruna",
-  "ashok",
-  "bhavna",
-  "chandra",
-  "deepak",
-  "deepika",
-  "dev",
-  "devika",
-  "dhruv",
-  "dinesh",
-  "divya",
-  "ganesh",
-  "gaurav",
-  "geeta",
-  "hari",
-  "harish",
-  "indira",
-  "isha",
-  "jagdish",
-  "kamala",
-  "karan",
-  "kavita",
-  "kishore",
-  "krishna",
-  "kumar",
-  "lakshmi",
-  "mahesh",
-  "manish",
-  "meera",
-  "mohan",
-  "mohit",
-  "nandini",
-  "naresh",
-  "neha",
-  "nikhil",
-  "nisha",
-  "pankaj",
-  "pooja",
-  "prakash",
-  "priya",
-  "rahul",
-  "rajesh",
-  "rajiv",
-  "raman",
-  "ramesh",
-  "rani",
-  "ravi",
-  "rekha",
-  "rohit",
-  "sachin",
-  "sandeep",
-  "sanjay",
-  "sapna",
-  "saroj",
-  "seema",
-  "shanti",
-  "sharma",
-  "shivani",
-  "shobha",
-  "shreya",
-  "sita",
-  "sneha",
-  "sunil",
-  "sunita",
-  "suresh",
-  "swati",
-  "tanvi",
-  "usha",
-  "varun",
-  "vijay",
-  "vikram",
-  "vinod",
-  "vishal",
-  "vivek",
-  "yash",
-  "yogesh",
-  "john",
-  "james",
-  "robert",
-  "michael",
-  "william",
-  "david",
-  "richard",
-  "joseph",
-  "thomas",
-  "charles",
-  "mary",
-  "patricia",
-  "jennifer",
-  "linda",
-  "elizabeth",
-  "barbara",
-  "susan",
-  "jessica",
-  "sarah",
-  "karen",
-  "mohammed",
-  "ahmed",
-  "ali",
-  "hassan",
-  "hussein",
-  "omar",
-  "fatima",
-  "aisha",
-  "zainab",
-  "khadija"
-]), m = /* @__PURE__ */ new Set([
-  "sharma",
-  "verma",
-  "gupta",
-  "singh",
-  "kumar",
-  "patel",
-  "joshi",
-  "mishra",
-  "agarwal",
-  "mehta",
-  "reddy",
-  "rao",
-  "nair",
-  "menon",
-  "pillai",
-  "iyer",
-  "iyengar",
-  "mukherjee",
-  "chatterjee",
-  "banerjee",
-  "das",
-  "bose",
-  "sen",
-  "ghosh",
-  "roy",
-  "dutta",
-  "sinha",
-  "jain",
-  "shah",
-  "desai",
-  "kulkarni",
-  "patil",
-  "deshpande",
-  "kaur",
-  "gill",
-  "bajwa",
-  "chopra",
-  "kapoor",
-  "malhotra",
-  "khanna",
-  "saxena",
-  "pandey",
-  "tiwari",
-  "dubey",
-  "trivedi",
-  "dwivedi",
-  "shukla",
-  "chauhan",
-  "yadav",
-  "thakur",
-  "smith",
-  "johnson",
-  "williams",
-  "brown",
-  "jones",
-  "davis",
-  "miller",
-  "wilson",
-  "moore",
-  "taylor"
-]), j = /* @__PURE__ */ new Set([
-  "andhra pradesh",
-  "arunachal pradesh",
-  "assam",
-  "bihar",
-  "chhattisgarh",
-  "goa",
-  "gujarat",
-  "haryana",
-  "himachal pradesh",
-  "jharkhand",
-  "karnataka",
-  "kerala",
-  "madhya pradesh",
-  "maharashtra",
-  "manipur",
-  "meghalaya",
-  "mizoram",
-  "nagaland",
-  "odisha",
-  "punjab",
-  "rajasthan",
-  "sikkim",
-  "tamil nadu",
-  "telangana",
-  "tripura",
-  "uttar pradesh",
-  "uttarakhand",
-  "west bengal",
-  "delhi",
-  "chandigarh",
-  "puducherry",
-  "jammu and kashmir",
-  "ladakh"
-]), u = /* @__PURE__ */ new Set([
-  "diabetes",
-  "hypertension",
-  "asthma",
-  "cancer",
-  "HIV",
-  "AIDS",
-  "tuberculosis",
-  "TB",
-  "hepatitis",
-  "malaria",
-  "dengue",
-  "cholesterol",
-  "thyroid",
-  "arthritis",
-  "epilepsy",
-  "pneumonia",
-  "bronchitis",
-  "anemia",
-  "leukemia",
-  "lymphoma",
-  "insulin",
-  "metformin",
-  "blood pressure",
-  "heart disease",
-  "kidney disease",
-  "liver disease",
-  "lung disease",
-  "chemotherapy",
-  "radiation",
-  "surgery",
-  "biopsy",
-  "diagnosis",
-  "prognosis",
-  "prescription",
-  "medication",
-  "dosage",
-  "allergic",
-  "allergy",
-  "positive",
-  "negative",
-  "report",
-  "pathology",
-  "radiology",
-  "MRI",
-  "CT scan",
-  "X-ray",
-  "ultrasound",
-  "ECG",
-  "EKG",
-  "patient",
-  "hospital",
-  "clinic",
-  "doctor",
-  "physician",
-  "surgeon"
-]), I = /* @__PURE__ */ new Set([
-  "road",
-  "rd",
-  "street",
-  "st",
-  "avenue",
-  "ave",
-  "lane",
-  "ln",
-  "nagar",
-  "colony",
-  "sector",
-  "block",
-  "plot",
-  "flat",
-  "floor",
-  "building",
-  "bldg",
-  "apartment",
-  "apt",
-  "house",
-  "no",
-  "near",
-  "opposite",
-  "opp",
-  "behind",
-  "beside",
-  "next to",
-  "main",
-  "cross",
-  "layout",
-  "extension",
-  "extn",
-  "phase",
-  "village",
-  "town",
-  "city",
-  "district",
-  "taluk",
-  "tehsil",
-  "post",
-  "pin",
-  "pincode",
-  "zip"
-]);
-function p() {
-  return "nlp_" + Math.random().toString(36).substring(2, 11);
-}
-function M(a) {
-  const n = [];
-  for (let e = 0; e < a.length; e++) {
-    const o = a[e].text.replace(/[^a-zA-Z]/g, "").toLowerCase();
-    if (o.length < 2) continue;
-    const t = f.has(o), i = m.has(o);
-    if (t || i) {
-      let l = a[e].text, s = { ...a[e].bbox }, r = 0.65;
-      if (e + 1 < a.length) {
-        const h = a[e + 1].text.replace(/[^a-zA-Z]/g, "").toLowerCase();
-        (f.has(h) || m.has(h)) && (l += " " + a[e + 1].text, s.w = a[e + 1].bbox.x + a[e + 1].bbox.w - s.x, r = 0.82, e++);
-      }
-      if (e + 1 < a.length) {
-        const h = a[e + 1].text.replace(/[^a-zA-Z]/g, "").toLowerCase();
-        m.has(h) && (l += " " + a[e + 1].text, s.w = a[e + 1].bbox.x + a[e + 1].bbox.w - s.x, r = 0.88, e++);
-      }
-      a[e - (l.split(" ").length - 1)]?.text[0]?.match(/[A-Z]/) && (r += 0.05), n.push({
-        id: p(),
-        type: "NAME",
-        value: l,
-        confidence: Math.min(r, 0.95),
-        bbox: s,
-        masked: !0,
-        layer: 2
-      });
-    }
-  }
-  return n;
-}
-function A(a) {
-  const n = [], e = a.map((i) => i.text).join(" ").toLowerCase(), o = /\b\d{6}\b/g;
-  let t;
-  for (; (t = o.exec(e)) !== null; ) {
-    const i = parseInt(t[0]);
-    if (i >= 110001 && i <= 855117) {
-      const l = k(a, t.index, e);
-      if (l >= 0) {
-        const s = Math.max(0, l - 8), r = a.slice(s, l + 1), h = r.some(
-          (x) => I.has(x.text.toLowerCase().replace(/[^a-z]/g, ""))
-        );
-        if (h || r.length >= 3) {
-          const x = r[0], d = r[r.length - 1];
-          n.push({
-            id: p(),
-            type: "ADDRESS",
-            value: r.map((c) => c.text).join(" "),
-            confidence: h ? 0.78 : 0.55,
-            bbox: {
-              x: x.bbox.x,
-              y: Math.min(...r.map((c) => c.bbox.y)),
-              w: d.bbox.x + d.bbox.w - x.bbox.x,
-              h: Math.max(...r.map((c) => c.bbox.y + c.bbox.h)) - Math.min(...r.map((c) => c.bbox.y)),
-              pageIndex: x.bbox.pageIndex
-            },
-            masked: !0,
-            layer: 2
-          });
-        }
-      }
-    }
-  }
-  for (const i of j) {
-    const l = i.split(" ");
-    for (let s = 0; s <= a.length - l.length; s++)
-      if (a.slice(s, s + l.length).map((h) => h.text.toLowerCase().replace(/[^a-z ]/g, "")).join(" ") === i) {
-        const h = a[s], x = a[s + l.length - 1];
-        n.push({
-          id: p(),
-          type: "ADDRESS",
-          value: a.slice(s, s + l.length).map((d) => d.text).join(" "),
-          confidence: 0.72,
-          bbox: {
-            x: h.bbox.x,
-            y: h.bbox.y,
-            w: x.bbox.x + x.bbox.w - h.bbox.x,
-            h: Math.max(h.bbox.h, x.bbox.h),
-            pageIndex: h.bbox.pageIndex
-          },
-          masked: !0,
-          layer: 2
-        });
-      }
-  }
-  return n;
-}
-function S(a) {
-  const n = [];
-  for (let e = 0; e < a.length; e++) {
-    const o = a[e].text.toLowerCase().replace(/[^a-z]/g, "");
-    if ((u.has(o) || u.has(a[e].text)) && n.push({
-      id: p(),
-      type: "MEDICAL",
-      value: a[e].text,
-      confidence: 0.75,
-      bbox: { ...a[e].bbox },
-      masked: !0,
-      layer: 2
-    }), e + 1 < a.length) {
-      const t = a[e].text + " " + a[e + 1].text;
-      u.has(t.toLowerCase()) && (n.push({
-        id: p(),
-        type: "MEDICAL",
-        value: t,
-        confidence: 0.8,
-        bbox: {
-          x: a[e].bbox.x,
-          y: a[e].bbox.y,
-          w: a[e + 1].bbox.x + a[e + 1].bbox.w - a[e].bbox.x,
-          h: Math.max(a[e].bbox.h, a[e + 1].bbox.h),
-          pageIndex: a[e].bbox.pageIndex
-        },
-        masked: !0,
-        layer: 2
-      }), e++);
-    }
-  }
-  return n;
-}
-function E(a) {
-  const n = [], e = /^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$/;
-  for (const o of a)
-    e.test(o.text) && n.push({
-      id: p(),
-      type: "EMAIL",
-      value: o.text,
-      confidence: 0.95,
-      bbox: { ...o.bbox },
-      masked: !0,
-      layer: 2
-    });
-  return n;
-}
-function L(a) {
-  const n = [], e = a.map((i) => i.text).join(" "), o = /\b(\d{1,2})[\/\-.](\d{1,2})[\/\-.](\d{4})\b/g;
-  let t;
-  for (; (t = o.exec(e)) !== null; ) {
-    const i = parseInt(t[1]), l = parseInt(t[2]), s = parseInt(t[3]);
-    if (i >= 1 && i <= 31 && l >= 1 && l <= 12 && s >= 1920 && s <= 2010) {
-      const r = Math.max(0, t.index - 30), h = e.substring(r, t.index).toLowerCase(), x = /\b(dob|date of birth|birth date|born|birthday|d\.o\.b)\b/.test(h), d = k(a, t.index, e);
-      if (d >= 0) {
-        const c = [];
-        let g = t.index;
-        for (let b = d; b < a.length && g < t.index + t[0].length; b++)
-          c.push(a[b]), g += a[b].text.length + 1;
-        if (c.length > 0) {
-          const b = c[0], y = c[c.length - 1];
-          n.push({
-            id: p(),
-            type: "DOB",
-            value: t[0],
-            confidence: x ? 0.9 : 0.6,
-            bbox: {
-              x: b.bbox.x,
-              y: b.bbox.y,
-              w: y.bbox.x + y.bbox.w - b.bbox.x,
-              h: Math.max(...c.map((v) => v.bbox.h)),
-              pageIndex: b.bbox.pageIndex
-            },
-            masked: !0,
-            layer: 2
-          });
-        }
-      }
-    }
-  }
-  return n;
-}
-function k(a, n, e) {
-  let o = 0;
-  for (let t = 0; t < a.length; t++) {
-    const i = e.indexOf(a[t].text, o);
-    if (i <= n && n < i + a[t].text.length)
-      return t;
-    o = i + a[t].text.length;
-  }
-  return -1;
-}
-self.onmessage = (a) => {
-  const { type: n, words: e, pageIndex: o } = a.data;
-  if (n === "NLP_ANALYZE")
-    try {
-      const t = [
-        ...M(e),
-        ...A(e),
-        ...S(e),
-        ...E(e),
-        ...L(e)
-      ];
-      for (const i of t)
-        i.bbox.pageIndex = o ?? 0;
-      self.postMessage({
-        type: "NLP_RESULT",
-        entities: t
-      });
-    } catch (t) {
-      self.postMessage({
-        type: "NLP_ERROR",
-        error: t instanceof Error ? t.message : "NLP analysis failed"
-      });
-    }
-};
-//# sourceMappingURL=nlp.worker-u7Lr_A3c.js.map

package/dist/assets/ocr.worker-D5s6dY7M.js

@@ -1,139 +0,0 @@
-function d(t) {
-  for (let e = 0; e < t.length; e += 4) {
-    const r = Math.round(
-      0.299 * t[e] + // R
-      0.587 * t[e + 1] + // G
-      0.114 * t[e + 2]
-      // B
-    );
-    t[e] = r, t[e + 1] = r, t[e + 2] = r;
-  }
-}
-function y(t) {
-  let e = 255, r = 0;
-  for (let s = 0; s < t.length; s += 4) {
-    const o = t[s];
-    o < e && (e = o), o > r && (r = o);
-  }
-  const n = r - e;
-  if (n !== 0)
-    for (let s = 0; s < t.length; s += 4) {
-      const o = Math.round((t[s] - e) / n * 255);
-      t[s] = o, t[s + 1] = o, t[s + 2] = o;
-    }
-}
-function R(t) {
-  const e = new Array(256).fill(0), r = t.length / 4;
-  for (let i = 0; i < t.length; i += 4)
-    e[t[i]]++;
-  let n = 0;
-  for (let i = 0; i < 256; i++)
-    n += i * e[i];
-  let s = 0, o = 0, l = 0, f = 0;
-  for (let i = 0; i < 256; i++) {
-    if (o += e[i], o === 0) continue;
-    const a = r - o;
-    if (a === 0) break;
-    s += i * e[i];
-    const g = s / o, u = (n - s) / a, p = o * a * (g - u) * (g - u);
-    p > l && (l = p, f = i);
-  }
-  return f;
-}
-function w(t, e) {
-  for (let r = 0; r < t.length; r += 4) {
-    const n = t[r] >= e ? 255 : 0;
-    t[r] = n, t[r + 1] = n, t[r + 2] = n;
-  }
-}
-function O(t, e, r) {
-  let n = 0;
-  const s = e * r;
-  for (let l = 0; l < t.length; l += 4)
-    (t[l] + t[l + 1] + t[l + 2]) / 3 < 240 && n++;
-  const o = n / s;
-  return o < 0.2 ? 11 : o > 0.5 ? 6 : o > 0.3 ? 4 : 3;
-}
-async function k(t) {
-  try {
-    const e = await createImageBitmap(t), r = new OffscreenCanvas(e.width, e.height), n = r.getContext("2d", { willReadFrequently: !0 });
-    if (!n)
-      return e.close(), { blob: t, psm: 3 };
-    n.drawImage(e, 0, 0);
-    const s = n.getImageData(0, 0, e.width, e.height), o = s.data;
-    d(o), y(o);
-    const l = O(o, e.width, e.height), f = R(o);
-    w(o, f), n.putImageData(s, 0, 0);
-    const i = await r.convertToBlob({ type: "image/png" });
-    return e.close(), { blob: i, psm: l };
-  } catch (e) {
-    return console.warn("[OCR Worker] Preprocessing failed, using original image:", e), { blob: t, psm: 3 };
-  }
-}
-let h = null;
-async function C() {
-  return h || (h = await (await import("./index-C62fEJ4q.js").then(function(e) {
-    return e.i;
-  })).createWorker("eng", void 0, {
-    logger: (e) => {
-      self.postMessage({
-        type: "OCR_PROGRESS",
-        progress: e.progress,
-        message: e.status
-      });
-    }
-  }), h);
-}
-self.onmessage = async (t) => {
-  const { type: e, fileBuffer: r, fileType: n, pageIndex: s } = t.data;
-  if (e === "OCR_START")
-    try {
-      const o = await C();
-      let l;
-      n === "application/pdf" ? l = new Blob([r], { type: "image/png" }) : l = new Blob([r], { type: n }), self.postMessage({
-        type: "OCR_PROGRESS",
-        progress: 0.1,
-        message: "Preprocessing image..."
-      });
-      const { blob: f, psm: i } = await k(l);
-      console.log(`[OCR Worker] Detected optimal PSM: ${i}`), await o.setParameters({
-        tessedit_pageseg_mode: i.toString()
-      });
-      const a = await o.recognize(f, {}, { text: !0, blocks: !0 }), g = [], u = (c) => {
-        !c || !c.text || !c.bbox || g.push({
-          text: c.text,
-          confidence: (c.confidence ?? 0) / 100,
-          bbox: {
-            x: c.bbox.x0,
-            y: c.bbox.y0,
-            w: c.bbox.x1 - c.bbox.x0,
-            h: c.bbox.y1 - c.bbox.y0,
-            pageIndex: s ?? 0
-          }
-        });
-      };
-      if (a.data.blocks && a.data.blocks.length > 0)
-        for (const c of a.data.blocks)
-          for (const m of c.paragraphs ?? [])
-            for (const x of m.lines ?? [])
-              for (const b of x.words ?? [])
-                u(b);
-      else if (a.data.words && a.data.words.length > 0)
-        for (const c of a.data.words)
-          u(c);
-      console.log(`[OCR Worker] Extracted ${g.length} words from Tesseract`);
-      let p = a.data.text ?? "";
-      !p.trim() && g.length > 0 && (p = g.map((c) => c.text).join(" "), console.log("[OCR Worker] Reconstructed fullText from words")), console.log(`[OCR Worker] fullText length: ${p.length}`), self.postMessage({
-        type: "OCR_RESULT",
-        words: g,
-        fullText: p,
-        pageIndex: s ?? 0
-      });
-    } catch (o) {
-      self.postMessage({
-        type: "OCR_ERROR",
-        error: o instanceof Error ? o.message : "OCR processing failed"
-      });
-    }
-};
-//# sourceMappingURL=ocr.worker-D5s6dY7M.js.map