kordoc 4.6.1 → 4.7.0

package/dist/{chunk-VCA4SZ2D.js → chunk-P2NWNJMR.js}

@@ -1,7 +1,7 @@
 #!/usr/bin/env node
 import {
   extractPageBlocksWithLines
-} from "./chunk-BGSGUONP.js";
+} from "./chunk-PEVPKD6Z.js";
 import {
   OCR_DET_MODEL,
   OCR_REC_DICT,
@@ -513,4 +513,4 @@ export {
   runPdfOcr,
   ocrItemsToBlocks
 };
-//# sourceMappingURL=chunk-VCA4SZ2D.js.map
+//# sourceMappingURL=chunk-P2NWNJMR.js.map

package/dist/{chunk-BGSGUONP.js → chunk-PEVPKD6Z.js}

@@ -3043,6 +3043,123 @@ function buildGridTable(lines, columns) {
   return md.join("\n");
 }
 
+// src/pdf/underline.ts
+var UNDER_MAX_THICKNESS = 2;
+var UNDER_BELOW_EM = 0.72;
+var UNDER_BELOW_MIN_PT = 3;
+var UNDER_ABOVE_PT = 1;
+var UNDER_MIN_OVERLAP_RATIO = 0.6;
+var UNDER_OWNER_PAD_EM = 0.75;
+var UNDER_OWNER_PAD_MIN_PT = 4;
+var UNDER_MIN_COVERAGE = 0.6;
+var UNDER_GRID_EPS = 2;
+var UNDER_COLUMN_GAP_EM = 2;
+var UNDER_REPEATED_SPAN_LEVELS = 3;
+var UNDER_SPAN_OVERLAP_RATIO = 0.8;
+var UNDER_BOX_PAIR_MIN_EM = 0.5;
+var UNDER_BOX_PAIR_MAX_EM = 2.2;
+var UNDER_BOX_PAIR_OVERLAP = 0.8;
+function markUnderlineItems(items, horizontals, verticals) {
+  if (items.length === 0 || horizontals.length === 0) return;
+  for (const line of horizontals) {
+    if (line.lineWidth > UNDER_MAX_THICKNESS) continue;
+    if (touchesVertical(line, verticals)) continue;
+    if (isRepeatedSpanRule(line, horizontals)) continue;
+    const matches = [];
+    for (const item of items) {
+      const h = item.h > 0 ? item.h : item.fontSize;
+      if (h <= 0 || item.w <= 0 || !item.text.trim()) continue;
+      const below = Math.max(h * UNDER_BELOW_EM, UNDER_BELOW_MIN_PT);
+      if (line.y1 < item.y - below || line.y1 > item.y + UNDER_ABOVE_PT) continue;
+      const overlap = Math.min(line.x2, item.x + item.w) - Math.max(line.x1, item.x);
+      if (overlap / item.w < UNDER_MIN_OVERLAP_RATIO) continue;
+      matches.push(item);
+    }
+    if (matches.length === 0) continue;
+    let x1 = Infinity, x2 = -Infinity, maxH = 0, covered = 0;
+    for (const m of matches) {
+      x1 = Math.min(x1, m.x);
+      x2 = Math.max(x2, m.x + m.w);
+      maxH = Math.max(maxH, m.h > 0 ? m.h : m.fontSize);
+      covered += m.w;
+    }
+    const pad = Math.max(maxH * UNDER_OWNER_PAD_EM, UNDER_OWNER_PAD_MIN_PT);
+    if (line.x1 < x1 - pad || line.x2 > x2 + pad) continue;
+    if (covered < (line.x2 - line.x1) * UNDER_MIN_COVERAGE) continue;
+    if (hasBoxTopPair(line, maxH, horizontals)) continue;
+    matches.sort((a, b) => a.x - b.x);
+    let hole = false;
+    for (let i = 1; i < matches.length; i++) {
+      if (matches[i].x - (matches[i - 1].x + matches[i - 1].w) > maxH * UNDER_COLUMN_GAP_EM) {
+        hole = true;
+        break;
+      }
+    }
+    if (hole) continue;
+    for (const m of matches) m.underline = true;
+  }
+}
+function touchesVertical(line, verticals) {
+  for (const v of verticals) {
+    if (v.x1 < line.x1 - UNDER_GRID_EPS || v.x1 > line.x2 + UNDER_GRID_EPS) continue;
+    const lo = Math.min(v.y1, v.y2), hi = Math.max(v.y1, v.y2);
+    if (lo <= line.y1 + UNDER_GRID_EPS && hi >= line.y1 - UNDER_GRID_EPS) return true;
+  }
+  return false;
+}
+function hasBoxTopPair(line, maxH, horizontals) {
+  const w = line.x2 - line.x1;
+  if (w <= 0) return false;
+  for (const o of horizontals) {
+    if (o === line) continue;
+    const dy = o.y1 - line.y1;
+    if (dy < maxH * UNDER_BOX_PAIR_MIN_EM || dy > maxH * UNDER_BOX_PAIR_MAX_EM) continue;
+    const ow = o.x2 - o.x1;
+    if (ow <= 0) continue;
+    const overlap = Math.min(line.x2, o.x2) - Math.max(line.x1, o.x1);
+    if (overlap / Math.max(w, ow) >= UNDER_BOX_PAIR_OVERLAP) return true;
+  }
+  return false;
+}
+function isRepeatedSpanRule(line, horizontals) {
+  const w = line.x2 - line.x1;
+  if (w <= 0) return false;
+  const ys = [];
+  for (const o of horizontals) {
+    const ow = o.x2 - o.x1;
+    if (ow <= 0) continue;
+    const overlap = Math.min(line.x2, o.x2) - Math.max(line.x1, o.x1);
+    if (overlap / Math.max(w, ow) < UNDER_SPAN_OVERLAP_RATIO) continue;
+    if (!ys.some((y) => Math.abs(y - o.y1) < 2)) ys.push(o.y1);
+  }
+  return ys.length >= UNDER_REPEATED_SPAN_LEVELS;
+}
+function wrapUnderlineRuns(items) {
+  const marked = items.filter((i) => i.underline);
+  if (marked.length === 0) return;
+  const lines = /* @__PURE__ */ new Map();
+  for (const item of marked) {
+    const key = Math.round(item.y / 3);
+    const arr = lines.get(key) || [];
+    arr.push(item);
+    lines.set(key, arr);
+  }
+  for (const arr of lines.values()) {
+    arr.sort((a, b) => a.x - b.x);
+    let runStart = 0;
+    for (let i = 1; i <= arr.length; i++) {
+      const prev = arr[i - 1];
+      const em = Math.max(prev.h > 0 ? prev.h : prev.fontSize, 1);
+      const gap = i < arr.length ? arr[i].x - (prev.x + prev.w) : Infinity;
+      if (gap > em) {
+        arr[runStart].text = "<u>" + arr[runStart].text;
+        prev.text = prev.text + "</u>";
+        runStart = i;
+      }
+    }
+  }
+}
+
 // src/pdf/page-blocks.ts
 function extractPageBlocksWithLines(items, pageNum, opList, pageWidth, pageHeight, extraLines, detectTables = true) {
   if (items.length === 0) return [];
@@ -3058,6 +3175,8 @@ function extractPageBlocksWithLines(items, pageNum, opList, pageWidth, pageHeigh
   verticals = bridgeSplitColumnVerticals(horizontals, verticals);
   markStrikethroughItems(items, horizontals);
   wrapStrikethroughRuns(items);
+  markUnderlineItems(items, horizontals, verticals);
+  wrapUnderlineRuns(items);
   const grids = detectTables ? buildTableGrids(horizontals, verticals) : [];
   if (grids.length > 0) {
     return extractBlocksWithGrids(items, pageNum, grids, horizontals, verticals);
@@ -3568,4 +3687,4 @@ export {
   extractPageBlocksWithLines,
   mergeCrossPageTables
 };
-//# sourceMappingURL=chunk-BGSGUONP.js.map
+//# sourceMappingURL=chunk-PEVPKD6Z.js.map