Improve text-selection for Type3 fonts with bogus /FontBBox-entries (issue 14999)

This extends PR 13461, by also building a fallback bounding box for Type3 fonts that contain a much too small /FontBBox-entry.

*Please note:* While this patch improves things overall, copy-and-pasting still doesn't work perfectly for this document. In particular the lowercase letter "c" cannot be selected/copied, however this can be reproduced in both Adobe Reader and PDFium (in Google Chrome) too, which is caused by a lack of proper /ToUnicode-data in the PDF document.

src/core/evaluator.js

@@ -4404,8 +4404,10 @@ class TranslatedFont {
     const fontResources = this.dict.get("Resources") || resources;
     const charProcOperatorList = Object.create(null);
 
-    const isEmptyBBox =
+    const fontBBox = Util.normalizeRect(translatedFont.bbox || [0, 0, 0, 0]),
-      !translatedFont.bbox || isArrayEqual(translatedFont.bbox, [0, 0, 0, 0]);
+      width = fontBBox[2] - fontBBox[0],
+      height = fontBBox[3] - fontBBox[1];
+    const fontBBoxSize = Math.hypot(width, height);
 
     for (const key of charProcs.getKeys()) {
       loadCharProcsPromise = loadCharProcsPromise.then(() => {
@@ -4426,7 +4428,7 @@ class TranslatedFont {
             //   colour-related parameters) in the graphics state;
             //   any use of such operators shall be ignored."
             if (operatorList.fnArray[0] === OPS.setCharWidthAndBounds) {
-              this._removeType3ColorOperators(operatorList, isEmptyBBox);
+              this._removeType3ColorOperators(operatorList, fontBBoxSize);
             }
             charProcOperatorList[key] = operatorList.getIR();
 
@@ -4454,7 +4456,7 @@ class TranslatedFont {
   /**
    * @private
    */
-  _removeType3ColorOperators(operatorList, isEmptyBBox = false) {
+  _removeType3ColorOperators(operatorList, fontBBoxSize = NaN) {
     if (
       typeof PDFJSDev === "undefined" ||
       PDFJSDev.test("!PRODUCTION || TESTING")
@@ -4467,12 +4469,19 @@ class TranslatedFont {
     const charBBox = Util.normalizeRect(operatorList.argsArray[0].slice(2)),
       width = charBBox[2] - charBBox[0],
       height = charBBox[3] - charBBox[1];
+    const charBBoxSize = Math.hypot(width, height);
 
     if (width === 0 || height === 0) {
       // Skip the d1 operator when its bounds are bogus (fixes issue14953.pdf).
       operatorList.fnArray.splice(0, 1);
       operatorList.argsArray.splice(0, 1);
-    } else if (isEmptyBBox) {
+    } else if (
+      fontBBoxSize === 0 ||
+      Math.round(charBBoxSize / fontBBoxSize) >= 10
+    ) {
+      // Override the fontBBox when it's undefined/empty, or when it's at least
+      // (approximately) one order of magnitude smaller than the charBBox
+      // (fixes issue14999_reduced.pdf).
       if (!this._bbox) {
         this._bbox = [Infinity, Infinity, -Infinity, -Infinity];
       }

test/pdfs/.gitignore

@@ -121,6 +121,7 @@
 !issue13916.pdf
 !issue14023.pdf
 !issue14438.pdf
+!issue14999_reduced.pdf
 !bad-PageLabels.pdf
 !decodeACSuccessive.pdf
 !issue13003.pdf

test/test_manifest.json

@@ -2848,6 +2848,12 @@
        "link": false,
        "type": "text"
     },
+    {  "id": "issue14999",
+       "file": "pdfs/issue14999_reduced.pdf",
+       "md5": "a4e664e734f6869aa66245e72d448874",
+       "rounds": 1,
+       "type": "text"
+    },
     {  "id": "issue6901-eq",
        "file": "pdfs/issue6901.pdf",
        "md5": "1a0604b1a7a3aaf2162b425a9a84230b",