SkPDF: Implement /ActualText to make text extraction correct.

src/pdf/SkPDFMakeToUnicodeCmap.cpp

 /*
  * Copyright 2011 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #include "SkPDFMakeToUnicodeCmap.h"
 #include "SkPDFUtils.h"
 #include "SkUtils.h"
@@ skipping 51 matching lines @@
     SkUnichar fUnicode;
 };
 
 struct BFRange {
     SkGlyphID fStart;
     SkGlyphID fEnd;
     SkUnichar fUnicode;
 };
 }  // namespace
 
-static void write_utf16be(SkDynamicMemoryWStream* wStream, SkUnichar utf32) {
-    SkGlyphID utf16[2] = {0, 0};
-    size_t len = SkUTF16_FromUnichar(utf32, utf16);
-    SkASSERT(len == 1 || len == 2);
-    SkPDFUtils::WriteUInt16BE(wStream, utf16[0]);
-    if (len == 2) {
-        SkPDFUtils::WriteUInt16BE(wStream, utf16[1]);
-    }
-}
-
 static void write_glyph(SkDynamicMemoryWStream* cmap,
                         bool multiByte,
                         SkGlyphID gid) {
     if (multiByte) {
         SkPDFUtils::WriteUInt16BE(cmap, gid);
     } else {
         SkPDFUtils::WriteUInt8(cmap, SkToU8(gid));
     }
 }
 
 static void append_bfchar_section(const SkTDArray<BFChar>& bfchar,
                                   bool multiByte,
                                   SkDynamicMemoryWStream* cmap) {
     // PDF spec defines that every bf* list can have at most 100 entries.
     for (int i = 0; i < bfchar.count(); i += 100) {
         int count = bfchar.count() - i;
         count = SkMin32(count, 100);
         cmap->writeDecAsText(count);
         cmap->writeText(" beginbfchar\n");
         for (int j = 0; j < count; ++j) {
             cmap->writeText("<");
             write_glyph(cmap, multiByte, bfchar[i + j].fGlyphId);
             cmap->writeText("> <");
-            write_utf16be(cmap, bfchar[i + j].fUnicode);
+            SkPDFUtils::WriteUTF16beHex(cmap, bfchar[i + j].fUnicode);
             cmap->writeText(">\n");
         }
         cmap->writeText("endbfchar\n");
     }
 }
 
 static void append_bfrange_section(const SkTDArray<BFRange>& bfrange,
                                    bool multiByte,
                                    SkDynamicMemoryWStream* cmap) {
     // PDF spec defines that every bf* list can have at most 100 entries.
     for (int i = 0; i < bfrange.count(); i += 100) {
         int count = bfrange.count() - i;
         count = SkMin32(count, 100);
         cmap->writeDecAsText(count);
         cmap->writeText(" beginbfrange\n");
         for (int j = 0; j < count; ++j) {
             cmap->writeText("<");
             write_glyph(cmap, multiByte, bfrange[i + j].fStart);
             cmap->writeText("> <");
             write_glyph(cmap, multiByte, bfrange[i + j].fEnd);
             cmap->writeText("> <");
-            write_utf16be(cmap, bfrange[i + j].fUnicode);
+            SkPDFUtils::WriteUTF16beHex(cmap, bfrange[i + j].fUnicode);
             cmap->writeText(">\n");
         }
         cmap->writeText("endbfrange\n");
     }
 }
 
 // Generate <bfchar> and <bfrange> table according to PDF spec 1.4 and Adobe
 // Technote 5014.
 // The function is not static so we can test it in unit tests.
 //
@@ skipping 88 matching lines @@
         SkGlyphID firstGlyphID,
         SkGlyphID lastGlyphID) {
     SkDynamicMemoryWStream cmap;
     append_tounicode_header(&cmap, multiByteGlyphs);
     SkPDFAppendCmapSections(glyphToUnicode, subset, &cmap, multiByteGlyphs,
                             firstGlyphID, lastGlyphID);
     append_cmap_footer(&cmap);
     return sk_make_sp<SkPDFStream>(
             std::unique_ptr<SkStreamAsset>(cmap.detachAsStream()));
 }