PDF: On Linux, convert font names to UTF-8 if needed.

pdf/pdfium/pdfium_engine.cc

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "pdf/pdfium/pdfium_engine.h"
 
 #include <math.h>
 
+#include "base/i18n/icu_encoding_detection.h"
+#include "base/i18n/icu_string_conversions.h"
 #include "base/json/json_writer.h"
 #include "base/logging.h"
 #include "base/memory/scoped_ptr.h"
 #include "base/numerics/safe_conversions.h"
 #include "base/stl_util.h"
 #include "base/strings/string_number_conversions.h"
 #include "base/strings/string_piece.h"
 #include "base/strings/string_util.h"
 #include "base/strings/utf_string_conversions.h"
 #include "base/values.h"
@@ skipping 148 matching lines @@
     // MS P?(Mincho|Gothic) are the most notable fonts in Japanese PDF files
     // without embedding the glyphs. Sometimes the font names are encoded
     // in Japanese Windows's locale (CP932/Shift_JIS) without space.
     // Most Linux systems don't have the exact font, but for outsourcing
     // fontconfig to find substitutable font in the system, we pass ASCII
     // font names to it.
     {"MS-PGothic", "MS PGothic", false, false},
     {"MS-Gothic", "MS Gothic", false, false},
     {"MS-PMincho", "MS PMincho", false, false},
     {"MS-Mincho", "MS Mincho", false, false},
     // MS PGothic in Shift_JIS encoding.

[kochi, 2015/05/20 06:24:27]

With the code added in this CL, the following should not be
necessary - as the Shift_JIS encoded font name will be passed
as UTF-8 to fontconfig and fontconfig should resolve it.

But anyway, these entries even translates into ASCII representation
of the name, so fontconfig still have a chance to fail resolving
the font with the UTF-8 name, so keep these.

     {"\x82\x6C\x82\x72\x82\x6F\x83\x53\x83\x56\x83\x62\x83\x4E",
      "MS PGothic", false, false},
     // MS Gothic in Shift_JIS encoding.
     {"\x82\x6C\x82\x72\x83\x53\x83\x56\x83\x62\x83\x4E",
      "MS Gothic", false, false},
     // MS PMincho in Shift_JIS encoding.
     {"\x82\x6C\x82\x72\x82\x6F\x96\xBE\x92\xA9",
      "MS PMincho", false, false},
     // MS Mincho in Shift_JIS encoding.
     {"\x82\x6C\x82\x72\x96\xBE\x92\xA9",
@@ skipping 28 matching lines @@
       description.set_face(PDFFontSubstitutions[i].face);
       if (PDFFontSubstitutions[i].bold)
         description.set_weight(PP_BROWSERFONT_TRUSTED_WEIGHT_BOLD);
       if (PDFFontSubstitutions[i].italic)
         description.set_italic(true);
       break;
     }
   }
 
   if (i == arraysize(PDFFontSubstitutions)) {
-    // TODO(kochi): Pass the face in UTF-8. If face is not encoded in UTF-8,
-    // convert to UTF-8 before passing.
-    description.set_face(face);
+    // Convert to UTF-8 before calling set_face().
+    std::string face_utf8;
+    if (base::IsStringUTF8(face)) {
+      face_utf8 = face;
+    } else {
+      std::string encoding;
+      if (base::DetectEncoding(face, &encoding)) {
+        // ConvertToUtf8AndNormalize() clears |face_utf8| on failure.
+        base::ConvertToUtf8AndNormalize(face, encoding, &face_utf8);
+      }
+    }
 
+    if (face_utf8.empty())
+      return nullptr;
+
+    description.set_face(face_utf8);
     description.set_weight(WeightToBrowserFontTrustedWeight(weight));
     description.set_italic(italic > 0);
   }
 
   if (!pp::PDF::IsAvailable()) {
     NOTREACHED();
     return NULL;
   }
 
   PP_Resource font_resource = pp::PDF::GetFontFileWithFallback(
@@ skipping 3765 matching lines @@
     double* height) {
   FPDF_DOCUMENT doc = FPDF_LoadMemDocument(pdf_buffer, pdf_buffer_size, NULL);
   if (!doc)
     return false;
   bool success = FPDF_GetPageSizeByIndex(doc, page_number, width, height) != 0;
   FPDF_CloseDocument(doc);
   return success;
 }
 
 }  // namespace chrome_pdf