Cache object numbers in CPDF_Parser::ParseIndirectObject().

core/src/fpdfapi/fpdf_parser/fpdf_parser_parser.cpp

 // Copyright 2014 PDFium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 // Original code copyright 2014 Foxit Software Inc. http://www.foxitsoftware.com
 
 #include "parser_int.h"
 
 #include <set>
 #include <utility>
@@ skipping 123 matching lines @@
     m_Syntax.m_pFileAccess = NULL;
   }
   FX_POSITION pos = m_ObjectStreamMap.GetStartPosition();
   while (pos) {
     void* objnum;
     CPDF_StreamAcc* pStream;
     m_ObjectStreamMap.GetNextAssoc(pos, objnum, (void*&)pStream);
     delete pStream;
   }
   m_ObjectStreamMap.RemoveAll();
+  m_ObjCache.clear();
+
   m_SortedOffset.RemoveAll();
   m_CrossRef.RemoveAll();
   m_V5Type.RemoveAll();
   m_ObjVersion.RemoveAll();
   int32_t iLen = m_Trailers.GetSize();
   for (int32_t i = 0; i < iLen; ++i) {
     if (CPDF_Dictionary* trailer = m_Trailers.GetAt(i))
       trailer->Release();
   }
   m_Trailers.RemoveAll();
@@ skipping 1040 matching lines @@
 
   CPDF_StreamAcc* pObjStream = GetObjectStream((FX_DWORD)m_CrossRef[objnum]);
   if (!pObjStream)
     return nullptr;
 
   ScopedFileStream file(FX_CreateMemoryStream(
       (uint8_t*)pObjStream->GetData(), (size_t)pObjStream->GetSize(), FALSE));
   CPDF_SyntaxParser syntax;
   syntax.InitParser(file.get(), 0);
   int32_t offset = GetStreamFirst(pObjStream);
-  for (int32_t i = GetStreamNCount(pObjStream); i > 0; --i) {
-    FX_DWORD thisnum = syntax.GetDirectNum();
-    FX_DWORD thisoff = syntax.GetDirectNum();
-    if (thisnum == objnum) {
-      syntax.RestorePos(offset + thisoff);
-      return syntax.GetObject(pObjList, 0, 0, pContext);
+
+  // Read object numbers from |pObjStream| into a cache.
+  if (m_ObjCache.find(pObjStream) == m_ObjCache.end()) {
+    FX_FILESIZE saved_pos = syntax.SavePos();
+    for (int32_t i = GetStreamNCount(pObjStream); i > 0; --i) {
+      FX_DWORD thisnum = syntax.GetDirectNum();
+      FX_DWORD thisoff = syntax.GetDirectNum();
+      m_ObjCache[pObjStream][thisnum] = thisoff;
     }
+    syntax.RestorePos(saved_pos);

[jun_fang, 2015/11/19 16:19:45]

nit: It seems that there is no need to restore the position here.
     

[Lei Zhang, 2015/11/20 07:11:42]

Done. I needed it in an earlier iteration, but then I arranged the code and it
was cut + pasted.

   }
-  return nullptr;
+
+  const auto it = m_ObjCache[pObjStream].find(objnum);
+  if (it == m_ObjCache[pObjStream].end())
+    return nullptr;
+
+  syntax.RestorePos(offset + it->second);
+  return syntax.GetObject(pObjList, 0, 0, pContext);
 }
 
 CPDF_StreamAcc* CPDF_Parser::GetObjectStream(FX_DWORD objnum) {
   CPDF_StreamAcc* pStreamAcc = nullptr;
   if (m_ObjectStreamMap.Lookup((void*)(uintptr_t)objnum, (void*&)pStreamAcc))
     return pStreamAcc;
 
   const CPDF_Stream* pStream =
       ToStream(m_pDocument ? m_pDocument->GetIndirectObject(objnum) : nullptr);
   if (!pStream)
@@ skipping 429 matching lines @@
   }
   m_LastXRefOffset += dwCount;
   FX_POSITION pos = m_ObjectStreamMap.GetStartPosition();
   while (pos) {
     void* objnum;
     CPDF_StreamAcc* pStream;
     m_ObjectStreamMap.GetNextAssoc(pos, objnum, (void*&)pStream);
     delete pStream;
   }
   m_ObjectStreamMap.RemoveAll();
+  m_ObjCache.clear();
+
   if (!LoadLinearizedAllCrossRefV4(m_LastXRefOffset, m_dwXrefStartObjNum) &&
       !LoadLinearizedAllCrossRefV5(m_LastXRefOffset)) {
     m_LastXRefOffset = 0;
     m_Syntax.m_MetadataObjnum = dwSaveMetadataObjnum;
     return PDFPARSE_ERROR_FORMAT;
   }
   FXSYS_qsort(m_SortedOffset.GetData(), m_SortedOffset.GetSize(),
               sizeof(FX_FILESIZE), CompareFileSize);
   m_Syntax.m_MetadataObjnum = dwSaveMetadataObjnum;
   return PDFPARSE_ERROR_SUCCESS;
@@ skipping 3264 matching lines @@
   if (!m_pLinearizedDict)
     return -1;
   CPDF_Array* pRange = m_pLinearizedDict->GetArray(FX_BSTRC("H"));
   if (!pRange)
     return -1;
   CPDF_Object* pStreamLen = pRange->GetElementValue(1);
   if (!pStreamLen)
     return -1;
   return pStreamLen->GetInteger();
 }