Split off CPDF_Parser and CPDF_SimpleParser into .h/.cpp files

R=dsinclair@chromium.org

Review URL: https://codereview.chromium.org/1773103003 .

3 files changed, 209 insertions, 190 deletions

diff --git a/core/include/fpdfapi/cpdf_parser.h b/core/include/fpdfapi/cpdf_parser.h
new file mode 100644
index 0000000000..80a1107743
--- /dev/null
+++ b/core/include/fpdfapi/cpdf_parser.h
@@ -0,0 +1,173 @@
+// Copyright 2016 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
+
+#ifndef CORE_INCLUDE_FPDFAPI_CPDF_PARSER_H_
+#define CORE_INCLUDE_FPDFAPI_CPDF_PARSER_H_
+
+#include <map>
+#include <memory>
+#include <set>
+
+#include "core/include/fxcrt/fx_basic.h"
+
+class CPDF_Array;
+class CPDF_CryptoHandler;
+class CPDF_Dictionary;
+class CPDF_Document;
+class CPDF_IndirectObjectHolder;
+class CPDF_Object;
+class CPDF_StreamAcc;
+class CPDF_SyntaxParser;
+class IFX_FileRead;
+class IPDF_SecurityHandler;
+
+class CPDF_Parser {
+ public:
+  enum Error {
+    SUCCESS = 0,
+    FILE_ERROR,
+    FORMAT_ERROR,
+    PASSWORD_ERROR,
+    HANDLER_ERROR
+  };
+
+  CPDF_Parser();
+  ~CPDF_Parser();
+
+  Error StartParse(IFX_FileRead* pFile);
+  FX_DWORD GetPermissions(FX_BOOL bCheckRevision = FALSE);
+
+  void SetPassword(const FX_CHAR* password) { m_Password = password; }
+  CFX_ByteString GetPassword() { return m_Password; }
+  CPDF_Dictionary* GetTrailer() const { return m_pTrailer; }
+  FX_FILESIZE GetLastXRefOffset() const { return m_LastXRefOffset; }
+  CPDF_Document* GetDocument() const { return m_pDocument; }
+
+  FX_DWORD GetRootObjNum();
+  FX_DWORD GetInfoObjNum();
+  CPDF_Array* GetIDArray();
+
+  CPDF_Dictionary* GetEncryptDict() const { return m_pEncryptDict; }
+
+  CPDF_Object* ParseIndirectObject(CPDF_IndirectObjectHolder* pObjList,
+                                   FX_DWORD objnum);
+
+  FX_DWORD GetLastObjNum() const;
+  bool IsValidObjectNumber(FX_DWORD objnum) const;
+  FX_FILESIZE GetObjectPositionOrZero(FX_DWORD objnum) const;
+  uint8_t GetObjectType(FX_DWORD objnum) const;
+  uint16_t GetObjectGenNum(FX_DWORD objnum) const;
+  bool IsVersionUpdated() const { return m_bVersionUpdated; }
+  bool IsObjectFreeOrNull(FX_DWORD objnum) const;
+  FX_BOOL IsFormStream(FX_DWORD objnum, FX_BOOL& bForm);
+  CPDF_CryptoHandler* GetCryptoHandler();
+  IFX_FileRead* GetFileAccess() const;
+
+  FX_FILESIZE GetObjectOffset(FX_DWORD objnum) const;
+  FX_FILESIZE GetObjectSize(FX_DWORD objnum) const;
+
+  void GetIndirectBinary(FX_DWORD objnum, uint8_t*& pBuffer, FX_DWORD& size);
+  int GetFileVersion() const { return m_FileVersion; }
+  FX_BOOL IsXRefStream() const { return m_bXRefStream; }
+
+  CPDF_Object* ParseIndirectObjectAt(CPDF_IndirectObjectHolder* pObjList,
+                                     FX_FILESIZE pos,
+                                     FX_DWORD objnum);
+
+  CPDF_Object* ParseIndirectObjectAtByStrict(
+      CPDF_IndirectObjectHolder* pObjList,
+      FX_FILESIZE pos,
+      FX_DWORD objnum,
+      FX_FILESIZE* pResultPos);
+
+  Error StartAsyncParse(IFX_FileRead* pFile);
+
+  FX_DWORD GetFirstPageNo() const { return m_dwFirstPageNo; }
+
+ protected:
+  struct ObjectInfo {
+    ObjectInfo() : pos(0), type(0), gennum(0) {}
+
+    FX_FILESIZE pos;
+    uint8_t type;
+    uint16_t gennum;
+  };
+
+  void CloseParser();
+  CPDF_Object* ParseDirect(CPDF_Object* pObj);
+  FX_BOOL LoadAllCrossRefV4(FX_FILESIZE pos);
+  FX_BOOL LoadAllCrossRefV5(FX_FILESIZE pos);
+  bool LoadCrossRefV4(FX_FILESIZE pos, FX_FILESIZE streampos, FX_BOOL bSkip);
+  FX_BOOL LoadCrossRefV5(FX_FILESIZE* pos, FX_BOOL bMainXRef);
+  CPDF_Dictionary* LoadTrailerV4();
+  FX_BOOL RebuildCrossRef();
+  Error SetEncryptHandler();
+  void ReleaseEncryptHandler();
+  FX_BOOL LoadLinearizedAllCrossRefV4(FX_FILESIZE pos, FX_DWORD dwObjCount);
+  FX_BOOL LoadLinearizedCrossRefV4(FX_FILESIZE pos, FX_DWORD dwObjCount);
+  FX_BOOL LoadLinearizedAllCrossRefV5(FX_FILESIZE pos);
+  Error LoadLinearizedMainXRefTable();
+  CPDF_StreamAcc* GetObjectStream(FX_DWORD number);
+  FX_BOOL IsLinearizedFile(IFX_FileRead* pFileAccess, FX_DWORD offset);
+  void SetEncryptDictionary(CPDF_Dictionary* pDict);
+  void ShrinkObjectMap(FX_DWORD size);
+
+  CPDF_Document* m_pDocument;
+  std::unique_ptr<CPDF_SyntaxParser> m_pSyntax;
+  bool m_bOwnFileRead;
+  int m_FileVersion;
+  CPDF_Dictionary* m_pTrailer;
+  CPDF_Dictionary* m_pEncryptDict;
+  FX_FILESIZE m_LastXRefOffset;
+  FX_BOOL m_bXRefStream;
+  std::unique_ptr<IPDF_SecurityHandler> m_pSecurityHandler;
+  CFX_ByteString m_bsRecipient;
+  CFX_ByteString m_FilePath;
+  CFX_ByteString m_Password;
+  std::map<FX_DWORD, ObjectInfo> m_ObjectInfo;
+  std::set<FX_FILESIZE> m_SortedOffset;
+  CFX_ArrayTemplate<CPDF_Dictionary*> m_Trailers;
+  FX_BOOL m_bVersionUpdated;
+  CPDF_Object* m_pLinearized;
+  FX_DWORD m_dwFirstPageNo;
+  FX_DWORD m_dwXrefStartObjNum;
+
+  // A map of object numbers to indirect streams. Map owns the streams.
+  std::map<FX_DWORD, std::unique_ptr<CPDF_StreamAcc>> m_ObjectStreamMap;
+
+  // Mapping of object numbers to offsets. The offsets are relative to the first
+  // object in the stream.
+  using StreamObjectCache = std::map<FX_DWORD, FX_DWORD>;
+
+  // Mapping of streams to their object caches. This is valid as long as the
+  // streams in |m_ObjectStreamMap| are valid.
+  std::map<CPDF_StreamAcc*, StreamObjectCache> m_ObjCache;
+
+  // All indirect object numbers that are being parsed.
+  std::set<FX_DWORD> m_ParsingObjNums;
+
+  friend class CPDF_DataAvail;
+
+ private:
+  enum class ParserState {
+    kDefault,
+    kComment,
+    kWhitespace,
+    kString,
+    kHexString,
+    kEscapedString,
+    kXref,
+    kObjNum,
+    kPostObjNum,
+    kGenNum,
+    kPostGenNum,
+    kTrailer,
+    kBeginObj,
+    kEndObj
+  };
+};
+
+#endif  // CORE_INCLUDE_FPDFAPI_CPDF_PARSER_H_
diff --git a/core/include/fpdfapi/cpdf_simple_parser.h b/core/include/fpdfapi/cpdf_simple_parser.h
new file mode 100644
index 0000000000..dc1441d18e
--- /dev/null
+++ b/core/include/fpdfapi/cpdf_simple_parser.h
@@ -0,0 +1,35 @@
+// Copyright 2016 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
+
+#ifndef CORE_INCLUDE_FPDFAPI_CPDF_SIMPLE_PARSER_H_
+#define CORE_INCLUDE_FPDFAPI_CPDF_SIMPLE_PARSER_H_
+
+#include "core/include/fxcrt/fx_string.h"
+#include "core/include/fxcrt/fx_system.h"
+
+class CPDF_SimpleParser {
+ public:
+  CPDF_SimpleParser(const uint8_t* pData, FX_DWORD dwSize);
+  CPDF_SimpleParser(const CFX_ByteStringC& str);
+
+  CFX_ByteStringC GetWord();
+
+  // Find the token and its |nParams| parameters from the start of data,
+  // and move the current position to the start of those parameters.
+  bool FindTagParamFromStart(const CFX_ByteStringC& token, int nParams);
+
+  // For testing only.
+  FX_DWORD GetCurPos() const { return m_dwCurPos; }
+
+ private:
+  void ParseWord(const uint8_t*& pStart, FX_DWORD& dwSize);
+
+  const uint8_t* m_pData;
+  FX_DWORD m_dwSize;
+  FX_DWORD m_dwCurPos;
+};
+
+#endif  // CORE_INCLUDE_FPDFAPI_CPDF_SIMPLE_PARSER_H_
diff --git a/core/include/fpdfapi/fpdf_parser.h b/core/include/fpdfapi/fpdf_parser.h
index ec7a98cb91..06030701b4 100644
--- a/core/include/fpdfapi/fpdf_parser.h
+++ b/core/include/fpdfapi/fpdf_parser.h
@@ -1,4 +1,4 @@
-// Copyright 2014 PDFium Authors. All rights reserved.
+// Copyright 2016 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.
 
@@ -47,26 +47,6 @@ class IPDF_SecurityHandler;
 using ScopedFileStream =
     std::unique_ptr<IFX_FileStream, ReleaseDeleter<IFX_FileStream>>;
 
-// Use the accessors below instead of directly accessing PDF_CharType.
-extern const char PDF_CharType[256];
-
-inline bool PDFCharIsWhitespace(uint8_t c) {
-  return PDF_CharType[c] == 'W';
-}
-inline bool PDFCharIsNumeric(uint8_t c) {
-  return PDF_CharType[c] == 'N';
-}
-inline bool PDFCharIsDelimiter(uint8_t c) {
-  return PDF_CharType[c] == 'D';
-}
-inline bool PDFCharIsOther(uint8_t c) {
-  return PDF_CharType[c] == 'R';
-}
-
-inline bool PDFCharIsLineEnding(uint8_t c) {
-  return c == '\r' || c == '\n';
-}
-
 template <typename T>
 class ScopedSetInsertion {
  public:
@@ -84,175 +64,6 @@ class ScopedSetInsertion {
 // Indexed by 8-bit char code, contains unicode code points.
 extern const FX_WORD PDFDocEncoding[256];
 
-
-class CPDF_SimpleParser {
- public:
-  CPDF_SimpleParser(const uint8_t* pData, FX_DWORD dwSize);
-  CPDF_SimpleParser(const CFX_ByteStringC& str);
-
-  CFX_ByteStringC GetWord();
-
-  // Find the token and its |nParams| parameters from the start of data,
-  // and move the current position to the start of those parameters.
-  bool FindTagParamFromStart(const CFX_ByteStringC& token, int nParams);
-
-  // For testing only.
-  FX_DWORD GetCurPos() const { return m_dwCurPos; }
-
- private:
-  void ParseWord(const uint8_t*& pStart, FX_DWORD& dwSize);
-
-  const uint8_t* m_pData;
-  FX_DWORD m_dwSize;
-  FX_DWORD m_dwCurPos;
-};
-
-class CPDF_Parser {
- public:
-  enum Error {
-    SUCCESS = 0,
-    FILE_ERROR,
-    FORMAT_ERROR,
-    PASSWORD_ERROR,
-    HANDLER_ERROR
-  };
-
-  CPDF_Parser();
-  ~CPDF_Parser();
-
-  Error StartParse(IFX_FileRead* pFile);
-  FX_DWORD GetPermissions(FX_BOOL bCheckRevision = FALSE);
-
-  void SetPassword(const FX_CHAR* password) { m_Password = password; }
-  CFX_ByteString GetPassword() { return m_Password; }
-  CPDF_Dictionary* GetTrailer() const { return m_pTrailer; }
-  FX_FILESIZE GetLastXRefOffset() const { return m_LastXRefOffset; }
-  CPDF_Document* GetDocument() const { return m_pDocument; }
-
-  FX_DWORD GetRootObjNum();
-  FX_DWORD GetInfoObjNum();
-  CPDF_Array* GetIDArray();
-
-  CPDF_Dictionary* GetEncryptDict() const { return m_pEncryptDict; }
-
-  CPDF_Object* ParseIndirectObject(CPDF_IndirectObjectHolder* pObjList,
-                                   FX_DWORD objnum);
-
-  FX_DWORD GetLastObjNum() const;
-  bool IsValidObjectNumber(FX_DWORD objnum) const;
-  FX_FILESIZE GetObjectPositionOrZero(FX_DWORD objnum) const;
-  uint8_t GetObjectType(FX_DWORD objnum) const;
-  uint16_t GetObjectGenNum(FX_DWORD objnum) const;
-  bool IsVersionUpdated() const { return m_bVersionUpdated; }
-  bool IsObjectFreeOrNull(FX_DWORD objnum) const;
-  FX_BOOL IsFormStream(FX_DWORD objnum, FX_BOOL& bForm);
-  CPDF_CryptoHandler* GetCryptoHandler();
-  IFX_FileRead* GetFileAccess() const;
-
-  FX_FILESIZE GetObjectOffset(FX_DWORD objnum) const;
-  FX_FILESIZE GetObjectSize(FX_DWORD objnum) const;
-
-  void GetIndirectBinary(FX_DWORD objnum, uint8_t*& pBuffer, FX_DWORD& size);
-  int GetFileVersion() const { return m_FileVersion; }
-  FX_BOOL IsXRefStream() const { return m_bXRefStream; }
-
-  CPDF_Object* ParseIndirectObjectAt(CPDF_IndirectObjectHolder* pObjList,
-                                     FX_FILESIZE pos,
-                                     FX_DWORD objnum);
-
-  CPDF_Object* ParseIndirectObjectAtByStrict(
-      CPDF_IndirectObjectHolder* pObjList,
-      FX_FILESIZE pos,
-      FX_DWORD objnum,
-      FX_FILESIZE* pResultPos);
-
-  Error StartAsyncParse(IFX_FileRead* pFile);
-
-  FX_DWORD GetFirstPageNo() const { return m_dwFirstPageNo; }
-
- protected:
-  struct ObjectInfo {
-    ObjectInfo() : pos(0), type(0), gennum(0) {}
-
-    FX_FILESIZE pos;
-    uint8_t type;
-    uint16_t gennum;
-  };
-
-  void CloseParser();
-  CPDF_Object* ParseDirect(CPDF_Object* pObj);
-  FX_BOOL LoadAllCrossRefV4(FX_FILESIZE pos);
-  FX_BOOL LoadAllCrossRefV5(FX_FILESIZE pos);
-  bool LoadCrossRefV4(FX_FILESIZE pos, FX_FILESIZE streampos, FX_BOOL bSkip);
-  FX_BOOL LoadCrossRefV5(FX_FILESIZE* pos, FX_BOOL bMainXRef);
-  CPDF_Dictionary* LoadTrailerV4();
-  FX_BOOL RebuildCrossRef();
-  Error SetEncryptHandler();
-  void ReleaseEncryptHandler();
-  FX_BOOL LoadLinearizedAllCrossRefV4(FX_FILESIZE pos, FX_DWORD dwObjCount);
-  FX_BOOL LoadLinearizedCrossRefV4(FX_FILESIZE pos, FX_DWORD dwObjCount);
-  FX_BOOL LoadLinearizedAllCrossRefV5(FX_FILESIZE pos);
-  Error LoadLinearizedMainXRefTable();
-  CPDF_StreamAcc* GetObjectStream(FX_DWORD number);
-  FX_BOOL IsLinearizedFile(IFX_FileRead* pFileAccess, FX_DWORD offset);
-  void SetEncryptDictionary(CPDF_Dictionary* pDict);
-  void ShrinkObjectMap(FX_DWORD size);
-
-  CPDF_Document* m_pDocument;
-  std::unique_ptr<CPDF_SyntaxParser> m_pSyntax;
-  bool m_bOwnFileRead;
-  int m_FileVersion;
-  CPDF_Dictionary* m_pTrailer;
-  CPDF_Dictionary* m_pEncryptDict;
-  FX_FILESIZE m_LastXRefOffset;
-  FX_BOOL m_bXRefStream;
-  std::unique_ptr<IPDF_SecurityHandler> m_pSecurityHandler;
-  CFX_ByteString m_bsRecipient;
-  CFX_ByteString m_FilePath;
-  CFX_ByteString m_Password;
-  std::map<FX_DWORD, ObjectInfo> m_ObjectInfo;
-  std::set<FX_FILESIZE> m_SortedOffset;
-  CFX_ArrayTemplate<CPDF_Dictionary*> m_Trailers;
-  FX_BOOL m_bVersionUpdated;
-  CPDF_Object* m_pLinearized;
-  FX_DWORD m_dwFirstPageNo;
-  FX_DWORD m_dwXrefStartObjNum;
-
-  // A map of object numbers to indirect streams. Map owns the streams.
-  std::map<FX_DWORD, std::unique_ptr<CPDF_StreamAcc>> m_ObjectStreamMap;
-
-  // Mapping of object numbers to offsets. The offsets are relative to the first
-  // object in the stream.
-  using StreamObjectCache = std::map<FX_DWORD, FX_DWORD>;
-
-  // Mapping of streams to their object caches. This is valid as long as the
-  // streams in |m_ObjectStreamMap| are valid.
-  std::map<CPDF_StreamAcc*, StreamObjectCache> m_ObjCache;
-
-  // All indirect object numbers that are being parsed.
-  std::set<FX_DWORD> m_ParsingObjNums;
-
-  friend class CPDF_DataAvail;
-
- private:
-  enum class ParserState {
-    kDefault,
-    kComment,
-    kWhitespace,
-    kString,
-    kHexString,
-    kEscapedString,
-    kXref,
-    kObjNum,
-    kPostObjNum,
-    kGenNum,
-    kPostGenNum,
-    kTrailer,
-    kBeginObj,
-    kEndObj
-  };
-};
-
 #define FXCIPHER_NONE 0
 #define FXCIPHER_RC4 1
 #define FXCIPHER_AES 2