Allow retrieving the XML tree from the CFX_XMLDoc

This CL allows the CXFA_SimpleParser to retrieve the XML tree from the
CFX_XMLDoc. This way, we don't have to keep the doc around and can store
the pointer to the tree in the CXFA_SimpleParser.

Change-Id: I5b478acbe61e6f1ca5fa04d03133a2b327a0cb1c
Reviewed-on: https://pdfium-review.googlesource.com/30210
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: dsinclair <dsinclair@chromium.org>

3 files changed, 27 insertions, 57 deletions

diff --git a/xfa/fxfa/parser/cxfa_document_parser.cpp b/xfa/fxfa/parser/cxfa_document_parser.cpp
index 18f9955210..883f684768 100644
--- a/xfa/fxfa/parser/cxfa_document_parser.cpp
+++ b/xfa/fxfa/parser/cxfa_document_parser.cpp
@@ -20,18 +20,15 @@ CXFA_DocumentParser::~CXFA_DocumentParser() {
 
 int32_t CXFA_DocumentParser::Parse(const RetainPtr<IFX_SeekableStream>& pStream,
                                    XFA_PacketType ePacketID) {
-  m_pDocument.reset();
-  m_nodeParser.CloseParser();
-
   m_pDocument = pdfium::MakeUnique<CXFA_Document>(GetNotify());
   m_nodeParser.SetFactory(m_pDocument.get());
 
-  int32_t nRetStatus = m_nodeParser.Parse(pStream, ePacketID);
-  if (nRetStatus == XFA_PARSESTATUS_Done) {
-    ASSERT(m_pDocument);
-    m_pDocument->SetRoot(m_nodeParser.GetRootNode());
-  }
-  return nRetStatus;
+  if (!m_nodeParser.Parse(pStream, ePacketID))
+    return XFA_PARSESTATUS_StatusErr;
+
+  ASSERT(m_pDocument);
+  m_pDocument->SetRoot(m_nodeParser.GetRootNode());
+  return XFA_PARSESTATUS_Done;
 }
 
 CXFA_FFNotify* CXFA_DocumentParser::GetNotify() const {
diff --git a/xfa/fxfa/parser/cxfa_simple_parser.cpp b/xfa/fxfa/parser/cxfa_simple_parser.cpp
index 225271fae8..232f1fbec8 100644
--- a/xfa/fxfa/parser/cxfa_simple_parser.cpp
+++ b/xfa/fxfa/parser/cxfa_simple_parser.cpp
@@ -100,11 +100,8 @@ const PacketInfo* GetPacketByName(const WideStringView& wsName) {
   return nullptr;
 }
 
-CFX_XMLNode* GetDocumentNode(CFX_XMLDoc* pXMLDoc) {
-  if (!pXMLDoc)
-    return nullptr;
-
-  for (CFX_XMLNode* pXMLNode = pXMLDoc->GetRoot()->GetFirstChild(); pXMLNode;
+CFX_XMLNode* GetDocumentNode(CFX_XMLNode* pRootNode) {
+  for (CFX_XMLNode* pXMLNode = pRootNode->GetFirstChild(); pXMLNode;
        pXMLNode = pXMLNode->GetNextSibling()) {
     if (pXMLNode->GetType() != FX_XMLNODE_Element)
       continue;
@@ -338,49 +335,33 @@ void CXFA_SimpleParser::SetFactory(CXFA_Document* pFactory) {
   m_pFactory = pFactory;
 }
 
-int32_t CXFA_SimpleParser::Parse(const RetainPtr<IFX_SeekableStream>& pStream,
-                                 XFA_PacketType ePacketID) {
-  CloseParser();
-
-  m_pFileRead = pStream;
-  m_pStream = pdfium::MakeRetain<CFX_SeekableStreamProxy>(pStream, false);
-  uint16_t wCodePage = m_pStream->GetCodePage();
+bool CXFA_SimpleParser::Parse(const RetainPtr<IFX_SeekableStream>& pStream,
+                              XFA_PacketType ePacketID) {
+  auto pStreamProxy =
+      pdfium::MakeRetain<CFX_SeekableStreamProxy>(pStream, false);
+  uint16_t wCodePage = pStreamProxy->GetCodePage();
   if (wCodePage != FX_CODEPAGE_UTF16LE && wCodePage != FX_CODEPAGE_UTF16BE &&
       wCodePage != FX_CODEPAGE_UTF8) {
-    m_pStream->SetCodePage(FX_CODEPAGE_UTF8);
+    pStreamProxy->SetCodePage(FX_CODEPAGE_UTF8);
   }
 
-  m_pXMLDoc = pdfium::MakeUnique<CFX_XMLDoc>(m_pStream);
-  if (!m_pXMLDoc)
-    return XFA_PARSESTATUS_StatusErr;
-
-  int32_t iRet = m_pXMLDoc->Load();
-  if (iRet < 0)
-    return XFA_PARSESTATUS_SyntaxErr;
-  if (iRet < 100)
-    return iRet / 2;
-
-  m_pRootNode = ParseAsXDPPacket(GetDocumentNode(m_pXMLDoc.get()), ePacketID);
-  m_pXMLDoc->CloseXML();
-  m_pStream.Reset();
-
-  if (!m_pRootNode)
-    return XFA_PARSESTATUS_StatusErr;
+  CFX_XMLDoc doc;
+  if (!doc.Load(pStreamProxy))
+    return false;
 
-  return XFA_PARSESTATUS_Done;
+  m_pNodeTree = doc.GetTree();
+  m_pRootNode = ParseAsXDPPacket(GetDocumentNode(m_pNodeTree.get()), ePacketID);
+  return !!m_pRootNode;
 }
 
 CFX_XMLNode* CXFA_SimpleParser::ParseXMLData(const ByteString& wsXML) {
-  CloseParser();
-
   auto pStream = pdfium::MakeRetain<CFX_SeekableStreamProxy>(
       const_cast<uint8_t*>(wsXML.raw_str()), wsXML.GetLength());
-  m_pXMLDoc = pdfium::MakeUnique<CFX_XMLDoc>(pStream);
+  CFX_XMLDoc doc;
+  if (doc.Load(pStream))
+    m_pNodeTree = doc.GetTree();
 
-  int32_t iRet = m_pXMLDoc->Load();
-  if (iRet < 0 || iRet >= 100)
-    m_pXMLDoc->CloseXML();
-  return iRet < 100 ? nullptr : GetDocumentNode(m_pXMLDoc.get());
+  return m_pNodeTree ? GetDocumentNode(m_pNodeTree.get()) : nullptr;
 }
 
 void CXFA_SimpleParser::ConstructXFANode(CXFA_Node* pXFANode,
@@ -1180,8 +1161,3 @@ void CXFA_SimpleParser::ParseInstruction(CXFA_Node* pXFANode,
     }
   }
 }
-
-void CXFA_SimpleParser::CloseParser() {
-  m_pXMLDoc.reset();
-  m_pStream.Reset();
-}
diff --git a/xfa/fxfa/parser/cxfa_simple_parser.h b/xfa/fxfa/parser/cxfa_simple_parser.h
index cef9c9af54..61edf8e8e6 100644
--- a/xfa/fxfa/parser/cxfa_simple_parser.h
+++ b/xfa/fxfa/parser/cxfa_simple_parser.h
@@ -25,13 +25,12 @@ class CXFA_SimpleParser {
   explicit CXFA_SimpleParser(CXFA_Document* pFactory);
   ~CXFA_SimpleParser();
 
-  int32_t Parse(const RetainPtr<IFX_SeekableStream>& pStream,
-                XFA_PacketType ePacketID);
+  bool Parse(const RetainPtr<IFX_SeekableStream>& pStream,
+             XFA_PacketType ePacketID);
 
   CFX_XMLNode* ParseXMLData(const ByteString& wsXML);
   void ConstructXFANode(CXFA_Node* pXFANode, CFX_XMLNode* pXMLNode);
   CXFA_Node* GetRootNode() const;
-  void CloseParser();
 
   // Called later for the ctor with no parameters.
   void SetFactory(CXFA_Document* pFactory);
@@ -71,10 +70,8 @@ class CXFA_SimpleParser {
                         CFX_XMLInstruction* pXMLInstruction,
                         XFA_PacketType ePacketID);
 
-  std::unique_ptr<CFX_XMLDoc> m_pXMLDoc;
-  RetainPtr<CFX_SeekableStreamProxy> m_pStream;
-  RetainPtr<IFX_SeekableStream> m_pFileRead;
   UnownedPtr<CXFA_Document> m_pFactory;
+  std::unique_ptr<CFX_XMLNode> m_pNodeTree;
   // TODO(dsinclair): Figure out who owns this.
   CXFA_Node* m_pRootNode = nullptr;
   const bool m_bDocumentParser;