Remove CFX_DSPATemplate usage in CXFA_TextParser

This CL changes the check to use std::binary_search instead of the custom
lookup method. The tag validation has been split out to a separate method
with unit tests added.

Review-Url: https://codereview.chromium.org/1996623002

5 files changed, 77 insertions, 17 deletions

diff --git a/BUILD.gn b/BUILD.gn
index 7541de8aaa..448fe9d588 100644
--- a/BUILD.gn
+++ b/BUILD.gn
@@ -1643,6 +1643,7 @@ test("pdfium_unittests") {
     sources += [
       "xfa/fde/xml/fde_xml_imp_unittest.cpp",
       "xfa/fxbarcode/pdf417/BC_PDF417HighLevelEncoder_unittest.cpp",
+      "xfa/fxfa/app/xfa_textlayout_unittest.cpp",
       "xfa/fxfa/parser/xfa_utils_imp_unittest.cpp",
     ]
   }
diff --git a/pdfium.gyp b/pdfium.gyp
index 54d4058eec..fa0ab01cd6 100644
--- a/pdfium.gyp
+++ b/pdfium.gyp
@@ -947,6 +947,7 @@
           'sources': [
             'xfa/fde/xml/fde_xml_imp_unittest.cpp',
             'xfa/fxbarcode/pdf417/BC_PDF417HighLevelEncoder_unittest.cpp',
+            'xfa/fxfa/app/xfa_textlayout_unittest.cpp',
             'xfa/fxfa/parser/xfa_utils_imp_unittest.cpp',
           ],
         }],
diff --git a/xfa/fxfa/app/xfa_textlayout.cpp b/xfa/fxfa/app/xfa_textlayout.cpp
index b7dfa3f168..91dd814e8d 100644
--- a/xfa/fxfa/app/xfa_textlayout.cpp
+++ b/xfa/fxfa/app/xfa_textlayout.cpp
@@ -220,14 +220,14 @@ void CXFA_TextParser::DoParse(CFDE_XMLNode* pXMLContainer,
 }
 void CXFA_TextParser::ParseRichText(CFDE_XMLNode* pXMLNode,
                                     IFDE_CSSComputedStyle* pParentStyle) {
-  if (pXMLNode == NULL) {
+  if (!pXMLNode)
     return;
-  }
+
   CXFA_CSSTagProvider tagProvider;
   ParseTagInfo(pXMLNode, tagProvider);
-  if (!tagProvider.m_bTagAviliable) {
+  if (!tagProvider.m_bTagAvailable)
     return;
-  }
+
   IFDE_CSSComputedStyle* pNewStyle = NULL;
   if ((tagProvider.GetTagName() != FX_WSTRC(L"body")) ||
       (tagProvider.GetTagName() != FX_WSTRC(L"html"))) {
@@ -263,30 +263,45 @@ void CXFA_TextParser::ParseRichText(CFDE_XMLNode* pXMLNode,
   if (pNewStyle)
     pNewStyle->Release();
 }
-void CXFA_TextParser::ParseTagInfo(CFDE_XMLNode* pXMLNode,
-                                   CXFA_CSSTagProvider& tagProvider) {
+
+bool CXFA_TextParser::TagValidate(const CFX_WideString& wsName) const {
   static const uint32_t s_XFATagName[] = {
-      0x61,       0x62,       0x69,       0x70,       0x0001f714,
-      0x00022a55, 0x000239bb, 0x00025881, 0x0bd37faa, 0x0bd37fb8,
-      0xa73e3af2, 0xb182eaae, 0xdb8ac455,
+      0x61,        // a
+      0x62,        // b
+      0x69,        // i
+      0x70,        // p
+      0x0001f714,  // br
+      0x00022a55,  // li
+      0x000239bb,  // ol
+      0x00025881,  // ul
+      0x0bd37faa,  // sub
+      0x0bd37fb8,  // sup
+      0xa73e3af2,  // span
+      0xb182eaae,  // body
+      0xdb8ac455,  // html
   };
+  static const int32_t s_iCount = FX_ArraySize(s_XFATagName);
+
+  return std::binary_search(s_XFATagName, s_XFATagName + s_iCount,
+                            FX_HashCode_GetW(wsName.AsStringC(), true));
+}
+
+void CXFA_TextParser::ParseTagInfo(CFDE_XMLNode* pXMLNode,
+                                   CXFA_CSSTagProvider& tagProvider) {
   CFX_WideString wsName;
   if (pXMLNode->GetType() == FDE_XMLNODE_Element) {
     CFDE_XMLElement* pXMLElement = static_cast<CFDE_XMLElement*>(pXMLNode);
     pXMLElement->GetLocalTagName(wsName);
     tagProvider.SetTagNameObj(wsName);
-    uint32_t dwHashCode = FX_HashCode_GetW(wsName.AsStringC(), true);
-    static const int32_t s_iCount = sizeof(s_XFATagName) / sizeof(uint32_t);
-    CFX_DSPATemplate<uint32_t> lookup;
-    tagProvider.m_bTagAviliable =
-        lookup.Lookup(dwHashCode, s_XFATagName, s_iCount) > -1;
+    tagProvider.m_bTagAvailable = TagValidate(wsName);
+
     CFX_WideString wsValue;
     pXMLElement->GetString(L"style", wsValue);
     if (!wsValue.IsEmpty()) {
       tagProvider.SetAttribute(L"style", wsValue);
     }
   } else if (pXMLNode->GetType() == FDE_XMLNODE_Text) {
-    tagProvider.m_bTagAviliable = TRUE;
+    tagProvider.m_bTagAvailable = TRUE;
     tagProvider.m_bContent = TRUE;
   }
 }
diff --git a/xfa/fxfa/app/xfa_textlayout.h b/xfa/fxfa/app/xfa_textlayout.h
index 4d03aee302..c34f181746 100644
--- a/xfa/fxfa/app/xfa_textlayout.h
+++ b/xfa/fxfa/app/xfa_textlayout.h
@@ -26,7 +26,7 @@ class CXFA_TextTabstopsContext;
 
 class CXFA_CSSTagProvider {
  public:
-  CXFA_CSSTagProvider() : m_bTagAviliable(FALSE), m_bContent(FALSE) {}
+  CXFA_CSSTagProvider() : m_bTagAvailable(FALSE), m_bContent(FALSE) {}
   ~CXFA_CSSTagProvider() {}
 
   CFX_WideString GetTagName() { return m_wsTagName; }
@@ -43,7 +43,7 @@ class CXFA_CSSTagProvider {
     m_Attributes.insert({wsAttr, wsValue});
   }
 
-  FX_BOOL m_bTagAviliable;
+  FX_BOOL m_bTagAvailable;
   FX_BOOL m_bContent;
 
  protected:
@@ -124,6 +124,9 @@ class CXFA_TextParser {
                        CFX_WideString& wsValue);
   CXFA_TextParseContext* GetParseContextFromMap(CFDE_XMLNode* pXMLNode);
 
+ protected:
+  bool TagValidate(const CFX_WideString& str) const;
+
  private:
   void InitCSSData(CXFA_TextProvider* pTextProvider);
   void ParseRichText(CFDE_XMLNode* pXMLNode,
diff --git a/xfa/fxfa/app/xfa_textlayout_unittest.cpp b/xfa/fxfa/app/xfa_textlayout_unittest.cpp
new file mode 100644
index 0000000000..7354631dbb
--- /dev/null
+++ b/xfa/fxfa/app/xfa_textlayout_unittest.cpp
@@ -0,0 +1,40 @@
+// 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.
+
+#include "xfa/fxfa/app/xfa_textlayout.h"
+
+#include "testing/gtest/include/gtest/gtest.h"
+
+class CXFA_TestTextParser : public CXFA_TextParser {
+ public:
+  CXFA_TestTextParser() : CXFA_TextParser() {}
+
+ private:
+  // Add test cases as friends to access protected member functions.
+  FRIEND_TEST(CXFA_TextParser, TagValidate);
+};
+
+TEST(CXFA_TextParser, TagValidate) {
+  CXFA_TestTextParser parser;
+  EXPECT_TRUE(parser.TagValidate(L"br"));
+  EXPECT_TRUE(parser.TagValidate(L"Br"));
+  EXPECT_TRUE(parser.TagValidate(L"BR"));
+  EXPECT_TRUE(parser.TagValidate(L"a"));
+  EXPECT_TRUE(parser.TagValidate(L"b"));
+  EXPECT_TRUE(parser.TagValidate(L"i"));
+  EXPECT_TRUE(parser.TagValidate(L"p"));
+  EXPECT_TRUE(parser.TagValidate(L"li"));
+  EXPECT_TRUE(parser.TagValidate(L"ol"));
+  EXPECT_TRUE(parser.TagValidate(L"ul"));
+  EXPECT_TRUE(parser.TagValidate(L"sub"));
+  EXPECT_TRUE(parser.TagValidate(L"sup"));
+  EXPECT_TRUE(parser.TagValidate(L"span"));
+  EXPECT_TRUE(parser.TagValidate(L"body"));
+  EXPECT_TRUE(parser.TagValidate(L"html"));
+
+  EXPECT_FALSE(parser.TagValidate(L""));
+  EXPECT_FALSE(parser.TagValidate(L"tml"));
+  EXPECT_FALSE(parser.TagValidate(L"xhtml"));
+  EXPECT_FALSE(parser.TagValidate(L"htmlx"));
+}