Refactor CPDF_SimpleParser and add unit tests.

Remove unused member functions and simplify calls to find tag parameters.

R=thestig@chromium.org

Review URL: https://codereview.chromium.org/1681403007 .

2 files changed, 112 insertions, 53 deletions

diff --git a/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility.cpp b/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility.cpp
index e6e2f2aabe..d1a7231d45 100644
--- a/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility.cpp
+++ b/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility.cpp
@@ -65,11 +65,13 @@ CPDF_SimpleParser::CPDF_SimpleParser(const uint8_t* pData, FX_DWORD dwSize) {
   m_dwSize = dwSize;
   m_dwCurPos = 0;
 }
+
 CPDF_SimpleParser::CPDF_SimpleParser(const CFX_ByteStringC& str) {
   m_pData = str.GetPtr();
   m_dwSize = str.GetLength();
   m_dwCurPos = 0;
 }
+
 void CPDF_SimpleParser::ParseWord(const uint8_t*& pStart, FX_DWORD& dwSize) {
   pStart = NULL;
   dwSize = 0;
@@ -146,6 +148,7 @@ void CPDF_SimpleParser::ParseWord(const uint8_t*& pStart, FX_DWORD& dwSize) {
     dwSize++;
   }
 }
+
 CFX_ByteStringC CPDF_SimpleParser::GetWord() {
   const uint8_t* pStart;
   FX_DWORD dwSize;
@@ -190,60 +193,14 @@ CFX_ByteStringC CPDF_SimpleParser::GetWord() {
   }
   return CFX_ByteStringC(pStart, dwSize);
 }
-FX_BOOL CPDF_SimpleParser::SearchToken(const CFX_ByteStringC& token) {
-  int token_len = token.GetLength();
-  while (m_dwCurPos < m_dwSize - token_len) {
-    if (FXSYS_memcmp(m_pData + m_dwCurPos, token.GetPtr(), token_len) == 0) {
-      break;
-    }
-    m_dwCurPos++;
-  }
-  if (m_dwCurPos == m_dwSize - token_len) {
-    return FALSE;
-  }
-  m_dwCurPos += token_len;
-  return TRUE;
-}
-FX_BOOL CPDF_SimpleParser::SkipWord(const CFX_ByteStringC& token) {
-  while (1) {
-    CFX_ByteStringC word = GetWord();
-    if (word.IsEmpty()) {
-      return FALSE;
-    }
-    if (word == token) {
-      return TRUE;
-    }
-  }
-  return FALSE;
-}
-FX_BOOL CPDF_SimpleParser::FindTagPair(const CFX_ByteStringC& start_token,
-                                       const CFX_ByteStringC& end_token,
-                                       FX_DWORD& start_pos,
-                                       FX_DWORD& end_pos) {
-  if (!start_token.IsEmpty()) {
-    if (!SkipWord(start_token)) {
-      return FALSE;
-    }
-    start_pos = m_dwCurPos;
-  }
-  while (1) {
-    end_pos = m_dwCurPos;
-    CFX_ByteStringC word = GetWord();
-    if (word.IsEmpty()) {
-      return FALSE;
-    }
-    if (word == end_token) {
-      return TRUE;
-    }
-  }
-  return FALSE;
-}
-FX_BOOL CPDF_SimpleParser::FindTagParam(const CFX_ByteStringC& token,
-                                        int nParams) {
+
+bool CPDF_SimpleParser::FindTagParamFromStart(const CFX_ByteStringC& token,
+                                              int nParams) {
   nParams++;
   FX_DWORD* pBuf = FX_Alloc(FX_DWORD, nParams);
   int buf_index = 0;
   int buf_count = 0;
+  m_dwCurPos = 0;
   while (1) {
     pBuf[buf_index++] = m_dwCurPos;
     if (buf_index == nParams) {
@@ -256,7 +213,7 @@ FX_BOOL CPDF_SimpleParser::FindTagParam(const CFX_ByteStringC& token,
     CFX_ByteStringC word = GetWord();
     if (word.IsEmpty()) {
       FX_Free(pBuf);
-      return FALSE;
+      return false;
     }
     if (word == token) {
       if (buf_count < nParams) {
@@ -264,10 +221,10 @@ FX_BOOL CPDF_SimpleParser::FindTagParam(const CFX_ByteStringC& token,
       }
       m_dwCurPos = pBuf[buf_index];
       FX_Free(pBuf);
-      return TRUE;
+      return true;
     }
   }
-  return FALSE;
+  return false;
 }
 
 CFX_ByteString PDF_NameDecode(const CFX_ByteStringC& bstr) {
@@ -291,12 +248,14 @@ CFX_ByteString PDF_NameDecode(const CFX_ByteStringC& bstr) {
   result.ReleaseBuffer((FX_STRSIZE)(pDest - pDestStart));
   return result;
 }
+
 CFX_ByteString PDF_NameDecode(const CFX_ByteString& orig) {
   if (!FXSYS_memchr(orig.c_str(), '#', orig.GetLength())) {
     return orig;
   }
   return PDF_NameDecode(CFX_ByteStringC(orig));
 }
+
 CFX_ByteString PDF_NameEncode(const CFX_ByteString& orig) {
   uint8_t* src_buf = (uint8_t*)orig.c_str();
   int src_len = orig.GetLength();
@@ -332,6 +291,7 @@ CFX_ByteString PDF_NameEncode(const CFX_ByteString& orig) {
   res.ReleaseBuffer();
   return res;
 }
+
 CFX_ByteTextBuf& operator<<(CFX_ByteTextBuf& buf, const CPDF_Object* pObj) {
   if (!pObj) {
     buf << " null";
@@ -402,6 +362,7 @@ CFX_ByteTextBuf& operator<<(CFX_ByteTextBuf& buf, const CPDF_Object* pObj) {
   }
   return buf;
 }
+
 FX_FLOAT PDF_ClipFloat(FX_FLOAT f) {
   if (f < 0) {
     return 0;
@@ -411,6 +372,7 @@ FX_FLOAT PDF_ClipFloat(FX_FLOAT f) {
   }
   return f;
 }
+
 static CPDF_Object* SearchNumberNode(CPDF_Dictionary* pNode, int num) {
   CPDF_Array* pLimits = pNode->GetArrayBy("Limits");
   if (pLimits &&
@@ -447,6 +409,7 @@ static CPDF_Object* SearchNumberNode(CPDF_Dictionary* pNode, int num) {
   }
   return NULL;
 }
+
 CPDF_Object* CPDF_NumberTree::LookupValue(int num) {
   return SearchNumberNode(m_pRoot, num);
 }
diff --git a/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility_unittest.cpp b/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility_unittest.cpp
new file mode 100644
index 0000000000..1673798139
--- /dev/null
+++ b/core/src/fpdfapi/fpdf_parser/fpdf_parser_utility_unittest.cpp
@@ -0,0 +1,96 @@
+// 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 "core/include/fpdfapi/fpdf_parser.h"
+
+#include <string>
+
+#include "testing/gtest/include/gtest/gtest.h"
+#include "testing/test_support.h"
+
+TEST(SimpleParserTest, GetWord) {
+  pdfium::StrFuncTestData test_data[] = {
+      // Empty src string.
+      STR_TEST_CASE("", ""),
+      // Content with whitespaces only.
+      STR_TEST_CASE(" \t \0 \n", ""),
+      // Content with comments only.
+      STR_TEST_CASE("%this is a test case\r\n%2nd line", ""),
+      // Mixed whitespaces and comments.
+      STR_TEST_CASE(" \t \0%try()%haha\n %another line \aa", ""),
+      // Name.
+      STR_TEST_CASE(" /Tester ", "/Tester"),
+      // String.
+      STR_TEST_CASE("\t(nice day)!\n ", "(nice day)"),
+      // String with nested braces.
+      STR_TEST_CASE("\t(It is a (long) day)!\n ", "(It is a (long) day)"),
+      // String with escaped chars.
+      STR_TEST_CASE("\t(It is a \\(long\\) day!)hi\n ",
+                    "(It is a \\(long\\) day!)"),
+      // Hex string.
+      STR_TEST_CASE(" \n<4545acdfedertt>abc ", "<4545acdfedertt>"),
+      STR_TEST_CASE(" \n<4545a<ed>ertt>abc ", "<4545a<ed>"),
+      // Dictionary.
+      STR_TEST_CASE("<</oc 234 /color 2 3 R>>", "<<"),
+      STR_TEST_CASE("\t\t<< /abc>>", "<<"),
+      // Handling ending delimiters.
+      STR_TEST_CASE("> little bear", ">"),
+      STR_TEST_CASE(") another bear", ")"),
+      STR_TEST_CASE(">> end ", ">>"),
+      // No ending delimiters.
+      STR_TEST_CASE("(sdfgfgbcv", "(sdfgfgbcv"),
+      // Regular cases.
+      STR_TEST_CASE("apple pear", "apple"),
+      STR_TEST_CASE(" pi=3.1415 ", "pi=3.1415"),
+      STR_TEST_CASE(" p t x c ", "p"),
+      STR_TEST_CASE(" pt\0xc ", "pt"),
+      STR_TEST_CASE(" $^&&*\t\0sdff ", "$^&&*"),
+      STR_TEST_CASE("\n\r+3.5656 -11.0", "+3.5656"),
+  };
+  for (size_t i = 0; i < FX_ArraySize(test_data); ++i) {
+    const pdfium::StrFuncTestData& data = test_data[i];
+    CPDF_SimpleParser parser(data.input, data.input_size);
+    CFX_ByteStringC word = parser.GetWord();
+    EXPECT_EQ(std::string(reinterpret_cast<const char*>(data.expected),
+                          data.expected_size),
+              std::string(word.GetCStr(), word.GetLength()))
+        << " for case " << i;
+  }
+}
+
+TEST(SimpleParserTest, FindTagParamFromStart) {
+  struct FindTagTestStruct {
+    const unsigned char* input;
+    unsigned int input_size;
+    const char* token;
+    int num_params;
+    bool result;
+    unsigned int result_pos;
+  } test_data[] = {
+      // Empty strings.
+      STR_IN_TEST_CASE("", "Tj", 1, false, 0),
+      STR_IN_TEST_CASE("", "", 1, false, 0),
+      // Empty token.
+      STR_IN_TEST_CASE("  T j", "", 1, false, 5),
+      // No parameter.
+      STR_IN_TEST_CASE("Tj", "Tj", 1, false, 2),
+      STR_IN_TEST_CASE("(Tj", "Tj", 1, false, 3),
+      // Partial token match.
+      STR_IN_TEST_CASE("\r12\t34  56 78Tj", "Tj", 1, false, 15),
+      // Regular cases with various parameters.
+      STR_IN_TEST_CASE("\r\0abd Tj", "Tj", 1, true, 0),
+      STR_IN_TEST_CASE("12 4 Tj 3 46 Tj", "Tj", 1, true, 2),
+      STR_IN_TEST_CASE("er^ 2 (34) (5667) Tj", "Tj", 2, true, 5),
+      STR_IN_TEST_CASE("<344> (232)\t343.4\n12 45 Tj", "Tj", 3, true, 11),
+      STR_IN_TEST_CASE("1 2 3 4 5 6 7 8 cm", "cm", 6, true, 3),
+  };
+  for (size_t i = 0; i < FX_ArraySize(test_data); ++i) {
+    const FindTagTestStruct& data = test_data[i];
+    CPDF_SimpleParser parser(data.input, data.input_size);
+    EXPECT_EQ(data.result,
+              parser.FindTagParamFromStart(data.token, data.num_params))
+        << " for case " << i;
+    EXPECT_EQ(data.result_pos, parser.GetCurPos()) << " for case " << i;
+  }
+}