Introduce FX_Number class as a replacement for FX_atonum().

The issue with FX_atonum() is that it doesn't return any information
about whether it range-checked its integer values as a signed or
unsigned type, even though it knows this as part of its processing.

Rather than adding another out parameter to that function, create
a class to hold all this information together.

This is the first place things went astray while diagnosing
bug 882959, in that a large positive value was cast to float as a
negative value. Unfortunately, this doesn't affect the related bug,
but is a step in the right direction.

Change-Id: I0977ec8fccf85e2632a962507bdd30a1cbe6d33c
Reviewed-on: https://pdfium-review.googlesource.com/42353
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Tom Sepez <tsepez@chromium.org>

6 files changed, 310 insertions, 111 deletions

diff --git a/core/fxcrt/fx_number.cpp b/core/fxcrt/fx_number.cpp
new file mode 100644
index 0000000000..68d5bd9b32
--- /dev/null
+++ b/core/fxcrt/fx_number.cpp
@@ -0,0 +1,104 @@
+// Copyright 2018 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 "core/fxcrt/fx_number.h"
+
+#include <limits>
+
+#include "core/fxcrt/fx_extension.h"
+#include "core/fxcrt/fx_string.h"
+
+FX_Number::FX_Number()
+    : m_bInteger(true), m_bSigned(false), m_UnsignedValue(0) {}
+
+FX_Number::FX_Number(uint32_t value)
+    : m_bInteger(true), m_bSigned(false), m_UnsignedValue(value) {}
+
+FX_Number::FX_Number(int32_t value)
+    : m_bInteger(true), m_bSigned(true), m_SignedValue(value) {}
+
+FX_Number::FX_Number(float value)
+    : m_bInteger(false), m_bSigned(true), m_FloatValue(value) {}
+
+FX_Number::FX_Number(const ByteStringView& strc)
+    : m_bInteger(true), m_bSigned(false), m_UnsignedValue(0) {
+  if (strc.IsEmpty())
+    return;
+
+  if (strc.Contains('.')) {
+    m_bInteger = false;
+    m_bSigned = true;
+    m_FloatValue = FX_atof(strc);
+    return;
+  }
+
+  // Note, numbers in PDF are typically of the form 123, -123, etc. But,
+  // for things like the Permissions on the encryption hash the number is
+  // actually an unsigned value. We use a uint32_t so we can deal with the
+  // unsigned and then check for overflow if the user actually signed the value.
+  // The Permissions flag is listed in Table 3.20 PDF 1.7 spec.
+  pdfium::base::CheckedNumeric<uint32_t> unsigned_val = 0;
+  bool bNegative = false;
+  size_t cc = 0;
+  if (strc[0] == '+') {
+    cc++;
+    m_bSigned = true;
+  } else if (strc[0] == '-') {
+    bNegative = true;
+    m_bSigned = true;
+    cc++;
+  }
+
+  while (cc < strc.GetLength() && std::isdigit(strc[cc])) {
+    unsigned_val = unsigned_val * 10 + FXSYS_DecimalCharToInt(strc.CharAt(cc));
+    if (!unsigned_val.IsValid())
+      break;
+    cc++;
+  }
+
+  uint32_t uValue = unsigned_val.ValueOrDefault(0);
+  if (!m_bSigned) {
+    m_UnsignedValue = uValue;
+    return;
+  }
+
+  // We have a sign, so if the value was greater then the signed integer
+  // limits, then we've overflowed and must reset to the default value.
+  constexpr uint32_t uLimit =
+      static_cast<uint32_t>(std::numeric_limits<int>::max());
+
+  if (uValue > (bNegative ? uLimit + 1 : uLimit))
+    uValue = 0;
+
+  // Switch back to the int space so we can flip to a negative if we need.
+  int32_t value = static_cast<int32_t>(uValue);
+  if (bNegative) {
+    // |value| is usually positive, except in the corner case of "-2147483648",
+    // where |uValue| is 2147483648. When it gets casted to an int, |value|
+    // becomes -2147483648. For this case, avoid undefined behavior, because
+    // an int32_t cannot represent 2147483648.
+    static constexpr int kMinInt = std::numeric_limits<int>::min();
+    m_SignedValue = LIKELY(value != kMinInt) ? -value : kMinInt;
+  } else {
+    m_SignedValue = value;
+  }
+}
+
+uint32_t FX_Number::GetUnsigned() const {
+  return m_bInteger ? m_UnsignedValue : static_cast<uint32_t>(m_FloatValue);
+}
+
+int32_t FX_Number::GetSigned() const {
+  return m_bInteger ? m_SignedValue : static_cast<int32_t>(m_FloatValue);
+}
+
+float FX_Number::GetFloat() const {
+  if (!m_bInteger)
+    return m_FloatValue;
+
+  return m_bSigned ? static_cast<float>(m_SignedValue)
+                   : static_cast<float>(m_UnsignedValue);
+}
diff --git a/core/fxcrt/fx_number.h b/core/fxcrt/fx_number.h
new file mode 100644
index 0000000000..4c789b4130
--- /dev/null
+++ b/core/fxcrt/fx_number.h
@@ -0,0 +1,39 @@
+// Copyright 2018 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_FXCRT_FX_NUMBER_H_
+#define CORE_FXCRT_FX_NUMBER_H_
+
+#include <stdint.h>
+
+#include "core/fxcrt/bytestring.h"
+
+class FX_Number {
+ public:
+  FX_Number();
+  explicit FX_Number(uint32_t value);
+  explicit FX_Number(int32_t value);
+  explicit FX_Number(float value);
+  explicit FX_Number(const ByteStringView& str);
+
+  bool IsInteger() const { return m_bInteger; }
+  bool IsSigned() const { return m_bSigned; }
+
+  uint32_t GetUnsigned() const;  // Underflow possible.
+  int32_t GetSigned() const;     // Underflow/Overflow possible.
+  float GetFloat() const;
+
+ private:
+  bool m_bInteger;  // One of the two integers vs. float type.
+  bool m_bSigned;   // Only valid if |m_bInteger|.
+  union {
+    uint32_t m_UnsignedValue;
+    int32_t m_SignedValue;
+    float m_FloatValue;
+  };
+};
+
+#endif  // CORE_FXCRT_FX_NUMBER_H_
diff --git a/core/fxcrt/fx_number_unittest.cpp b/core/fxcrt/fx_number_unittest.cpp
new file mode 100644
index 0000000000..83702dadcf
--- /dev/null
+++ b/core/fxcrt/fx_number_unittest.cpp
@@ -0,0 +1,165 @@
+// Copyright 2018 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 <limits>
+
+#include "core/fxcrt/fx_number.h"
+#include "testing/gtest/include/gtest/gtest.h"
+
+TEST(fxnumber, Default) {
+  FX_Number number;
+  EXPECT_TRUE(number.IsInteger());
+  EXPECT_FALSE(number.IsSigned());
+  EXPECT_EQ(0u, number.GetUnsigned());
+  EXPECT_EQ(0, number.GetSigned());
+  EXPECT_FLOAT_EQ(0.0f, number.GetFloat());
+}
+
+TEST(fxnumber, FromUnsigned) {
+  FX_Number number(42u);
+  EXPECT_TRUE(number.IsInteger());
+  EXPECT_FALSE(number.IsSigned());
+  EXPECT_EQ(42u, number.GetUnsigned());
+  EXPECT_EQ(42, number.GetSigned());
+  EXPECT_FLOAT_EQ(42.0f, number.GetFloat());
+
+  // Show that assignment works.
+  FX_Number number2 = number;
+  EXPECT_TRUE(number2.IsInteger());
+  EXPECT_FALSE(number2.IsSigned());
+  EXPECT_EQ(42u, number2.GetUnsigned());
+  EXPECT_EQ(42, number2.GetSigned());
+  EXPECT_FLOAT_EQ(42.0F, number2.GetFloat());
+}
+
+TEST(fxnumber, FromSigned) {
+  FX_Number number(-128);
+  EXPECT_TRUE(number.IsInteger());
+  EXPECT_TRUE(number.IsSigned());
+  EXPECT_EQ(4294967168u, number.GetUnsigned());
+  EXPECT_EQ(-128, number.GetSigned());
+  EXPECT_FLOAT_EQ(-128.0f, number.GetFloat());
+
+  // Show that assignment works.
+  FX_Number number2 = number;
+  EXPECT_TRUE(number2.IsInteger());
+  EXPECT_TRUE(number2.IsSigned());
+  EXPECT_EQ(4294967168u, number2.GetUnsigned());
+  EXPECT_EQ(-128, number2.GetSigned());
+  EXPECT_FLOAT_EQ(-128.0f, number2.GetFloat());
+}
+
+TEST(fxnumber, FromFloat) {
+  FX_Number number(-100.001f);
+  EXPECT_FALSE(number.IsInteger());
+  EXPECT_TRUE(number.IsSigned());
+  EXPECT_EQ(4294967196u, number.GetUnsigned());
+  EXPECT_EQ(-100, number.GetSigned());
+  EXPECT_FLOAT_EQ(-100.001f, number.GetFloat());
+
+  // Show that assignment works.
+  FX_Number number2 = number;
+  EXPECT_FALSE(number2.IsInteger());
+  EXPECT_TRUE(number2.IsSigned());
+  EXPECT_EQ(4294967196u, number2.GetUnsigned());
+  EXPECT_EQ(-100, number2.GetSigned());
+  EXPECT_FLOAT_EQ(-100.001f, number2.GetFloat());
+}
+
+TEST(fxnumber, FromStringUnsigned) {
+  {
+    FX_Number number("");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_FALSE(number.IsSigned());
+    EXPECT_EQ(0u, number.GetUnsigned());
+  }
+  {
+    FX_Number number("0");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_FALSE(number.IsSigned());
+    EXPECT_EQ(0u, number.GetUnsigned());
+  }
+  {
+    FX_Number number("10");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_FALSE(number.IsSigned());
+    EXPECT_EQ(10u, number.GetUnsigned());
+  }
+  {
+    FX_Number number("4294967295");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_FALSE(number.IsSigned());
+    EXPECT_EQ(std::numeric_limits<uint32_t>::max(), number.GetUnsigned());
+  }
+  {
+    // Value overflows.
+    FX_Number number("4223423494965252");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_FALSE(number.IsSigned());
+    EXPECT_EQ(0u, number.GetUnsigned());
+  }
+  {
+    // No explicit sign will allow the number to go negative if we retrieve
+    // it as a signed value. This is needed for things like the encryption
+    // Permissions flag (Table 3.20 PDF 1.7 spec)
+    FX_Number number("4294965252");
+    EXPECT_EQ(-2044, number.GetSigned());
+  }
+}
+
+TEST(fxnumber, FromStringSigned) {
+  {
+    FX_Number number("-0");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(0, number.GetSigned());
+  }
+  {
+    FX_Number number("+0");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(0, number.GetSigned());
+  }
+  {
+    FX_Number number("-10");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(-10, number.GetSigned());
+  }
+  {
+    FX_Number number("+10");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(10, number.GetSigned());
+  }
+  {
+    FX_Number number("-2147483648");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(std::numeric_limits<int32_t>::min(), number.GetSigned());
+  }
+  {
+    FX_Number number("+2147483647");
+    EXPECT_TRUE(number.IsInteger());
+    EXPECT_TRUE(number.IsSigned());
+    EXPECT_EQ(std::numeric_limits<int32_t>::max(), number.GetSigned());
+  }
+  {
+    // Value underflows.
+    FX_Number number("-2147483649");
+    EXPECT_EQ(0u, number.GetUnsigned());
+    EXPECT_EQ(0, number.GetSigned());
+  }
+  {
+    // Value overflows.
+    FX_Number number("+2147483648");
+    EXPECT_EQ(0u, number.GetUnsigned());
+    EXPECT_EQ(0, number.GetSigned());
+  }
+}
+
+TEST(fxnumber, FromStringFloat) {
+  FX_Number number("3.24");
+  EXPECT_FLOAT_EQ(3.24f, number.GetFloat());
+}
diff --git a/core/fxcrt/fx_string.cpp b/core/fxcrt/fx_string.cpp
index c9993f9ab8..31eb8e917d 100644
--- a/core/fxcrt/fx_string.cpp
+++ b/core/fxcrt/fx_string.cpp
@@ -47,69 +47,6 @@ float FractionalScale(size_t scale_factor, int value) {
 
 }  // namespace
 
-bool FX_atonum(const ByteStringView& strc, void* pData) {
-  if (strc.Contains('.')) {
-    float* pFloat = static_cast<float*>(pData);
-    *pFloat = FX_atof(strc);
-    return false;
-  }
-
-  // Note, numbers in PDF are typically of the form 123, -123, etc. But,
-  // for things like the Permissions on the encryption hash the number is
-  // actually an unsigned value. We use a uint32_t so we can deal with the
-  // unsigned and then check for overflow if the user actually signed the value.
-  // The Permissions flag is listed in Table 3.20 PDF 1.7 spec.
-  pdfium::base::CheckedNumeric<uint32_t> integer = 0;
-  bool bNegative = false;
-  bool bSigned = false;
-  size_t cc = 0;
-  if (strc[0] == '+') {
-    cc++;
-    bSigned = true;
-  } else if (strc[0] == '-') {
-    bNegative = true;
-    bSigned = true;
-    cc++;
-  }
-
-  while (cc < strc.GetLength() && std::isdigit(strc[cc])) {
-    integer = integer * 10 + FXSYS_DecimalCharToInt(strc.CharAt(cc));
-    if (!integer.IsValid())
-      break;
-    cc++;
-  }
-
-  // We have a sign, and the value was greater then a regular integer
-  // we've overflowed, reset to the default value.
-  if (bSigned) {
-    if (bNegative) {
-      if (integer.ValueOrDefault(0) >
-          static_cast<uint32_t>(std::numeric_limits<int>::max()) + 1) {
-        integer = 0;
-      }
-    } else if (integer.ValueOrDefault(0) >
-               static_cast<uint32_t>(std::numeric_limits<int>::max())) {
-      integer = 0;
-    }
-  }
-
-  // Switch back to the int space so we can flip to a negative if we need.
-  uint32_t uValue = integer.ValueOrDefault(0);
-  int32_t value = static_cast<int>(uValue);
-  if (bNegative) {
-    // |value| is usually positive, except in the corner case of "-2147483648",
-    // where |uValue| is 2147483648. When it gets casted to an int, |value|
-    // becomes -2147483648. For this case, avoid undefined behavior, because an
-    // integer cannot represent 2147483648.
-    static constexpr int kMinInt = std::numeric_limits<int>::min();
-    value = LIKELY(value != kMinInt) ? -value : kMinInt;
-  }
-
-  int* pInt = static_cast<int*>(pData);
-  *pInt = value;
-  return true;
-}
-
 float FX_atof(const ByteStringView& strc) {
   if (strc.IsEmpty())
     return 0.0;
diff --git a/core/fxcrt/fx_string.h b/core/fxcrt/fx_string.h
index 2cf823738a..670aa4d4b7 100644
--- a/core/fxcrt/fx_string.h
+++ b/core/fxcrt/fx_string.h
@@ -7,6 +7,8 @@
 #ifndef CORE_FXCRT_FX_STRING_H_
 #define CORE_FXCRT_FX_STRING_H_
 
+#include <stdint.h>
+
 #include "core/fxcrt/bytestring.h"
 #include "core/fxcrt/widestring.h"
 
@@ -19,7 +21,6 @@ WideString FX_UTF8Decode(const ByteStringView& bsStr);
 
 float FX_atof(const ByteStringView& str);
 float FX_atof(const WideStringView& wsStr);
-bool FX_atonum(const ByteStringView& str, void* pData);
 size_t FX_ftoa(float f, char* buf);
 
 #endif  // CORE_FXCRT_FX_STRING_H_
diff --git a/core/fxcrt/fx_string_unittest.cpp b/core/fxcrt/fx_string_unittest.cpp
index 60e7f07523..3059de3e5c 100644
--- a/core/fxcrt/fx_string_unittest.cpp
+++ b/core/fxcrt/fx_string_unittest.cpp
@@ -2,56 +2,9 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
-#include <limits>
-
 #include "core/fxcrt/fx_string.h"
 #include "testing/gtest/include/gtest/gtest.h"
 
-TEST(fxstring, FX_atonum) {
-  int i;
-  EXPECT_TRUE(FX_atonum("10", &i));
-  EXPECT_EQ(10, i);
-
-  EXPECT_TRUE(FX_atonum("-10", &i));
-  EXPECT_EQ(-10, i);
-
-  EXPECT_TRUE(FX_atonum("+10", &i));
-  EXPECT_EQ(10, i);
-
-  EXPECT_TRUE(FX_atonum("-2147483648", &i));
-  EXPECT_EQ(std::numeric_limits<int>::min(), i);
-
-  EXPECT_TRUE(FX_atonum("2147483647", &i));
-  EXPECT_EQ(2147483647, i);
-
-  // Value overflows.
-  EXPECT_TRUE(FX_atonum("-2147483649", &i));
-  EXPECT_EQ(0, i);
-
-  // Value overflows.
-  EXPECT_TRUE(FX_atonum("+2147483648", &i));
-  EXPECT_EQ(0, i);
-
-  // Value overflows.
-  EXPECT_TRUE(FX_atonum("4223423494965252", &i));
-  EXPECT_EQ(0, i);
-
-  // No explicit sign will allow the number to go negative. This is for things
-  // like the encryption Permissions flag (Table 3.20 PDF 1.7 spec)
-  EXPECT_TRUE(FX_atonum("4294965252", &i));
-  EXPECT_EQ(-2044, i);
-
-  EXPECT_TRUE(FX_atonum("-4294965252", &i));
-  EXPECT_EQ(0, i);
-
-  EXPECT_TRUE(FX_atonum("+4294965252", &i));
-  EXPECT_EQ(0, i);
-
-  float f;
-  EXPECT_FALSE(FX_atonum("3.24", &f));
-  EXPECT_FLOAT_EQ(3.24f, f);
-}
-
 TEST(fxstring, FX_UTF8Encode) {
   EXPECT_EQ("", FX_UTF8Encode(WideStringView()));
   EXPECT_EQ(