Merge to XFA: Simplify UTF16LE_Encode and add unittest.
Previously, UTF16LE_Encode take an optional flag to indicate
if the returned byte string has trailing zeros. In fact, no where
needs the flag to be false. So just get rid of it so callers won't
misuse.
The bug is found by https://codereview.chromium.org/837723009
R=tsepez@chromium.org
Review URL: https://codereview.chromium.org/860973002
diff --git a/BUILD.gn b/BUILD.gn
index 6324e85..bd3d3bc 100644
--- a/BUILD.gn
+++ b/BUILD.gn
@@ -1546,6 +1546,7 @@
test("pdfium_unittests") {
sources = [
"core/src/fxcrt/fx_basic_bstring_unittest.cpp",
+ "core/src/fxcrt/fx_basic_wstring_unittest.cpp",
"testing/fx_string_testhelpers.cpp",
"testing/fx_string_testhelpers.h",
"xfa/src/fxbarcode/pdf417/BC_PDF417HighLevelEncoder_unittest.cpp",
diff --git a/core/include/fxcrt/fx_string.h b/core/include/fxcrt/fx_string.h
index eae1c15..33634f8 100644
--- a/core/include/fxcrt/fx_string.h
+++ b/core/include/fxcrt/fx_string.h
@@ -774,7 +774,7 @@
CFX_ByteString UTF8Encode() const;
- CFX_ByteString UTF16LE_Encode(FX_BOOL bTerminate = TRUE) const;
+ CFX_ByteString UTF16LE_Encode() const;
void ConvertFrom(const CFX_ByteString& str, CFX_CharMap* pCharMap = NULL);
protected:
diff --git a/core/src/fxcrt/fx_basic_wstring.cpp b/core/src/fxcrt/fx_basic_wstring.cpp
index 0827fb6..0945d2f 100644
--- a/core/src/fxcrt/fx_basic_wstring.cpp
+++ b/core/src/fxcrt/fx_basic_wstring.cpp
@@ -323,25 +323,21 @@
{
return FX_UTF8Encode(*this);
}
-CFX_ByteString CFX_WideString::UTF16LE_Encode(FX_BOOL bTerminate) const
+CFX_ByteString CFX_WideString::UTF16LE_Encode() const
{
if (m_pData == NULL) {
- return bTerminate ? CFX_ByteString(FX_BSTRC("\0\0")) : CFX_ByteString();
+ return CFX_ByteString(FX_BSTRC("\0\0"));
}
int len = m_pData->m_nDataLength;
CFX_ByteString result;
- FX_LPSTR buffer = result.GetBuffer(len * 2 + (bTerminate ? 2 : 0));
+ FX_LPSTR buffer = result.GetBuffer(len * 2 + 2);
for (int i = 0; i < len; i ++) {
buffer[i * 2] = m_pData->m_String[i] & 0xff;
buffer[i * 2 + 1] = m_pData->m_String[i] >> 8;
}
- if (bTerminate) {
- buffer[len * 2] = 0;
- buffer[len * 2 + 1] = 0;
- result.ReleaseBuffer(len * 2 + 2);
- } else {
- result.ReleaseBuffer(len * 2);
- }
+ buffer[len * 2] = 0;
+ buffer[len * 2 + 1] = 0;
+ result.ReleaseBuffer(len * 2 + 2);
return result;
}
void CFX_WideString::ConvertFrom(const CFX_ByteString& str, CFX_CharMap* pCharMap)
diff --git a/core/src/fxcrt/fx_basic_wstring_unittest.cpp b/core/src/fxcrt/fx_basic_wstring_unittest.cpp
new file mode 100644
index 0000000..c6e5c2f
--- /dev/null
+++ b/core/src/fxcrt/fx_basic_wstring_unittest.cpp
@@ -0,0 +1,29 @@
+// Copyright 2014 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 "testing/gtest/include/gtest/gtest.h"
+#include "../../../testing/fx_string_testhelpers.h"
+#include "../../include/fxcrt/fx_basic.h"
+
+#define ByteStringLiteral(str) CFX_ByteString(FX_BSTRC(str))
+
+TEST(fxcrt, WideStringUTF16LE_Encode) {
+ struct UTF16LEEncodeCase {
+ CFX_WideString ws;
+ CFX_ByteString bs;
+ } utf16le_encode_cases[] = {
+ { L"", ByteStringLiteral("\0\0") },
+ { L"abc", ByteStringLiteral("a\0b\0c\0\0\0") },
+ { L"abcdef", ByteStringLiteral("a\0b\0c\0d\0e\0f\0\0\0") },
+ { L"abc\0def", ByteStringLiteral("a\0b\0c\0\0\0") },
+ { L"\xaabb\xccdd", ByteStringLiteral("\xbb\xaa\xdd\xcc\0\0") },
+ { L"\x3132\x6162", ByteStringLiteral("\x32\x31\x62\x61\0\0") },
+ };
+
+ for (size_t i = 0; i < FX_ArraySize(utf16le_encode_cases); ++i) {
+ EXPECT_EQ(utf16le_encode_cases[i].bs,
+ utf16le_encode_cases[i].ws.UTF16LE_Encode())
+ << " for case number " << i;
+ }
+}
diff --git a/fpdfsdk/src/fpdfdoc.cpp b/fpdfsdk/src/fpdfdoc.cpp
index 23c3c2e..19b4581 100644
--- a/fpdfsdk/src/fpdfdoc.cpp
+++ b/fpdfsdk/src/fpdfdoc.cpp
@@ -55,14 +55,12 @@
return 0;
CPDF_Bookmark bookmark((CPDF_Dictionary*)pDict);
CFX_WideString title = bookmark.GetTitle();
- CFX_ByteString encodedTitle = title.UTF16LE_Encode(FALSE);
+ CFX_ByteString encodedTitle = title.UTF16LE_Encode();
unsigned long len = encodedTitle.GetLength();
- if (buffer && buflen >= len + 2) {
+ if (buffer && buflen >= len) {
FXSYS_memcpy(buffer, encodedTitle.c_str(), len);
- ((FX_BYTE*)buffer)[len] = 0;
- ((FX_BYTE*)buffer)[len + 1] = 0;
}
- return len + 2;
+ return len;
}
DLLEXPORT FPDF_BOOKMARK STDCALL FPDFBookmark_Find(FPDF_DOCUMENT document, FPDF_WIDESTRING title)
@@ -296,11 +294,8 @@
// Use UTF-16LE encoding
CFX_ByteString encodedText = text.UTF16LE_Encode();
unsigned long len = encodedText.GetLength();
- if (buffer && buflen >= len + 2) {
+ if (buffer && buflen >= len) {
FXSYS_memcpy(buffer, encodedText.c_str(), len);
- // use double zero as trailer
- ((FX_BYTE*)buffer)[len] = 0;
- ((FX_BYTE*)buffer)[len + 1] = 0;
}
- return len+2;
+ return len;
}
diff --git a/fpdfsdk/src/fpdfview.cpp b/fpdfsdk/src/fpdfview.cpp
index 291b9f6..f084ce0 100644
--- a/fpdfsdk/src/fpdfview.cpp
+++ b/fpdfsdk/src/fpdfview.cpp
@@ -1055,11 +1055,9 @@
CFX_ByteString utf16Name = wsName.UTF16LE_Encode();
unsigned int len = utf16Name.GetLength();
if (!buffer) {
- buflen = len + 2;
- } else if (buflen >= len + 2) {
+ buflen = len;
+ } else if (buflen >= len) {
memcpy(buffer, utf16Name.c_str(), len);
- ((FX_BYTE*)buffer)[len] = 0;
- ((FX_BYTE*)buffer)[len + 1] = 0;
} else {
len = -1;
}
diff --git a/pdfium.gyp b/pdfium.gyp
index 6b8214e..3fba4ec 100644
--- a/pdfium.gyp
+++ b/pdfium.gyp
@@ -905,6 +905,7 @@
'testing/fx_string_testhelpers.h',
'testing/fx_string_testhelpers.cpp',
'core/src/fxcrt/fx_basic_bstring_unittest.cpp',
+ 'core/src/fxcrt/fx_basic_wstring_unittest.cpp',
'xfa/src/fxbarcode/pdf417/BC_PDF417HighLevelEncoder_unittest.cpp',
],
},
