From c2f0789bf90d8f3656abde8da8371e8975f7084e Mon Sep 17 00:00:00 2001 From: Dan Sinclair Date: Wed, 30 Aug 2017 15:46:58 -0400 Subject: Move CFX_UTF8Encoder out of fx_basic The CFX_UTF8Encoder is only used in FX_UTF8Encode(). This CL moves the class to the anonymous namespace with that method. The unused AppendStr method has been removed. Change-Id: Ie514686c4b4489bb0b0df83b7eeec14bb1876fcd Reviewed-on: https://pdfium-review.googlesource.com/12410 Reviewed-by: Tom Sepez Commit-Queue: dsinclair --- core/fxcrt/fx_basic.h | 16 -------- core/fxcrt/fx_basic_utf.cpp | 97 +++++++++++++++++++++++++-------------------- 2 files changed, 53 insertions(+), 60 deletions(-) diff --git a/core/fxcrt/fx_basic.h b/core/fxcrt/fx_basic.h index 8e3312d365..dcbb846196 100644 --- a/core/fxcrt/fx_basic.h +++ b/core/fxcrt/fx_basic.h @@ -103,22 +103,6 @@ class CFX_UTF8Decoder { CFX_WideTextBuf m_Buffer; }; -class CFX_UTF8Encoder { - public: - CFX_UTF8Encoder(); - ~CFX_UTF8Encoder(); - - void Input(wchar_t unicode); - void AppendStr(const CFX_ByteStringC& str); - - // The data returned by GetResult() is invalidated when this is modified by - // appending any data. - CFX_ByteStringC GetResult() const; - - private: - std::vector m_Buffer; -}; - template class CFX_FixedBufGrow { public: diff --git a/core/fxcrt/fx_basic_utf.cpp b/core/fxcrt/fx_basic_utf.cpp index 084553b671..1bcae61fe9 100644 --- a/core/fxcrt/fx_basic_utf.cpp +++ b/core/fxcrt/fx_basic_utf.cpp @@ -6,6 +6,59 @@ #include "core/fxcrt/fx_basic.h" +#include + +namespace { + +class CFX_UTF8Encoder { + public: + CFX_UTF8Encoder() {} + ~CFX_UTF8Encoder() {} + + void Input(wchar_t unicodeAsWchar) { + uint32_t unicode = static_cast(unicodeAsWchar); + if (unicode < 0x80) { + m_Buffer.push_back(unicode); + } else { + if (unicode >= 0x80000000) + return; + + int nbytes = 0; + if (unicode < 0x800) + nbytes = 2; + else if (unicode < 0x10000) + nbytes = 3; + else if (unicode < 0x200000) + nbytes = 4; + else if (unicode < 0x4000000) + nbytes = 5; + else + nbytes = 6; + + static uint8_t prefix[] = {0xc0, 0xe0, 0xf0, 0xf8, 0xfc}; + int order = 1 << ((nbytes - 1) * 6); + int code = unicodeAsWchar; + m_Buffer.push_back(prefix[nbytes - 2] | (code / order)); + for (int i = 0; i < nbytes - 1; i++) { + code = code % order; + order >>= 6; + m_Buffer.push_back(0x80 | (code / order)); + } + } + } + + // The data returned by GetResult() is invalidated when this is modified by + // appending any data. + CFX_ByteStringC GetResult() const { + return CFX_ByteStringC(m_Buffer.data(), m_Buffer.size()); + } + + private: + std::vector m_Buffer; +}; + +} // namespace + void CFX_UTF8Decoder::Clear() { m_Buffer.Clear(); m_PendingBytes = 0; @@ -46,50 +99,6 @@ void CFX_UTF8Decoder::Input(uint8_t byte) { } } -CFX_UTF8Encoder::CFX_UTF8Encoder() {} - -CFX_UTF8Encoder::~CFX_UTF8Encoder() {} - -void CFX_UTF8Encoder::Input(wchar_t unicodeAsWchar) { - uint32_t unicode = static_cast(unicodeAsWchar); - if (unicode < 0x80) { - m_Buffer.push_back(unicode); - } else { - if (unicode >= 0x80000000) { - return; - } - int nbytes = 0; - if (unicode < 0x800) { - nbytes = 2; - } else if (unicode < 0x10000) { - nbytes = 3; - } else if (unicode < 0x200000) { - nbytes = 4; - } else if (unicode < 0x4000000) { - nbytes = 5; - } else { - nbytes = 6; - } - static uint8_t prefix[] = {0xc0, 0xe0, 0xf0, 0xf8, 0xfc}; - int order = 1 << ((nbytes - 1) * 6); - int code = unicodeAsWchar; - m_Buffer.push_back(prefix[nbytes - 2] | (code / order)); - for (int i = 0; i < nbytes - 1; i++) { - code = code % order; - order >>= 6; - m_Buffer.push_back(0x80 | (code / order)); - } - } -} - -void CFX_UTF8Encoder::AppendStr(const CFX_ByteStringC& str) { - m_Buffer.insert(m_Buffer.end(), str.begin(), str.end()); -} - -CFX_ByteStringC CFX_UTF8Encoder::GetResult() const { - return CFX_ByteStringC(m_Buffer.data(), m_Buffer.size()); -} - CFX_ByteString FX_UTF8Encode(const CFX_WideStringC& wsStr) { FX_STRSIZE len = wsStr.GetLength(); const wchar_t* pStr = wsStr.unterminated_c_str(); -- cgit v1.2.3