int TextHandler::convert_pdf_to_text(const string & filepath, const TextID & tid, string & text)
{
//读取原文件
ByteBuffer data;
if(false == read_file(filepath, data) )
{
return ERR_READ_PDF_INPUT_FILE_FAILED;
}
//生成source文件
string source_file_path = m_default_file_path + tid + PDF_SOURCE_POSTFIX;
if(false == write_file(source_file_path, data.GetPtr(), data.GetLength()) )
{
return ERR_WRITE_PDF_SOURCE_FILE;
}
string pdf_ext_filepath = m_default_file_path + tid + PDF_EXT_TXT_POSTFIX;
//call xpdf
string cmd = "./mdata/pdf/pdftotext -cfg ./mdata/pdf/xpdf-chinese-simplified/xpdfrc -enc GBK " + source_file_path + " " + pdf_ext_filepath;
system(cmd.c_str());
//读取抽取的文件
ByteBuffer tmp_data;
if(false == read_file(pdf_ext_filepath, tmp_data))
{
return ERR_READ_PDF_EXT_FILE_FAILED;
}
tmp_data.String(text);
return SUCCESS;
}
static void TestMD5FromString(const char* value, const char* expectedBase64Hash)
{
Aws::String source(value);
ByteBuffer digest = HashingUtils::CalculateMD5(source);
ASSERT_EQ(16uL, digest.GetLength());
Aws::String base64Hash = HashingUtils::Base64Encode(digest);
ASSERT_STREQ(expectedBase64Hash, base64Hash.c_str());
}
HashResult Sha256HMACCommonCryptoImpl::Calculate(const ByteBuffer& toSign, const ByteBuffer& secret)
{
unsigned int length = CC_SHA256_DIGEST_LENGTH;
ByteBuffer digest(length);
std::memset(digest.GetUnderlyingData(), 0, length);
CCHmac(kCCHmacAlgSHA256, secret.GetUnderlyingData(), secret.GetLength(), toSign.GetUnderlyingData(), toSign.GetLength(), digest.GetUnderlyingData());
return HashResult(std::move(digest));
}
TEST(HashingUtilsTest, TestSHA256FromString)
{
AWS_BEGIN_MEMORY_TEST(16, 10)
Aws::String toHash = "TestToHash";
ByteBuffer digest = HashingUtils::CalculateSHA256(toHash);
ASSERT_EQ(32uL, digest.GetLength());
Aws::String base64Hash = HashingUtils::Base64Encode(digest);
EXPECT_STREQ("No9GqyFhBA5QWj9+YUchjN83IByaCH5Lqji0McSOKyg=", base64Hash.c_str());
AWS_END_MEMORY_TEST
}
TEST(HashingUtilsTest, TestSHA256HMAC)
{
const char* toHash = "TestHash";
const char* secret = "TestSecret";
AWS_BEGIN_MEMORY_TEST(16, 10)
ByteBuffer digest = HashingUtils::CalculateSHA256HMAC(
ByteBuffer((unsigned char*) toHash, 8), ByteBuffer((unsigned char*) secret, 10));
Aws::String computedHashAsHex = HashingUtils::HexEncode(digest);
ASSERT_EQ(32uL, digest.GetLength());
EXPECT_STREQ("43cf04fa24b873a456670d34ef9af2cb7870483327b5767509336fa66fb7986c", computedHashAsHex.c_str());
AWS_END_MEMORY_TEST
}
TEST(UUIDTest, TestPlatformGeneratesUUID)
{
Aws::Set<Aws::String> generatedUUids;
for(size_t i = 0u; i < 1000u; ++i)
{
UUID uuid = UUID::RandomUUID();
Aws::String uuidStr = uuid;
ASSERT_EQ(36u, uuidStr.length());
ByteBuffer rawUUID = uuid;
ASSERT_EQ(16u, rawUUID.GetLength());
ASSERT_EQ(0x40u, 0x40u & rawUUID[6]);
ASSERT_EQ(0x80u, 0x80u & rawUUID[8]);
ASSERT_EQ(generatedUUids.end(), generatedUUids.find(uuidStr));
generatedUUids.insert(uuidStr);
}
}
int TextHandler::convert_html_to_text(const string & filepath, const TextID & tid, string & text)
{
//读取原文件
ByteBuffer data;
if(false == read_file(filepath, data) )
{
return ERR_READ_HTML_INPUT_FILE_FAILED;
}
//生成source文件
string source_file_path = m_default_file_path + tid + HTML_SOURCE_POSTFIX;
if(false == write_file(source_file_path, data.GetPtr(), data.GetLength()) )
{
return ERR_WRITE_HTML_SOURCE_FILE;
}
string html_ext_filepath = m_default_file_path + tid + HTML_EXT_TXT_POSTFIX;
string html_info_filepath = m_default_file_path + tid + HTML_INFO_POSTFIX;
//call html parser
string cmd = "java -jar ./mdata/html/htmlAnalysis.jar -g \"" + source_file_path + "\" \"" + html_ext_filepath + "\" \"" + html_info_filepath + "\"";
system(cmd.c_str());
//读取抽取的文件
ByteBuffer tmp_data;
if(false == read_file(html_ext_filepath, tmp_data))
{
return ERR_READ_HTML_EXT_FILE_FAILED;
}
tmp_data.String(text);
return SUCCESS;
}
ByteBuffer BytecodeHeader::GetByteBuffer()
{
ByteBuffer buffer;
auto bodyOffsetReservation = buffer.Reserve(4);
// Write string constants
buffer.WriteU32(constStringTable->constants.size());
for (auto &key : constStringTable->constants)
buffer.WriteString(key.val);
// Write numeric constants
buffer.WriteU32(constNumberTable->constants.size());
for (auto &key : constNumberTable->constants)
buffer.WriteU32(*reinterpret_cast<uint32_t *>(&key.val));
buffer.WriteU32(buffer.GetLength(), bodyOffsetReservation);
// Write function names and offsets
return buffer;
}
HashResult BCryptHashImpl::Calculate(const ByteBuffer& toHash, const ByteBuffer& secret)
{
if (!IsValid())
{
return HashResult();
}
std::lock_guard<std::mutex> locker(m_algorithmMutex);
BCryptHashContext context(m_algorithmHandle, m_hashObject, m_hashObjectLength, secret);
if (!context.IsValid())
{
AWS_LOG_ERROR(logTag, "Error creating hash handle.");
return HashResult();
}
return HashData(context, static_cast<PBYTE>(toHash.GetUnderlyingData()), static_cast<ULONG>(toHash.GetLength()));
}
// tool functions for build result
int TextHandler::build_base_result_file(const TextID & tid,
const string & build_id,
const vector<boost::tuple<string, string, string, size_t>> & result_vec,
const bool is_blank_between_sent,
const bool is_bilingual,
string & base_result_file_path,
string & result_text)
{
//读取base source文件
ByteBuffer source;
string source_file_path = m_default_file_path + tid + BASE_TEXT_SOURCE_POSTFIX;
if(false == read_file(source_file_path, source))
{
return ERR_READ_BASE_SOURCE_FILE;
}
//读取base pos文件
ByteBuffer pos_data;
string pos_file_path = m_default_file_path + tid + BASE_TEXT_POS_POSTFIX;
if(false == read_file(pos_file_path, pos_data) )
{
return ERR_READ_BASE_POS_FILE;
}
//生成结果文件
TextFormat format;
if(false == format.UnSerialization(pos_data.String()) )
{
return ERR_PARSE_RESULT_FORMAT;
}
//为每个段落生成tgt
string curr_para_tgt;
vector<string> para_tgt_vec;
para_tgt_vec.resize(format.Size(), "");
size_t i = 0;
for(i=0; i<result_vec.size(); ++i)
{
string final_tgt;
if(result_vec[i].get<2>().size() > 0)
final_tgt += result_vec[i].get<2>();
else
final_tgt += result_vec[i].get<1>();
if(result_vec[i].get<3>() >= format.Size())
{
lerr << "formate restor failed. para_size = " << format.Size() << " result_vec[" << i << "].para_idx = " << result_vec[i].get<2>() << endl;
continue;
}
if(i+1 < result_vec.size() && is_blank_between_sent)
para_tgt_vec[result_vec[i].get<3>()] += final_tgt + " ";
else
para_tgt_vec[result_vec[i].get<3>()] += final_tgt;
}
//重新生成原文
result_text.clear();
size_t para_idx = 0;
size_t offset = 0;
while(offset < source.GetLength())
{
if(para_idx < format.Size())
{
size_t len = format[para_idx]._offset - offset;
//输出非翻译段落
if(len > 0)
{
result_text += source.String().substr(offset, len);
}
//输出翻译段落
if(is_bilingual)
result_text += source.String().substr(format[para_idx]._offset, format[para_idx]._len) + " [";
result_text += para_tgt_vec[para_idx];
if(is_bilingual)
result_text += "] ";
//更新偏移量
offset = format[para_idx]._offset + format[para_idx]._len;
++para_idx;
}
else
{
size_t len = source.GetLength() - offset;
//输出非翻译段落
if(len > 0)
{
result_text += source.String().substr(offset, len);
}
break;
}
}
//输出结果到文件
base_result_file_path = m_default_file_path + tid + "." + build_id + BASE_TEXT_TARGET_POSTFIX;
if(false == write_file(base_result_file_path, result_text.c_str(), result_text.size()) )
{
return ERR_WRITE_BASE_RESULT_FILE;
}
return SUCCESS;
}
BCryptHashContext(void* algorithmHandle, PBYTE hashObject, DWORD hashObjectLength, const ByteBuffer& secret) :
m_hashHandle(nullptr),
m_isValid(false)
{
NTSTATUS status = BCryptCreateHash(algorithmHandle, &m_hashHandle, hashObject, hashObjectLength, secret.GetUnderlyingData(), (ULONG)secret.GetLength(), 0);
m_isValid = NT_SUCCESS(status);
}
