// on_load_text_2
int TextHandler::convert_txt_to_text(const string & filepath, string & text)
{
ByteBuffer data;
if(false == read_file(filepath, data) )
{
return ERR_READ_TXT_INPUT_FILE_FAILED;
}
data.String(text);
return SUCCESS;
}
int TextHandler::convert_word_to_text(const string & filepath, const TextID & tid, const string & file_type, string & text)
{
string cmd = "java -jar ./mdata/word/WordDocumentExtractor.jar parse -t \"" + file_type + "\" -i \"" + filepath + "\" -m \"" + m_word_workspace + "/" + tid + ".word.tmp\" -x \"" + m_word_workspace + "/" + tid + ".word.xml\"";
system(cmd.c_str());
//读取原文件
string word_ext_result_file = m_word_workspace + "/" + tid + ".word.xml";
ByteBuffer data;
if(false == read_file(word_ext_result_file, data) )
{
return ERR_READ_WORD_EXT_FILE_FAILED;
}
//解析抽取出的XML
stringstream ss;
TiXmlDocument xmldoc;
xmldoc.Parse(data.String().c_str());
TiXmlHandle docHandle( &xmldoc );
try
{
size_t idx = 0;
TiXmlElement * elem = docHandle.FirstChild("document").Child("p", idx).ToElement();
while(elem)
{
const char * tmp = elem->GetText();
if(tmp)
ss << tmp << endl;
else
ss << endl;
++idx;
elem = docHandle.FirstChild("document").Child("p", idx).ToElement();
}
}catch (...)
{
return ERR_PARSE_WORD_XML;
}
//清理XML资源
xmldoc.Clear();
text = ss.str();
return SUCCESS;
}
// 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;
}