BOOST XML解析相同Tag子节点
主要用到了BOOST_FOREACH宏,对ptree进行遍历:
int SecSrcmConf::readFromXML() {
ifstream file;
file.open(m_confFilePath.c_str(), ios::in);
if (!file.is_open()) {
cerr << "File " << m_confFilePath << " open failed" << endl;
return -1;
}
file.close();
ptree pt;
read_xml(m_confFilePath, pt, xml_parser::trim_whitespace);
parseSystemInfoFromXML(pt.get_child("SRCM.System"));
parseMMIInfoFromXML(pt.get_child("SRCM.MMI"));
BOOST_FOREACH (ptree::value_type &v, pt.get_child("SRCM")) {
string tmp;
if (v.first == "Substation") {
SrcmSubConf subConf;
subConf.m_subName = v.second.get<string>("Name");
tmp = v.second.get<string>("Protocol");
if (tmp.length()) {
if (tmp == "ext104" || tmp == "EXT104") {
subConf.m_protocolType = PROTOCOL_EXT104;
} else if (tmp == "iec61850" || tmp == "IEC61850") {
subConf.m_protocolType = PROTOCOL_IEC61850;
} else {
subConf.m_protocolType = PROTOCOL_NULL;
}
}
tmp = v.second.get<string>("Compress");
if (tmp.length()) {
if (tmp == "ZIP" || tmp == "zip") {
subConf.m_compressType = COMPRESS_ZIP;
} else if (tmp == "GZIP" || tmp == "gzip" ||
tmp == "GZ" || tmp == "gz") {
subConf.m_compressType = COMPRESS_GZIP;
} else {
subConf.m_compressType = COMPRESS_NULL;
}
}
subConf.m_cimLastUpdate = str2time(
v.second.get<string>("DeviceCIMFile.LastUpdateTime"));
subConf.m_cimLastImport = str2time(
v.second.get<string>("DeviceCIMFile.LastImportTime"));
subConf.m_svgLastUpdate = str2time(
v.second.get<string>("DeviceSVGFile.LastUpdateTime"));
subConf.m_svgLastImport = str2time(
v.second.get<string>("DeviceSVGFile.LastImportTime"));
subConf.m_mapLastUpdate = str2time(
v.second.get<string>("DeviceMAPFile.LastUpdateTime"));
subConf.m_mapLastImport = str2time(
v.second.get<string>("DeviceMAPFile.LastImportTime"));
m_subConfMap.insert(pair<string, SrcmSubConf>(subConf.m_subName, subConf));
}
}
return 0;
}
BOOST XML解析中文问题
ptree解析主要的xml文件的格式是utf-8格式的,如果xml文件中含有unicode如中文字符,解析出来就是乱码。解析unicode要用wptree,该类的接口均支持宽字符并且接口和ptree保持一致。要支持中文解析仅仅wptree还不够,还需要一个unicode转换器的帮助,该转换器可以实现宽字符和窄字符的转换,宽窄的互相转换函数有很多实现,不过c++11中有更简单统一的方式实现宽窄字符的转换。
c++11中宽窄字符的转换:
std::wstring_convert<std::codecvt<wchar_t,char,std::mbstate_t>> conv
(newstd::codecvt<wchar_t,char,std::mbstate_t>("CHS"));
//宽字符转为窄字符
string str = conv.to_bytes(L"你好");
//窄字符转为宽字符
string wstr = conv.from_bytes(str);
boost.property_tree在解析含中文的xml文件时,需要先将该文件转换一下。boost解决方法:
#include "boost/program_options/detail/utf8_codecvt_facet.hpp"
void ParseChn()
{
std::wifstream f(fileName);
std::locale utf8Locale(std::locale(), new boost::program_options::detail::utf8_codecvt_facet());
f.imbue(utf8Locale); //先转换一下
//用wptree去解析
property_tree::wptree ptree;
property_tree::read_xml(f, ptree);
}
这种方法有个缺点就是要引入boost的libboost_program_options库,该库有二十多M,仅仅是为了解决一个中文问题,却要搞得这么麻烦,有点得不偿失。好在c++11提供更简单的方式,用c++11可以这样:
void Init(const wstring& fileName, wptree& ptree)
{
std::wifstream f(fileName);
std::locale utf8Locale(std::locale(), new std::codecvt_utf8<wchar_t>);
f.imbue(utf8Locale); //先转换一下
//用wptree去解析
property_tree::read_xml(f, ptree);
}
用c++11就不需要再引入boost的libboost_program_options库了,很简单。另外一种方法就是,仍然用ptree和string,只是在取出string字符串后,做一个转换为unicode的转换,就能得到中文字符串了。例如:
auto child = item.second.get_child("Scenes.Scene");
auto oname = child.get_optional<string>("<xmlattr>.Name");
//oname内部存了一个unicode字符串,需要将其转换为宽字符串得到中文
std::wstring_convert<std::codecvt_utf8_utf16<wchar_t>> converter;
std::wstring wide = converter.from_bytes(*oname);
//宽字符串转为窄字符串
//std::string narrow = converter.to_bytes(L"foo");
BOOST 写入XML文件格式问题
主要使用到了boost::property_tree::xml_writer_settings。如果编译器用到了C++11,用下面的写法表示缩进两个空格:
boost::property_tree::xml_writer_settings<char> settings(' ',2);
boost::property_tree::write_xml(filename, pt,std::local(),settings);
如果没有C++11特性,那么使用xml_writer_make_settings方法:
boost::property_tree::xml_writer_settings<string> settings =
bp::xml_writer_make_settings<string>(' ', 2);
boost::property_tree::write_xml(xml_path, root, locale(), settings);