Qt中三种解析xml的方式

在下面的随笔中，我会根据xml的结构，给出Qt中解析这个xml的三种方式的代码。虽然，这个代码时通过调用Qt的函数实现的，但是，很多开源的C++解析xml的库，甚至很多其他语言解析xml的库，都和下面三种解析xml采用相同的原理，所以就算你不是学习qt，也可以大致参看一下代码，对三种解析方式有一种大致的感觉。

先给出xml如下：

<?xml version="1.0" encoding="utf-8"?>
<school>
    <teacher>
        <entry name="Job">
            <age>30</age>
            <sport>soccer</sport>
        </entry>
        <entry name="Tom">
            <age>32</age>
            <sport>swimming</sport>
        </entry>
    </teacher>
    <student>
        <entry name="Lily">
            <age>20</age>
            <sport>dancing</sport>
        </entry>
        <entry name="Keith">
            <age>21</age>
            <sport>running</sport>
        </entry>
    </student>
</school>

下面给出qt中解析xml的三种方式，通过解析xml，创建student列表和teacher列表。先给出存储的结构体和辅助函数：

#include <string>
#include <ostream>

namespace School
{

struct Teacher
{
    std::string name;
    int age;
    std::string loveSport;

    Teacher(std::string name_, int age_, std::string loveSport_)
        : name(std::move(name_)), age(age_), loveSport(std::move(loveSport_))
    {

    }
};

struct Student
{
    std::string name;
    int age;
    std::string loveSport;

    Student(std::string name_, int age_, std::string loveSport_)
        : name(std::move(name_)), age(age_), loveSport(std::move(loveSport_))
    {

    }
};

inline void print(std::ostream &out, const Teacher& teacher)
{
    out << "teacher: " << teacher.name << std::endl;
    out << "\tage: " << teacher.age << std::endl;
    out << "\tfavorite sport: " << teacher.loveSport << std::endl;
}

inline void print(std::ostream& out, const Student& student)
{
    out << "student: " << student.name << std::endl;
    out << "\tage: " << student.age << std::endl;
    out << "\tfavorite sport: " << student.loveSport << std::endl;
}

}

另外需要注意在.pro中添加

QT += xml

（1）通过QXmlStreamReader：

#include <QXmlStreamReader>
#include "schooldefine.h"

class XmlStreamReader
{
public:
    XmlStreamReader();

    bool readFile(const QString& fileName);
    void printAllMembers();

private:
    void readSchoolMembers();
    void readTeacherMembers();
    void readTeacher(const QStringRef& teacherName);
    void readStudentMembers();
    void readStudent(const QStringRef& studentName);
    void skipUnknownElement();

    QXmlStreamReader reader;

    std::vector<School::Teacher> m_teachers;
    std::vector<School::Student> m_students;
};

#include "XmlStreamReader.h"
#include <QFile>
#include <iostream>
#include <QDebug>

XmlStreamReader::XmlStreamReader()
{

}

bool XmlStreamReader::readFile(const QString &fileName)
{
    QFile file(fileName);
    if (!file.open(QFile::ReadOnly | QFile::Text))
    {
        std::cerr << "Error: Cannot read file " << qPrintable(fileName)
                  << ": " << qPrintable(file.errorString())
                  << std::endl;
        return false;
    }
    reader.setDevice(&file);

    reader.readNext();
    while (!reader.atEnd())
    {
        if (reader.isStartElement())
        {
            if (reader.name() == "school")
            {
                readSchoolMembers();
            }
            else
            {
                reader.raiseError(QObject::tr("Not a school file"));
            }
        }
        else
        {
            reader.readNext();
        }
    }

    file.close();
    if (reader.hasError())
    {
        std::cerr << "Error: Failed to parse file "
                  << qPrintable(fileName) << ": "
                  << qPrintable(reader.errorString()) << std::endl;
        return false;
    }
    else if (file.error() != QFile::NoError)
    {
        std::cerr << "Error: Cannot read file " << qPrintable(fileName)
                  << ": " << qPrintable(file.errorString())
                  << std::endl;
        return false;
    }
    return true;
}

void XmlStreamReader::printAllMembers()
{
    std::cout << "All teachers: " << std::endl;
    for (const auto& teacher : m_teachers)
    {
        School::print(std::cout, teacher);
    }
    std::cout << "All students: " << std::endl;
    for (const auto& student : m_students)
    {
        School::print(std::cout, student);
    }
}

void XmlStreamReader::readSchoolMembers()
{
    reader.readNext();
    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            if (reader.name() == "teacher")
            {
                readTeacherMembers();
            }
            else if (reader.name() == "student")
            {
                readStudentMembers();
            }
            else
            {
                skipUnknownElement();
            }
        }
        else
        {
            reader.readNext();
        }
    }
}

void XmlStreamReader::readTeacherMembers()
{
    reader.readNext();
    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            if (reader.name() == "entry")
            {
                readTeacher(reader.attributes().value("name"));
            }
            else
            {
                skipUnknownElement();
            }
        }
        else
        {
            reader.readNext();
        }
    }
}

void XmlStreamReader::readTeacher(const QStringRef& teacherName)
{
    reader.readNext();

    int age = ;
    std::string favoriteSport;

    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            if (reader.name() == "age")
            {
                age = reader.readElementText().toInt();
            }
            else if (reader.name() == "sport")
            {
                favoriteSport = reader.readElementText().toStdString();
            }
            else
            {
                skipUnknownElement();
            }
        }
        reader.readNext();
    }

    m_teachers.emplace_back(teacherName.toString().toStdString(), age, favoriteSport);
}

void XmlStreamReader::readStudentMembers()
{
    reader.readNext();
    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            if (reader.name() == "entry")
            {
                readStudent(reader.attributes().value("name"));
            }
            else
            {
                skipUnknownElement();
            }
        }
        else
        {
            reader.readNext();
        }
    }
}

void XmlStreamReader::readStudent(const QStringRef &studentName)
{
    reader.readNext();

    int age = ;
    std::string favoriteSport;

    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            if (reader.name() == "age")
            {
                age = reader.readElementText().toInt();
            }
            else if (reader.name() == "sport")
            {
                favoriteSport = reader.readElementText().toStdString();
            }
            else
            {
                skipUnknownElement();
            }
        }
        reader.readNext();
    }

    m_students.emplace_back(studentName.toString().toStdString(), age, favoriteSport);
}

void XmlStreamReader::skipUnknownElement()
{
    reader.readNext();
    while (!reader.atEnd())
    {
        if (reader.isEndElement())
        {
            reader.readNext();
            break;
        }

        if (reader.isStartElement())
        {
            skipUnknownElement();
        }
        else
        {
            reader.readNext();
        }
    }
}

（2）通过DOM方式：

#include <QString>
#include <QDomElement>
#include "schooldefine.h"

class DomParser
{
public:
    DomParser();

    bool readFile(const QString &fileName);
    void printAllMembers();

private:
    void parseSchoolMembers(const QDomElement &element);
    void parseTeacherMembers(const QDomElement &element);
    void parseStudentMembers(const QDomElement &element);
    void parseTeacher(const QDomElement &element);
    void parseStudent(const QDomElement &element);

    std::vector<School::Teacher> m_teachers;
    std::vector<School::Student> m_students;
};

#include "domparser.h"
#include <QDomDocument>
#include <QFile>
#include <iostream>

DomParser::DomParser()
{

}

bool DomParser::readFile(const QString &fileName)
{
    QFile file(fileName);
    if (!file.open(QFile::ReadOnly | QFile::Text)) {
        std::cerr << "Error: Cannot read file " << qPrintable(fileName)
                  << ": " << qPrintable(file.errorString())
                  << std::endl;
        return false;
    }

    QString errorStr;
    int errorLine;
    int errorColumn;

    QDomDocument doc;
    if (!doc.setContent(&file, false, &errorStr, &errorLine, &errorColumn))
    {
        std::cerr << "Error: Parse error at line " << errorLine << ", "
                  << "column " << errorColumn << ": "
                  << qPrintable(errorStr) << std::endl;
        return false;
    }

    QDomElement root = doc.documentElement();
    if (root.tagName() != "school")
    {
        std::cerr << "Error: Not a school file" << std::endl;
        return false;
    }

    parseSchoolMembers(root);
    return true;
}

void DomParser::printAllMembers()
{
    std::cout << "All teachers: " << std::endl;
    for (const auto& teacher : m_teachers)
    {
        School::print(std::cout, teacher);
    }
    std::cout << "All students: " << std::endl;
    for (const auto& student : m_students)
    {
        School::print(std::cout, student);
    }
}

void DomParser::parseSchoolMembers(const QDomElement &element)
{
    QDomNode child = element.firstChild();
    while (!child.isNull())
    {
        if (child.toElement().tagName() == "teacher")
        {
            parseTeacherMembers(child.toElement());
        }
        else if (child.toElement().tagName() == "student")
        {
            parseStudentMembers(child.toElement());
        }
        child = child.nextSibling();
    }
}

void DomParser::parseTeacherMembers(const QDomElement &element)
{
    QDomNode child = element.firstChild();
    while (!child.isNull())
    {
        if (child.toElement().tagName() == "entry")
        {
            parseTeacher(child.toElement());
        }
        child = child.nextSibling();
    }
}

void DomParser::parseStudentMembers(const QDomElement &element)
{
    QDomNode child = element.firstChild();
    while (!child.isNull())
    {
        if (child.toElement().tagName() == "entry")
        {
            parseStudent(child.toElement());
        }
        child = child.nextSibling();
    }
}

void DomParser::parseTeacher(const QDomElement &element)
{
    auto children = element.childNodes();
    auto firstChild = children.at().toElement();
    auto secondChild = children.at().toElement();
    int age = firstChild.text().toInt();

    m_teachers.emplace_back(element.attribute("name").toStdString(),
                age, secondChild.text().toStdString());
}

void DomParser::parseStudent(const QDomElement &element)
{
    auto children = element.childNodes();
    auto firstChild = children.at().toElement();
    auto secondChild = children.at().toElement();
    int age = firstChild.text().toInt();

    m_students.emplace_back(element.attribute("name").toStdString(),
                age, secondChild.text().toStdString());
}

3. 采用QXmlSimpleReader方式，也就是回调函数方式：

#include <QXmlDefaultHandler>
#include "schooldefine.h"

class SaxHandler : public QXmlDefaultHandler
{
public:
    SaxHandler();

    bool readFile(const QString &fileName);
    void printAllMembers();

protected:
    bool startElement(const QString &namespaceURI,
                      const QString &localName,
                      const QString &qName,
                      const QXmlAttributes &atts) override;
    bool endElement(const QString &namespaceURL,
                    const QString &localName,
                    const QString &qName) override;
    bool characters(const QString &ch) override;
    bool fatalError(const QXmlParseException &exception) override;

private:
    bool m_isStudent = false;
    QString  m_currentContext;
    std::vector<School::Teacher> m_teachers;
    std::vector<School::Student> m_students;
};

#include "saxhandler.h"
#include <iostream>

SaxHandler::SaxHandler()
{

}

bool SaxHandler::readFile(const QString &fileName)
{
    QFile file(fileName);
    QXmlInputSource inputSource(&file);
    QXmlSimpleReader reader;
    reader.setContentHandler(this);
    reader.setErrorHandler(this);;
    return reader.parse(inputSource);
}

void SaxHandler::printAllMembers()
{
    std::cout << "All teachers: " << std::endl;
    for (const auto& teacher : m_teachers)
    {
        School::print(std::cout, teacher);
    }
    std::cout << "All students: " << std::endl;
    for (const auto& student : m_students)
    {
        School::print(std::cout, student);
    }
}

bool SaxHandler::startElement(const QString &namespaceURI,
                              const QString &localName,
                              const QString &qName,
                              const QXmlAttributes &atts)
{
    if (qName == "teacher")
    {
        m_isStudent = false;
    }
    else if (qName == "student")
    {
        m_isStudent = true;
    }
    else if (qName == "entry")
    {
        if (m_isStudent)
        {
            m_students.push_back(School::Student("", , ""));
            m_students.back().name = atts.value("name").toStdString();
        }
        else
        {
            m_teachers.push_back(School::Teacher("", , ""));
            m_teachers.back().name = atts.value("name").toStdString();
        }
    }
    else if (qName == "age")
    {
        m_currentContext.clear();
    }
    else if (qName == "sport")
    {
        m_currentContext.clear();
    }
    return true;
}

bool SaxHandler::characters(const QString &ch)
{
    m_currentContext += ch;
    return true;
}

bool SaxHandler::endElement(const QString &namespaceURL,
                            const QString &localName,
                            const QString &qName)
{
    if (qName == "age")
    {
        if (m_isStudent)
        {
            m_students.back().age = m_currentContext.toInt();
        }
        else
        {
            m_teachers.back().age = m_currentContext.toInt();
        }
    }
    else if (qName == "sport")
    {
        if (m_isStudent)
        {
            m_students.back().loveSport = m_currentContext.toStdString();
        }
        else
        {
            m_teachers.back().loveSport = m_currentContext.toStdString();
        }
    }
    m_currentContext.clear();
    return true;
}

bool SaxHandler::fatalError(const QXmlParseException &exception)
{
    std::cerr << "Parse error at line" << exception.lineNumber()
              << ", " << "column " << exception.columnNumber() << ": "
              << qPrintable(exception.message()) << std::endl;
    return false;
}

下面简单对上述三种方式予以说明：

(1) 从代码行数来看，采用DOM和QXmlSimpleReader的方式，代码行数比较少，而QXmlStreamReader代码行数较多。

(2) 从代码逻辑分析来看，采用DOM方式最容易理解，采用QXmlStreamReader的方式稍微难理解一些，而采用QXmlSimpleReader由于使用了较多的回调，引入了大量的类数据成员，使得代码会很难理解。

(3) 从内存占用来看，DOM的方式会耗费最多的内存，因为需要一次性将所有的内容构建成树，DOM和QXmlSimpleReader对内存要求都较低。

(4) 从运行时间消耗来看，DOM的消耗，可能会稍微大一些，因为DOM正常要经历2次的遍历，一次遍历构建树，一次遍历，构建自己需要的数据。而QXmlSimpleReader和QXmlStreamReader正常只需要遍历一次。

(5) 从处理异常来看，DOM和QXmlStreamReader应该会更容易一些，因为不涉及回调函数，但是对于xml来说，很多时候主要确认内容正确与否，如果错误就退出，查看xml中的错误。当然，这个也是比较重要的项。

对于我来说，因为大多数情况下，解析的xml不是很大，而且基本只涉及加载过程中，所以使用DOM的情况比较多。如果xml比较大，或者调用比较频繁，可以考虑使用QXmlStreamReader的方式。