XmlReader 不断读取

Question

我有一个非常大的 xml 文件。这是xml格式的简化版。

<?xml version='1.0' encoding='UTF-8'?>
<Sender>
 <SenderID>571099948</SenderID>
 <Sponsors>
  <Sponsor>
    <SponsorID>TEST01</SponsorID>
    <Contracts>
      <Contract>
        <ContractID>000001</ContractID>
        <Member>
          <SSN>1111111111</SSN>
          <Gender>M</Gender>
          <Benefits>
            <Benefit BenefitType="AAA">
            </Benefit>
            <Benefit BenefitType="BBB">
            </Benefit>
          </Benefits>
        </Member>
        <Member>
          <SSN>4444444444</SSN>
          <Gender>F</Gender>
          <Benefits>
            <Benefit BenefitType="AAA">
            </Benefit>
          </Benefits>
        </Member>
      </Contract>
      <Contract>
        <ContractID>0000002</ContractID>
        <Member>
          <SSN>2222222222</SSN>
          <Gender>F</Gender>
          <Benefits>
            <Benefit BenefitType="CCC">
            </Benefit>
            <Benefit BenefitType="DDD">
            </Benefit>
          </Benefits>
        </Member>
      </Contract>
      <Contract>
        <ContractID>0000003</ContractID>
        <Member>
          <SSN>333333333</SSN>
          <Gender>F</Gender>
          <Benefits> 
            <Benefit BenefitType="CCC">
            </Benefit>
          </Benefits>
        </Member>
      </Contract>
    </Contracts>
  </Sponsor>
  <Sponsor>
    <SponsorID>TEST02</SponsorID>
    <Contracts>
      <Contract>
        <ContractID>0000011</ContractID>
        <Member>
          <SSN>1111111111</SSN>
          <Gender>M</Gender>
          <Benefits>
          </Benefits>
        </Member>
      </Contract>
      <Contract>
        <ContractID>0000002</ContractID>
        <Member>
          <SSN>2222222222</SSN>
          <Gender>F</Gender>
          <Benefits>
          </Benefits>
        </Member>
      </Contract>
    </Contracts>
  </Sponsor>
</Sponsors>
</Sender>

我想从父节点获取合约节点的所有信息，以及 SponsorID。下面是使用 XmlReader 部分读取 xml 文件的代码：

        static IEnumerable<XElement> SimpleStreamAxis(string inputUrl, string elementName)      
    {

            using (XmlReader reader = XmlReader.Create(inputUrl))
            {
                reader.MoveToContent();
                while (reader.Read())
                {
                    if (reader.NodeType == XmlNodeType.Element)
                    {
                        if (reader.Name == elementName)
                        {
                            XElement el = XNode.ReadFrom(reader) as XElement;
                            if (el != null)
                            {
                                yield return el;
                            }
                        }
                    }
                }
            }                  
    }

这就是问题所在。我不能使用这个，因为整个赞助商树可能对内存来说太大了。

var sponsor = SimpleStreamAxis(file, "Sponsor");

我也不能使用这个，因为我不能只用合约节点信息告诉 SponsorID。

var contract = SimpleStreamAxis(file, "Contract");

有没有办法可以读取Sponsor中的SponsorID，向前移动光标，读取这个Sponsor下的所有Contract节点，然后移动到下一个Sponsor读取SponsorID及其Contract节点等等？

Answer 1

试试这个：

using (XmlReader xmlReader = XmlReader.Create("file.xml"))
{
    while (xmlReader.Read())
    {
        if (xmlReader.ReadToFollowing("SponsorID"))
        {
            string sponsorId = xmlReader.ReadElementContentAsString();

            // process SponsorID
            Console.WriteLine(sponsorId);

            if (xmlReader.ReadToFollowing("Contract"))
            {
                do
                {
                    XmlReader contractSubtree = xmlReader.ReadSubtree();
                    XElement contractElement = XElement.Load(contractSubtree);

                    // process Contract
                    Console.WriteLine(contractElement.Element("ContractID"));

                } while (xmlReader.ReadToNextSibling("Contract"));
            }
        }
    }
}

Answer 2

是的，这可以做到假设SponsorID 总是在Contract 节点之前。

基本思想是通读 XML 文件，直到找到具有所需名称 "SponsorID" 或"Contract" 的元素，然后将它们生成以进行更高的处理

    public static IEnumerable<XElement> StreamNamedElements(XmlReader reader, IEnumerable<XName> names)
    {
        var nameSet = new HashSet<XName>(names);

        while (reader.Read())
        {
            if (reader.NodeType == XmlNodeType.Element && nameSet.Contains(XName.Get(reader.LocalName, reader.NamespaceURI)))
            {
                XElement el = XNode.ReadFrom(reader) as XElement;
                if (el != null)
                    yield return el;
            }
        }
    }

如果SponsorID 始终存在并且在Contract 之前，这将正确枚举这些元素。但是，如果某个赞助商 ID 丢失或乱序，则可能会获取之前赞助商的赞助商 ID。可以通过使用 ReadSubtree() 将每个“SponsorID”的范围限制为包含“Sponsor”元素的范围来捕获此错误：

    public static IEnumerable<XmlReader> StreamNamedSubtrees(XmlReader reader, IEnumerable<XName> names)
    {
        var nameSet = new HashSet<XName>(names);

        while (reader.Read())
        {
            if (reader.NodeType == XmlNodeType.Element && nameSet.Contains(XName.Get(reader.LocalName, reader.NamespaceURI)))
            {
                var subReader = reader.ReadSubtree();
                yield return subReader;
                ((IDisposable)subReader).Dispose(); // Be sure to advance to the end of the subtree if the caller did not.
            }
        }
    }

然后像这样使用它：

        using (var sr = new StringReader(xml))
        using (var reader = XmlReader.Create(sr))
        {
            foreach (var subReader in StreamNamedSubtrees(reader, new[] { (XName)"Sponsor" }))
            {
                XElement sponsorID = null;
                foreach (var el in StreamNamedElements(subReader, new[] { (XName)"SponsorID", (XName)"Contract" }))
                {
                    if (el.Name == "SponsorID")
                    {
                        sponsorID = el;
                    }
                    else if (el.Name == "Contract")
                    {
                        if (sponsorID == null)
                            throw new InvalidOperationException();
                        // Example "higher processing"
                        Debug.WriteLine(string.Format("{0}: {1}", sponsorID.Value, el.ToString()));
                    }
                }
            }
        }