Question

我有一个使用时间戳的数据帧，结果显示每一秒，如果数据每60秒或1分钟带来数据，我怎样才能得到这个数据帧的聚合？框架

string fullPath="1$2$3";
List<string> nodesToBeAdded = fullPath.Split('$').ToList();

XmlDocument xmlDocument = new XmlDocument();
xmlDocument.Load(filePath);

XmlNode nodeContent = xmlDocument.CreateNode(XmlNodeType.Element, "Content", null);

XmlNode nodeUID = xmlDocument.CreateNode(XmlNodeType.Element, "UId", null);
nodeUID.InnerText = value.UId;

XmlNode nodeFileName = xmlDocument.CreateNode(XmlNodeType.Element, "FileName", null);
nodeFileName.InnerText = value.FileName;

XmlNode nodeImage = xmlDocument.CreateNode(XmlNodeType.Element, "Image", null);
nodeImage.InnerText = value.Image;

bool addNodeContent = false;
for (int i = 0; i < nodesToBeAdded.Count - 1; i++)
{
    XmlNode nodeFullPath = xmlDocument.CreateNode(XmlNodeType.Element, "FullPath", null);
    nodeFullPath.InnerText = nodesToBeAdded[i];

    if (xmlDocument.SelectNodes(string.Concat("//", "Content", '/', "FullPath", "[text()='" + nodesToBeAdded[i] + "']")).Count == 0)
    {
        nodeContent.AppendChild(nodeUID);
        nodeContent.AppendChild(nodeFileName);
        nodeContent.AppendChild(nodeImage);
        nodeContent.AppendChild(nodeFullPath);

        addNodeContent = true;
    }
}

if (addNodeContent)
{
   //add parent node to document
   xmlDocument.DocumentElement.AppendChild(nodeContent);
}

xmlDocument.Save(filePath);

我想使用spark SQL和Scala 非常感谢我想减少数据集的大小非常感谢

基于数据的聚合数据帧

0 个答案: