使用大型阵列时程序崩溃 [英] Program crashes when working with large array

查看：97
本文介绍了使用大型阵列时程序崩溃的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！

问题描述

我的程序循环遍历一个空元素数组，将每个元素分配给一个新对象(我制作的类的实例).数组变得非常大-大约有60,000个元素.有时，此循环会起作用，但通常程序会崩溃，并出现通用的Microsoft错误"x.exe遇到问题，需要关闭...".我已经得出结论，这不是内存 size 问题，但是可能是某种内存管理问题...

这是Point类:

My program loops through an array of null elements, assigning each element to a new object (an instance of a class I made). The array gets quite large - about 60,000 elements in size. On occasion, this loop will work, but generally the program will crash, giving a generic microsoft error "x.exe has encountered a problem and needs to close...". I have concluded that this is not a memory size issue, but could be a memory management issue of some sort...

Here is the Point class:

public partial class Point : System.Windows.Window
    {
        private string m_ID; // Theoretically to be used only for the transmitter point locations... Could be used for other points too depending on application...
        public double x, y, power;
        public string m_time;
        public Point()
        {
            power = 0;
        }
        public Point(double x, double y)
        {
            this.x = x; this.y = y; power = 0;
        }
        public Point(double x, double y, int column)
        {
            this.x = x; this.y = y; power = 0;
        }
        public Point(double x, double y, int column, string time)
        {
            m_time = time; this.x = x; this.y = y; power = 0;
        }
        public Point(double x, double y, int column, string time, double power)
        {
            m_time = time; this.x = x; this.y = y; this.power = power;
        }
        /// <summary>
        /// THE LABEL OR NAME OF THE GIVEN POINT.
        /// </summary>
        public string ID
        {
            get
            {
                return m_ID;
            }
            set
            {
                m_ID = value;
            }
        }
        public string time
        {
            get
            {
                return m_time;
            }
            set
            {
                m_time = value;
            }
        }
    }



这是构建Point's数组的函数:



And here is the function which builds the array of Point''s:

public Point[] getPoints(ExcelInterface excelData)
        {
            int longColumn = excelData.FindColumnNumber("Longitude");
            int latColumn = excelData.FindColumnNumber("Latitude");
            int timeColumn = excelData.FindColumnNumber("Time");
            int powerColumn = excelData.FindColumnNumber("Power_dBm");
            Point[] dataPoints = new Point[excelData.rows];
            string hh, mm, ss, MM, dd;
            hh = string.Empty; mm = string.Empty; ss = string.Empty; MM = string.Empty; dd = string.Empty;
            int hour, minute, second;
            double power;
            DateTime time;
            double[] thresholds = getPowerThresholds(numThresholds, excelData);
            string dateTime, timeString;
            if (System.DateTime.Now.Month < 10)
                MM = "0" + System.DateTime.Now.Month.ToString();
            else MM = System.DateTime.Now.Month.ToString();
            if (System.DateTime.Now.Day < 10)
                dd = "0" + System.DateTime.Now.Day.ToString ();
            else dd = System.DateTime.Now.Day.ToString();
            string date = System.DateTime.Now.Year + "-" + MM + "-" + dd;
            double lat, lon;
            // array stop between 1900 and 2000 in size...
            for (int i = 2; i < excelData.rows; i++)
            {
                if (i % 400 == 0)
                {
                }
                time = new DateTime(); // DEBUG
                lat = (double)excelData.GetCellValue(i, latColumn, 1);
                lon = (double)excelData.GetCellValue(i, longColumn, 1);
                time = (DateTime)excelData.GetCellValue(i, timeColumn, 1);
                power = (double)excelData.GetCellValue(i, powerColumn, 1);
                hour = time.Hour;
                if (hour < 10)
                    hh = "0" + hour.ToString();
                else hh = hour.ToString();
                minute = time.Minute;
                if (minute < 10)
                    mm = "0" + minute.ToString();
                else mm = minute.ToString();
                second = time.Second;
                if (second < 10)
                    ss = "0" + second.ToString();
                else ss = second.ToString();
                timeString = "T" + hh + ":" + mm + ":" + ss + "Z";
                dateTime = date + timeString;
                try
                {
                    dataPoints[i - 2] = new Point(lon, lat, i, dateTime);  
                }
                catch (Exception ex)
                {
                }
            }



我还通过仅使用构造函数简化了错误发生的循环，从而进行了测试:



I have also tested this by simplifying the loop where the error occurrs, by using only the constructor:

   for (int i = 2; i < excelData.rows; i++)
            {
                try
                {
                    dataPoints[i - 2] = new Point(0, 0, 0, "");  
                }
                catch (Exception ex)
                {
                }
}



在此测试方案下，程序仍然崩溃.我还构建了一个简单的测试程序，该程序仅创建一个点数组并将每个数组元素分配给一个新的Point.该程序可以很好地处理超过一百万个元素...

我真的很感谢您的帮助-我已经为此努力了几天！

Rich



The program still crashed with this test scenario. I have also built a simple test program which solely creates a point array and assigns each array element to a new Point. That program worked just fine with over 1million elements...

I would really appreciate any help - Ive been wrestling with this for days!

Rich

推荐答案

Erm ...您是否有任何理由要从 System.Windows.Window 继承，因为我怀疑这是您当前的原因问题?

您可以像在轻量级结构中一样轻松地存储点信息.您还可以消除日期内容周围的许多字符串操作...

我通常不喜欢仅仅给出答案，但是您的示例中有很多多余的内容，这能胜任吗?

Erm...is there any reason your inheriting from System.Windows.Window as I suspect this is cause of your current issue?

You could store your point information just as easily in a lightweight struct. You could also eliminate alot of the string operations around the date stuff...

I don''t usually like just giving out answers, but there is a lot of excess compexity in your example, will this do the job?

public struct Point
{
	public double Longitude{get;set;}
	
	public double Latitude{get;set;}
	
	public double Power{get;set;}
	
	public DateTime Time{get;set;} 
}
	
public Point[] getPoints(ExcelInterface excelData)
{
    int longColumn = excelData.FindColumnNumber("Longitude");
    int latColumn = excelData.FindColumnNumber("Latitude");
    int timeColumn = excelData.FindColumnNumber("Time");
    int powerColumn = excelData.FindColumnNumber("Power_dBm");
    Point[] dataPoints = new Point[excelData.rows];
  	DateTime date = DateTime.Now.Date()
    double[] thresholds = getPowerThresholds(numThresholds, excelData); // Don't know what this is for
    double lat, lon, power;
    // array stop between 1900 and 2000 in size...
	List<Point> points = new List<Point>();
    for (int i = 2; i < excelData.rows; i++)
    {
		Point p = new Point
		{
			Longitude =  (double)excelData.GetCellValue(i, longColumn, 1)
			,Latitude =  (double)excelData.GetCellValue(i, latColumn, 1)
			,Time = date.Add(((DateTime)excelData.GetCellValue(i, timeColumn, 1)).TimeOfDay)
			, Power =  (double)excelData.GetCellValue(i, powerColumn, 1)
		}
		points.Add(p)
    }
	return points.ToArray();
}


哦，对不起，我在这篇文章之前删除了Windows继承.不太确定为什么要这么做...但这不是问题的原因-删除继承并没有阻止错误.实际上，使用列表而不是数组可以解决它！

现在已经解决了这个问题，我遇到了另一个奇怪的问题……我有一个格式化某些点节点并将这些节点放入kml的函数.这些点由Google Earth显示.在foreach循环中循环时，出现通用的Windows错误消息，并且程序在任意行崩溃.实际上，当我调试和逐步执行代码时，在步骤之间出现错误...

这是代码:

Oh, sorry, the windows inheritance I removed prior to this post. Not too sure why I did that... That was not the cause of the issue though - removing the inheritance didn''t stop the error. Actually, using a list instead of an array solved it!

Now that this one is solved, I have run into another strange issue... I have a function that formats some point nodes and places these nodes into a kml. These points are displayed by google earth. While looping through the foreach loop, a generic windows error message appears and the program crashes at a random line. Actually, while I was debugging and stepping through the code, an error occurred in between steps...

Heres is the code:

private void plotPoints(Point[] points, double[] thresholds)
    {
        string node;
        string style = "";

        System.Xml.XPath.XPathNavigator navigator = xDoc.CreateNavigator();
        navigator.MoveToChild("kml", "http://www.opengis.net/kml/2.2");
        navigator.MoveToFirstChild();

        if (isNewKml())
            insertFolder("Points", navigator);

        else // delete folder
        {
            deleteFolder("Points");
            insertFolder("Points", navigator);
        }

        navigator = getNavigator(navigator, "Points");

        foreach (Point p in points)
        {
            if (p != null)
            {
                if (1 % 1000 == 0)
                { }

                // Errors occur here at random places...
                // Check power thresholding and apply colors accordingly
                for (int n = 0; n < thresholds.Length; n++) // Error here
                {// error here
                    if (p.power < thresholds[n])
                    {
                        style = "CustomStyle" + (n + 1).ToString();
                        break;
                    }
                    else if (n == (thresholds.Length - 1)) // Gets the case where a power measurement is above the average max (20 highest measurements)
                        style = "CustomStyle" + (n + 1).ToString();
                }
                object[] arguments = { pointIdex, System.Environment.NewLine, p.time, System.Environment.NewLine, style, System.Environment.NewLine, p.x, p.y };

                // Error here (above line) while waiting (not stepping)!

                node = String.Format("<name>Point{0}</name>{1}<TimeStamp> <when>{2}</when> </TimeStamp>{3}<styleUrl>#{4}</styleUrl>{5}<Point> <coordinates>{6},{7}, 0</coordinates> </Point>");

                navigator.MoveToFirstChild();
                navigator.InsertElementAfter(navigator.Prefix, "Placemark", navigator.LookupNamespace(navigator.Prefix), null); // "Point" + pointIdex
                navigator.MoveToNext();
                // navigator.MoveToFirstChild(); // not used?
                navigator.AppendChild(node);
                pointIdex++;
                navigator.MoveToParent();
            }
        }
    }

}



有谁知道为什么会这样?



Does anyone have any idea why this might be?


这篇关于使用大型阵列时程序崩溃的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！

查看全文
相关文章
其他开发语言最新文章
热门教程
热门工具
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆