我有一个xml文件,其中包含带有IP地址的所有学校名称,并且我正在尝试按计时器间隔(即SchoolFunctions.Backup
)备份每所学校。
问题是,在某些情况下,备份速度太慢,并且在先前的备份仍在运行时,备份方法将再次启动...
如果SchoolFunctions.Backup
仍在该学校运行,我该如何阻止它执行?
using (DataTable schools = new DataTable { TableName = "Schools" })
{
schools.ReadXml(AppSettings.Default.SettingsPath);
try
{
Parallel.ForEach(schools.AsEnumerable(), _opts, row =>
{
SchoolFunctions.Backup(row.Field<string>("IPAddress"), row.Field<string>("Name"));
}
}
...
}
答案 0 :(得分:3)
我个人会和凯文斯一起回答
但是,如果您希望这些备份独立运行(不必等待所有学校完成),并且大多数情况下使用您当前的模式。然后可以玩一些这样的想法
private ConcurrentDictionary<string,bool> _dict = new ConcurrentDictionary<string,bool>();
...
using (DataTable schools = new DataTable { TableName = "Schools" })
{
schools.ReadXml(AppSettings.Default.SettingsPath);
try
{
Parallel.ForEach(schools.AsEnumerable(), _opts, row =>
{
var ipAddress = row.Field<string>("IPAddress");
// check if there is an ip registered and if its processing
if (_dict.TryGetValue(ipAddress, out processing) && processing)
return;
// its not processing ,so update it
_dict.AddOrUpdate(ipAddress, true, (s, b) => true);
SchoolFunctions.Backup(ipAddress , row.Field<string>("Name"));
// when we are done update processing to false
_dict.AddOrUpdate(ipAddress, false, (s, b) => false);
}
}
}
注意,有很多方法可以做到这一点,但这看起来很有希望。也没有测试
此外,您可能只使用TryAdd
和TryRemove
而忽略了值
if (!_dict.TryAdd(ipAddress, false))
return;
SchoolFunctions.Backup(ipAddress , row.Field<string>("Name"));
_dict.TryRemove(ipAddress,out _);