Question

我有一个手机所有位置的数据集。（如果你有兴趣的话，我是通过谷歌外卖获得的。）数据的问题在于，在某个时刻，我得到了第二部手机。我拥有的数据集没有任何允许我通过特定手机跟踪数据的信息。因此，如果我在家里留下电话，那么它会立刻向我显示两个地方。我决定编写一个查询，试图通过确定最后5个中哪个点最接近消除来找到相邻点，并且我必须以超过150英里/小时的速度行进以便到达。

数据的表定义如下：

CREATE TABLE [dbo].[locationdata](
    [ID] [bigint] IDENTITY(1,1) NOT NULL,
    [t] [datetime] NULL,
    [lat] [float] NULL,
    [long] [float] NULL,
    [accuracy] [smallint] NULL,
    [activity] [varchar](14) NULL,
    [confidence] [int] NULL,
    [velocity] [varchar](2) NULL,
    [altitude] [smallint] NULL,
    [heading] [smallint] NULL,
    [point] [geography] NULL,
    [tag] [varchar](50) NULL,
 CONSTRAINT [PK_locationdata] PRIMARY KEY CLUSTERED 
(
    [ID] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY] TEXTIMAGE_ON [PRIMARY]

GO

行按时间顺序插入，因此ID按正确的顺序排列，但多个点可以存在相同的时间。

所以这是我尝试没有CURSOR写这个。问题是你不能在公用表表达式的递归部分使用“TOP”。

WITH tripdata(originid, endid, startid, speed, distance, startpoint, startt)
AS
(
SELECT originid, endid, startid, speed, distance, startpoint, startt
FROM
    (
        SELECT 
               origin.id as originid
             , NULL as endid
             , origin.id as startid
             , NULL as distance
             , NULL as speed
             , point as startpoint
             , t as startt
        FROM locationdata origin
    )  a
UNION ALL

        SELECT
              originid as originid
            , startid as endid
            , l.id as startid
            , origin.startpoint.STDistance(l.point) as distance
            , (origin.startpoint.STDistance(l.point)/(datediff(S, origin.startt, l.t))) * -2.23694  as speed
            , l.point as startpoint
            , l.t as startt
        FROM tripdata origin
        CROSS APPLY
        (       
            SELECT top 1
                  z.id
                 ,z.point
                 ,z.t
            FROM locationdata z
            where origin.startid > z.ID and origin.startid -5 < z.ID
                and z.t <> origin.startt
                and  (origin.startpoint.STDistance(z.point)/(datediff(S, origin.startt, z.t))) * -2.23694 < 150
            order by origin.startpoint.STDistance(z.point)
        ) l

)
SELECT *
FROM tripdata
WHERE originid = 218255
;

我愿意接受有关如何修复此查询或甚至是否可能的建议。

递归地查找最近的相邻点

0 个答案: