根据文档,2dSphere 地理空间索引也可用于复杂的 GeoJSON 形状,例如由多个节点组成的 LineString 和 Polygon。
当您使用$nearSphere搜索填充了此类复杂形状的字段以接近另一个 GeoJSON 对象(也可能是复杂形状)时,如何计算这些形状的距离?
它会比较形状的最近节点吗?最远的节点?节点的一些平均值?形状的任何边缘上的最近点?
根据文档,2dSphere 地理空间索引也可用于复杂的 GeoJSON 形状,例如由多个节点组成的 LineString 和 Polygon。
当您使用$nearSphere搜索填充了此类复杂形状的字段以接近另一个 GeoJSON 对象(也可能是复杂形状)时,如何计算这些形状的距离?
它会比较形状的最近节点吗?最远的节点?节点的一些平均值?形状的任何边缘上的最近点?
$nearSphere
运算符仅允许您指定一个点(您可以通过提供maxDistance
参数定义的“球体”的中心)。因此,唯一有意义的答案可能是关于形状如何排序最近到最远。
答案是所有具有等距“最近”点的形状都将排在距离最近点更远的形状之前。换言之,点与形状之间的距离是点与形状上最近点之间的距离。
我们可以看到,通过运行一个变体$nearSphere
是$geoNear
命令,因为它返回计算出的距离,然后我们可以检查。
我们的样本数据:
> db.near.find()
{ "_id" : 10, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 5 ], [ 5, 5 ], [ 5, 0 ], [ 0, 0 ] ] ] } }
{ "_id" : 11, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 1, 1 ], [ 1, 5 ], [ 5, 5 ], [ 5, 1 ], [ 1, 1 ] ] ] } }
{ "_id" : 12, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [ 0, 0 ] ] ] } }
{ "_id" : 13, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [0, 0 ] ], [ [ 1, 1 ], [ 1, 5 ], [ 5, 5 ], [ 5, 1 ], [ 1, 1 ] ] ] } }
{ "_id" : 14, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [ 0, 0 ] ], [ [ 0.1, 0.1 ], [ 0.1, 5 ], [ 5, 5 ], [ 5, 0.1 ], [ 0.1, 0.1 ] ] ] } }
{ "_id" : 15, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 1 ], [ 1, 1 ], [ 1, 0 ], [0, 0 ] ], [ [ 0.1, 0.1 ], [ 0.1, 0.5 ], [ 0.5, 0.5 ], [ 0.5, 0.1 ], [ 0.1, 0.1 ] ] ] } }
{ "_id" : 16, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0.1, 0.1 ] ] } }
{ "_id" : 17, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0, 0.1 ] ] } }
{ "_id" : 18, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0.1, 0 ] ] } }
我们使用 $nearSphere 的结果:
{ "_id" : 16, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0.1, 0.1 ] ] } }
{ "_id" : 18, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0.1, 0 ] ] } }
{ "_id" : 14, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [ 0, 0 ] ], [ [ 0.1, 0.1 ], [ 0.1, 5 ], [ 5, 5 ], [ 5, 0.1 ], [ 0.1, 0.1 ] ] ] } }
{ "_id" : 13, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [ 0, 0 ] ], [ [ 1, 1 ], [ 1, 5 ], [ 5, 5 ], [ 5, 1 ], [ 1, 1 ] ] ] } }
{ "_id" : 15, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 1 ], [ 1, 1 ], [ 1, 0 ], [ 0, 0 ] ], [ [ 0.1, 0.1 ], [ 0.1, 0.5 ], [ 0.5, 0.5 ], [ 0.5, 0.1 ], [ 0.1, 0.1 ] ] ] } }
{ "_id" : 12, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 6 ], [ 6, 6 ], [ 6, 0 ], [ 0, 0 ] ] ] } }
{ "_id" : 17, "loc" : { "type" : "LineString", "coordinates" : [ [ 0, 0 ], [ 0, 0.1 ] ] } }
{ "_id" : 10, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 0, 0 ], [ 0, 5 ], [ 5, 5 ], [ 5, 0 ], [ 0, 0 ] ] ] } }
{ "_id" : 11, "loc" : { "type" : "Polygon", "coordinates" : [ [ [ 1, 1 ], [ 1, 5 ], [ 5, 5 ], [ 5, 1 ], [ 1, 1 ] ] ] } }
运行$geoSphere命令,我们可以确认点到形状的距离(出于空间考虑,省略坐标):
> db.runCommand({geoNear:"near", near: {type:"Point", coordinates:[0,0]}, spherical:true})
[
{
"dis" : 0,
"obj" : {
"_id" : 16,
"loc" : {
"type" : "LineString"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 18,
"loc" : {
"type" : "LineString"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 14,
"loc" : {
"type" : "Polygon"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 13,
"loc" : {
"type" : "Polygon"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 15,
"loc" : {
"type" : "Polygon"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 12,
"loc" : {
"type" : "Polygon"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 17,
"loc" : {
"type" : "LineString"
}
}
},
{
"dis" : 0,
"obj" : {
"_id" : 10,
"loc" : {
"type" : "Polygon"
}
}
},
{
"dis" : 157424.6238723255,
"obj" : {
"_id" : 11,
"loc" : {
"type" : "Polygon"
}
}
}
]
如您所见,每个形状(除了从 [1,1] 开始的最后一个形状)到点 [0,0] 的距离都是 0,即使这些形状的大小和形状都不同。在到指定点的距离相同的文档中,不定义顺序。这意味着即使它看起来是确定性的,也不能保证它是相同的。