从IPointCollection读取点效率比较

void ReadPoints1(IPointCollection ptCol) { int iCount = ptCol.PointCount; IPoint pt; for (int i = 0; i < iCount; i++) { pt = ptCol.get_Point(i); } } void ReadPoints2(IPointCollection ptCol) { int iCount = ptCol.PointCount; IPoint pt; for (int i = 0; i < iCount; i++) { pt = ptCol.QueryPoint(i, pt); } }

 

我平时一般喜欢用第一种，思维定势吧。

 

但是，当读取一个线图层中，由20多w个点组成时，发现第一种很费时间。

 

于是，就寻找他路，终于，尝试方法2的时候，发现，找到了！

 

2比1的效率高太多！

 

再看帮助文档，如下：

IPointCollection.QueryPoint Method

Note: The output geometry must be co-created prior to the query. The output geometry is not co-created by the method; it is populated. This can be used in performance critical situations. For example, creating the geometry only once outside a loop and use the query method could improve performance.

 

看来，文档说的很清楚，只是平时用get_Point习惯了。