优化Java时空数据重叠查询：索引技术深度解析

import com.github.tzaeschke.tinspin.indexes.rtree.RTree; // 假设引入Tinspin的RTree实现
import java.util.ArrayList;
import java.util.List;
import java.util.function.Consumer;

/**
 * 代表一个具有空间和时间范围的事件。
 * 空间范围由kilometerInitial和kilometerFinal定义。
 * 时间范围由instantDateInitial和instantDateFinal定义。
 */
class SpatioTemporalEvent {
    double kilometerInitial;
    double kilometerFinal;
    long instantDateInitial; // 例如使用Unix时间戳
    long instantDateFinal;
    String eventId; // 事件唯一标识符

    public SpatioTemporalEvent(double ki, double kf, long di, long df, String id) {
        this.kilometerInitial = ki;
        this.kilometerFinal = kf;
        this.instantDateInitial = di;
        this.instantDateFinal = df;
        this.eventId = id;
    }

    // 获取事件的最小坐标 (Xmin, Ymin)
    public double[] getMinCoordinates() {
        return new double[]{kilometerInitial, (double)instantDateInitial};
    }

    // 获取事件的最大坐标 (Xmax, Ymax)
    public double[] getMaxCoordinates() {
        return new double[]{kilometerFinal, (double)instantDateFinal};
    }

    @Override
    public String toString() {
        return "Event{" + "id='" + eventId + '\'' +
               ", spatial=[" + String.format("%.2f", kilometerInitial) + "," + String.format("%.2f", kilometerFinal) + "]" +
               ", temporal=[" + instantDateInitial + "," + instantDateFinal + "]}";
    }
}

public class SpatioTemporalOverlapDetector {

    public static void main(String[] args) {
        // 初始化一个2维的R-Tree索引 (1维空间，1维时间)
        RTree<SpatioTemporalEvent> rTreeIndex = new RTree<>(2);

        // 插入一些示例时空事件
        SpatioTemporalEvent event1 = new SpatioTemporalEvent(0, 10, 100, 200, "EventA");
        SpatioTemporalEvent event2 = new SpatioTemporalEvent(5, 15, 150, 250, "EventB"); // 与A重叠
        SpatioTemporalEvent event3 = new SpatioTemporalEvent(20, 30, 50, 150, "EventC"); // 空间不重叠，时间部分重叠
        SpatioTemporalEvent event4 = new SpatioTemporalEvent(8, 12, 180, 220, "EventD"); // 完全在A/B重叠区域内
        SpatioTemporalEvent event5 = new SpatioTemporalEvent(3, 7, 120, 180, "EventE"); // 与A部分重叠

        rTreeIndex.insert(event1.getMinCoordinates(), event1.getMaxCoordinates(), event1);
        rTreeIndex.insert(event2.getMinCoordinates(), event2.getMaxCoordinates(), event2);
        rTreeIndex.insert(event3.getMinCoordinates(), event3.getMaxCoordinates(), event3);
        rTreeIndex.insert(event4.getMinCoordinates(), event4.getMaxCoordinates(), event4);
        rTreeIndex.insert(event5.getMinCoordinates(), event5.getMaxCoordinates(), event5);

        System.out.println("索引中事件总数: " + rTreeIndex.size());

        // 定义一个查询事件，查找所有与它重叠的事件
        SpatioTemporalEvent queryEvent = new SpatioTemporalEvent(7, 13, 170, 230, "QueryEvent");
        System.out.println("\n查询与 " + queryEvent + " 重叠的事件:");

        List<SpatioTemporalEvent> overlappingResults = new ArrayList<>();
        // Tinspin的queryWindow方法接受查询窗口的min/max坐标，并使用Consumer处理结果
        rTreeIndex.queryWindow(queryEvent.getMinCoordinates(), queryEvent.getMaxCoordinates(), new Consumer<SpatioTemporalEvent>() {
            @Override
            public void accept(SpatioTemporalEvent resultEvent) {
                overlappingResults.add(resultEvent);
            }
        });

        if (overlappingResults.isEmpty()) {
            System.out.println("  - 未发现重叠事件。");
        } else {
            for (SpatioTemporalEvent event : overlappingResults) {
                System.out.println("  - 发现重叠: " + event);
            }
        }

        // 另一个查询：查找与EventC重叠的事件
        System.out.println("\n查询与 " + event3 + " 重叠的事件:");
        overlappingResults.clear();
        rTreeIndex.queryWindow(event3.getMinCoordinates(), event3.getMaxCoordinates(), overlappingResults::add);
        if (overlappingResults.isEmpty()) {
            System.out.println("  - 未发现重叠事件。");
        } else {
            for (SpatioTemporalEvent event : overlappingResults) {
                System.out.println("  - 发现重叠: " + event);
            }
        }
    }
}