scr_005_od_rids.py

"""
Write the region IDs for each pickup and dropoff in the OD data.

For each row in --filename such as "./data/od_part-m-000.pkl",
  write the corresponding list of pickup region IDs to string -d
  write the corresponding list of dropoff region IDs to string -p
"""

import pandas as pd
import numpy as np
from math import floor, ceil
from matplotlib import path
import argparse

def process_line(line):
    # Not sure what these values are used for tbh!
    val1 = line[0]
    val2 = line[1]
    polyline = [float(v) for v in line[2:]]
    polyline = np.array(list(zip(polyline[0::2], polyline[1::2])))
    return val1, val2, polyline 

def point_to_rids(point, polypaths, radius=0.0):
    RIDs = []
    for rid, polypath in enumerate(polypaths):
        if polypath.contains_point(point, radius=radius):
            RIDs.append(rid)
    
    return RIDs


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--filename", "-f",
                        help="Filename of OD data to look at.",
                        type=str, nargs=1)
    parser.add_argument("--drop_outname", "-d",
                        help="Filename to explort CSV of drop RIDs to",
                        type=str, nargs=1)
    parser.add_argument("--pick_outname", "-p",
                        help="Filename to explort CSV of pick RIDs to",
                        type=str, nargs=1)
    
    args = parser.parse_args()
    fn = args.filename[0]
    print(f"Processing {fn}")
    
    with open("./DataForUConn/DecideRegion/area.txt") as f:
        area = [process_line(line.split(",")) for line in f.read().strip().split("\n")]
    
    polypaths = [path.Path(line[2][1:]) for line in area]
    
    df = pd.read_pickle(fn)
    
    pick_RIDs_outfile = open(args.pick_outname[0], "w+")
    drop_RIDs_outfile = open(args.drop_outname[0], "w+")
    
    for ii in range(len(df)):
        if ii % 10000 == 0:
            print(f"  Processing point {ii} of {len(df)}...")
        
        # Log points that correspond to 0 or multiple RIDs.
        pick_point = (df["plon"][ii], df["plat"][ii])
        drop_point = (df["dlon"][ii], df["dlat"][ii])
        # List of region IDs for pickup, dropoff
        list_of_pick_RIDs = point_to_rids(pick_point, polypaths=polypaths)
        list_of_drop_RIDs = point_to_rids(drop_point, polypaths=polypaths)
        
        pick_RIDs_outfile.write(str(list_of_pick_RIDs)[1:-1])
        pick_RIDs_outfile.write("\n")
        drop_RIDs_outfile.write(str(list_of_drop_RIDs)[1:-1])
        drop_RIDs_outfile.write("\n")
    
    pick_RIDs_outfile.close()
    drop_RIDs_outfile.close()




"""
Usage:

python -i scr_005_od_rids.py -f ./data/od_part-m-000.pkl -p ./data/od_pick-000.csv -d ./data/od_drop-000.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-001.pkl -p ./data/od_pick-001.csv -d ./data/od_drop-001.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-002.pkl -p ./data/od_pick-002.csv -d ./data/od_drop-002.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-003.pkl -p ./data/od_pick-003.csv -d ./data/od_drop-003.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-004.pkl -p ./data/od_pick-004.csv -d ./data/od_drop-004.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-005.pkl -p ./data/od_pick-005.csv -d ./data/od_drop-005.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-006.pkl -p ./data/od_pick-006.csv -d ./data/od_drop-006.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-007.pkl -p ./data/od_pick-007.csv -d ./data/od_drop-007.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-008.pkl -p ./data/od_pick-008.csv -d ./data/od_drop-008.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-009.pkl -p ./data/od_pick-009.csv -d ./data/od_drop-009.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-010.pkl -p ./data/od_pick-010.csv -d ./data/od_drop-010.csv
python -i scr_005_od_rids.py -f ./data/od_part-m-011.pkl -p ./data/od_pick-011.csv -d ./data/od_drop-011.csv
"""