add disassemble example

pymeos_examples/PyMEOS Examples/pymeos_disassemble_berlinmod.py

@@ -0,0 +1,70 @@
+import datetime
+from pymeos import Temporal
+from pymeos_cffi import meos_initialize, meos_finish
+from postgis import Geometry
+import pandas as pd
+
+import warnings
+warnings.filterwarnings("ignore")
+
+if __name__ == '__main__':
+    meos_initialize()
+
+    # define converters for parsing csv files
+    geom_converter = Geometry.from_ewkb
+    hexwkb_converter = Temporal.temporal_from_hexwkb
+
+    # read csv files
+    trips = pd.read_csv('../data/trips.csv', converters={'trip': hexwkb_converter, 'trajectory': geom_converter})
+    print("%d trip records read.\n\n" % len(trips))
+
+    # record the index of current instant of each trip
+    # in each iteration, the current instant with minimum timestamp 
+    # will be output and its index will move forward
+    curr_inst_idx = [1 for i in range(len(trips))]
+    num_insts = [t.num_instants for t in trips['trip']]
+    MAX_DATETIME = datetime.datetime.max.replace(tzinfo=datetime.timezone.utc)
+
+    # initialize the list of timestamp of current instance of each trip
+    curr_inst_ts = list()
+    for i in range(len(trips)):
+        if curr_inst_idx[i] > 0 and curr_inst_idx[i] <= num_insts[i]:
+            t = trips['trip'][i]
+            curr_inst = t.instant_n(curr_inst_idx[i])
+            curr_inst_ts.append(curr_inst.timestamp)
+        else:
+            curr_inst_ts.append(MAX_DATETIME)
+
+    count = 0
+
+    # output
+    f = open('trip_instants.csv', 'w')
+    f.write('vehicle,day,seq,geom,t\n')
+    print(f"There are supposed to be {sum(num_insts)} obeservations, which might take a while...")
+
+    while(True):
+        if len([ts for ts in curr_inst_ts if ts < MAX_DATETIME]) == 0:
+            break
+        # write the instance to file
+        selected_trip_idx = min(range(len(curr_inst_ts)), key=curr_inst_ts.__getitem__)
+        row = trips.iloc[selected_trip_idx]
+        selected_trip = row['trip']
+        curr_inst = selected_trip.instant_n(curr_inst_idx[selected_trip_idx])
+        inst_str = curr_inst.__str__()
+        str_split = inst_str.split('@')
+        f.write(f'{row.vehicle},{row.day},{row.seq},{str_split[0]},{str_split[1]}\n')
+        count += 1
+
+        # update curr instance index and timestamp
+        if curr_inst_idx[selected_trip_idx] >= num_insts[selected_trip_idx]:
+            # finish the trip
+            curr_inst_idx[selected_trip_idx] = -1
+            curr_inst_ts[selected_trip_idx] = MAX_DATETIME
+        else:
+            curr_inst_idx[selected_trip_idx] += 1
+            next_inst = selected_trip.instant_n(curr_inst_idx[selected_trip_idx])
+            curr_inst_ts[selected_trip_idx] = next_inst.timestamp
+
+    print(f"{count} observation records written.")
+    f.close()
+    meos_finish()