remove geopy dependency

backend/Pipfile

@@ -9,7 +9,6 @@ name = "pypi"
 numpy = "*"
 fastapi = "*"
 pydantic = "*"
-geopy = "*"
 shapely = "*"
 scipy = "*"
 osmpythontools = "*"

backend/src/persistence.py

@@ -21,8 +21,8 @@ if constants.MEMCACHED_HOST_PATH is None:
 else:
     client = Client(
         constants.MEMCACHED_HOST_PATH,
-        timeout=1,
+        timeout = 1,
-        allow_unicode_keys=True,
+        allow_unicode_keys = True,
-        encoding='utf-8',
+        encoding = 'utf-8',
-        serde=serde.pickle_serde
+        serde = serde.pickle_serde
     )

backend/src/structs/landmark.py

@@ -5,7 +5,7 @@ from uuid import uuid4
 
 # Output to frontend
 class Landmark(BaseModel) :
-    
+
     # Properties of the landmark
     name : str
     type: Literal['sightseeing', 'nature', 'shopping', 'start', 'finish']
@@ -22,22 +22,22 @@ class Landmark(BaseModel) :
 
     # Unique ID of a given landmark
     uuid: str = Field(default_factory=uuid4)
-    
+
     # Additional properties depending on specific tour
     must_do : Optional[bool] = False
     must_avoid : Optional[bool] = False
     is_secondary : Optional[bool] = False                       # TODO future   
-    
+
     time_to_reach_next : Optional[int] = 0
     next_uuid : Optional[str] = None
-    
+
     def __str__(self) -> str:
         time_to_next_str = f", time_to_next={self.time_to_reach_next}" if self.time_to_reach_next else ""
         is_secondary_str = f", secondary" if self.is_secondary else ""
         type_str = '(' + self.type + ')'
         if self.type in ["start", "finish", "nature", "shopping"] : type_str += '\t '
         return f'Landmark{type_str}: [{self.name} @{self.location}, score={self.attractiveness}{time_to_next_str}{is_secondary_str}]'
-    
+
     def distance(self, value: 'Landmark') -> float:
         return (self.location[0] - value.location[0])**2 + (self.location[1] - value.location[1])**2
 

backend/src/structs/trip.py

@@ -22,7 +22,8 @@ class Trip(BaseModel):
 
         # Store the trip in the cache
         cache_client.set(f"trip_{trip.uuid}", trip)
-        cache_client.set_many({f"landmark_{landmark.uuid}": landmark for landmark in landmarks}, expire=3600)
+        # make sure to await the result (noreply=False). Otherwise the cache might not be inplace when the trip is actually requested
+        cache_client.set_many({f"landmark_{landmark.uuid}": landmark for landmark in landmarks}, expire=3600, noreply=False)
         # is equivalent to:
         # for landmark in landmarks:
         #     cache_client.set(f"landmark_{landmark.uuid}", landmark, expire=3600)

backend/src/utils/get_time_separation.py

@@ -1,5 +1,5 @@
 import yaml
-from geopy.distance import geodesic
+from math import sin, cos, sqrt, atan2, radians
 
 import constants
 
@@ -8,6 +8,7 @@ with constants.OPTIMIZER_PARAMETERS_PATH.open('r') as f:
     DETOUR_FACTOR = parameters['detour_factor']
     AVERAGE_WALKING_SPEED = parameters['average_walking_speed']
 
+EARTH_RADIUS_KM = 6373
 
 def get_time(p1: tuple[float, float], p2: tuple[float, float]) -> int:
     """
@@ -22,16 +23,28 @@ def get_time(p1: tuple[float, float], p2: tuple[float, float]) -> int:
     """
 
 
-    # Compute the straight-line distance in km
+    if p1 == p2:
-    if p1 == p2 :
         return 0
-    else: 
+    else:
-        dist = geodesic(p1, p2).kilometers
+        # Compute the distance in km along the surface of the Earth
+        # (assume spherical Earth)
+        # this is the haversine formula, stolen from stackoverflow
+        # in order to not use any external libraries
+        lat1, lon1 = radians(p1[0]), radians(p1[1])
+        lat2, lon2 = radians(p2[0]), radians(p2[1])
 
-    # Consider the detour factor for average cityto deterline walking distance (in km)
+        dlon = lon2 - lon1
-    walk_dist = dist*DETOUR_FACTOR
+        dlat = lat2 - lat1
+
+        a = sin(dlat / 2)**2 + cos(lat1) * cos(lat2) * sin(dlon / 2)**2
+        c = 2 * atan2(sqrt(a), sqrt(1 - a))
+
+        distance = EARTH_RADIUS_KM * c
+
+    # Consider the detour factor for average an average city
+    walk_distance = distance * DETOUR_FACTOR
 
     # Time to walk this distance (in minutes)
-    walk_time = walk_dist/AVERAGE_WALKING_SPEED*60
+    walk_time = walk_distance / AVERAGE_WALKING_SPEED * 60
 
     return round(walk_time)

backend/src/utils/optimizer.py

@@ -3,7 +3,6 @@ import numpy as np
 
 from scipy.optimize import linprog
 from collections import defaultdict, deque
-from geopy.distance import geodesic
 
 from structs.landmark import Landmark
 from .get_time_separation import get_time