Merge pull request #12 from ParallelDots/typehints-fix

Fix: fix typehints to use comment typehints.

retailtree/logics/overlap.py

@@ -1,4 +1,8 @@
+from retailtree.structs.annotation_struct import Annotation
+
+
 def calculate_overlap(anno_1, anno_2, axis):
+    # type:(Annotation, Annotation, str) -> float
     """
     Calculate overlap of two annotations in given axis.
     Parameters

retailtree/logics/right_left.py

@@ -1,8 +1,10 @@
+from retailtree.structs.annotation_struct import Annotation
 from retailtree.logics.overlap import calculate_overlap
 
 
 # Establishing right-left connections
 def right_left_connections(i, j, given_overlap_percentage):
+    # type:(Annotation, Annotation, float) -> None
     # Checking the candidate annotations is in the range of base annotations corresponds to x axis
     if (i.x_min < j.x_min and i.x_max + i.length > j.x_min):
         # Finding overlap percentage
@@ -22,6 +24,7 @@ def right_left_connections(i, j, given_overlap_percentage):
 
 # Establishing left-right connection
 def left_right_connections(i, j, given_overlap_percentage):
+    # type:(Annotation, Annotation, float) -> None
     if j.right == None:
         # Checking the candidate annotations is in the range of base annotations corresponds to x axis
         if (i.x_min >= j.x_max and i.x_min - i.length < j.x_max):

retailtree/logics/top_bottom.py

@@ -1,8 +1,10 @@
+from retailtree.structs.annotation_struct import Annotation
 from retailtree.logics.overlap import calculate_overlap
 # Establishing tob-bottom connections
 
 
 def top_bottom_connections(i, j, given_overlap_percentage):
+    # type:(Annotation, Annotation, float) -> None
     if (i.y_max > j.y_max and i.y_min - i.width < j.y_max):
         # Finding overlap percentage
         overlap_percentage = calculate_overlap(i, j, axis='x')/j.length
@@ -24,6 +26,7 @@ def top_bottom_connections(i, j, given_overlap_percentage):
 
 # Establishing tob-bottom connections
 def bottom_top_connections(i, j, given_overlap_percentage):
+    # type:(Annotation, Annotation, float) -> None
     if j.top == None:
         if (i.y_max <= j.y_min and i.y_max + i.width > j.y_min):
             # Finding overlap percentage

retailtree/logics/vp_tree.py

@@ -1,10 +1,13 @@
-from typing import Callable
 import math
 import statistics as stats
+from typing import Callable
+
+from retailtree.structs.annotation_struct import Annotation
 
 
 class VPTree:
-    def __init__(self, points, dist_fn: Callable[[tuple[float, float], tuple[float, float]], float]):
+    def __init__(self, points, dist_fn):
+        # type:(list[Annotation], Callable[[tuple[float, float], tuple[float, float]], float]) -> None
         self.left = None
         self.right = None
         self.left_min = math.inf
@@ -56,7 +59,8 @@ def _is_leaf(self):
         return (self.left is None) and (self.right is None)
 
     def get_all_in_range(self, query, max_distance):
-        neighbors = list()
+        # type: (tuple[float, float], float) -> list[tuple[float, Annotation]]
+        neighbors = list() # type: list[tuple[float, Annotation]]
         nodes_to_visit = [(self, 0)]
 
         while len(nodes_to_visit) > 0:

retailtree/retailtree.py

@@ -1,16 +1,18 @@
+import math
+from random import sample
+import numpy as np
+from typing import Callable
+
 from retailtree.structs.annotation_struct import Annotation
 from retailtree.logics.vp_tree import VPTree
 from retailtree.logics.right_left import right_left_connections, left_right_connections
 from retailtree.logics.top_bottom import top_bottom_connections, bottom_top_connections
 from retailtree.utils.dist_func import euclidean
-import math
-from random import sample
-import numpy as np
 
 
 class RetailTree:
     def __init__(self) -> None:
-        self.annotations: dict[int, Annotation] = {}
+        self.annotations = {}  # type:dict[int, Annotation]
         self.tree = None
         self.__neighbors_radius = None
 
@@ -25,7 +27,8 @@ def add_annotation(self, annotation):
         """
         self.annotations[annotation.id] = annotation
 
-    def get(self, id: int):
+    def get(self, id):
+        # type:(int) -> Annotation
         """
         Retrieve an annotation by its ID.
 
@@ -38,6 +41,7 @@ def get(self, id: int):
         return self.annotations[id]
 
     def build_tree(self, dist_func=euclidean):
+        # type:(Callable[[tuple[float, float], tuple[float, float]], float]) -> None
         """
         Builds a Vantage Point Tree (VPTree) using the given distance function.
 
@@ -55,6 +59,7 @@ def build_tree(self, dist_func=euclidean):
         self.tree = obj
 
     def __get_neighbors_radius(self):
+        # type:() -> float
         """
         Method to get the radius within which neighbors will be searched for.
         Radius is the max of the diagonals of the annotations considered.
@@ -70,14 +75,15 @@ def __get_neighbors_radius(self):
             self.__neighbors_radius = radius
         return radius
 
-    def __fetch_neighbors(self, id: int, radius=1):
-
+    def __fetch_neighbors(self, id, radius=1):
+        # type:(int, int) -> list[tuple[float, Annotation]]
         radius = radius*self.__get_neighbors_radius()
         neighbors = self.tree.get_all_in_range(
             (self.annotations[id].x_mid, self.annotations[id].y_mid), radius)
         return neighbors
 
     def __finding_angle(self, origin, neighbor):
+        # type:(Annotation, tuple[float, Annotation]) -> int
         translated_point2 = np.array(
             [neighbor[1].x_mid, neighbor[1].y_mid]) - np.array([origin.x_mid, origin.y_mid])
         # print(translated_point2)
@@ -106,7 +112,8 @@ def __fetching_ann_in_range(self, result_dict, min_angle, max_angle, result_lst)
         # if min_angle is None:
         return result_lst
 
-    def neighbors_wa(self, id: int, radius=1, amin=None, amax=None):
+    def neighbors_wa(self, id, radius=1, amin=None, amax=None):
+        # type:(int, int, float, float) -> list[dict]
         """
         Retrieves neighboring elements within a specified angle range around a given element.
 
@@ -139,7 +146,8 @@ def neighbors_wa(self, id: int, radius=1, amin=None, amax=None):
 
         return result_lst
 
-    def neighbors(self, id: int, radius=1):
+    def neighbors(self, id, radius=1):
+        # type:(int, int) -> list[dict]
         """
         Finds neighboring annotations within a specified radius of a given annotation.(Radius specified is taken as a square rather than a circle)
 
@@ -177,7 +185,8 @@ def neighbors(self, id: int, radius=1):
 
         return result_lst
 
-    def TBLR(self, id: int, radius=1, overlap=0.5):
+    def TBLR(self, id, radius=1, overlap=0.5):
+        # type:(int, int, float) -> (dict[str, int | bool] | str)
         """
         Computes top, bottom, left, and right connections for a given annotation within a specified radius.
 
@@ -189,8 +198,9 @@ def TBLR(self, id: int, radius=1, overlap=0.5):
         -   overlap (float, optional): The overlap percentage used to compute connections. Defaults to 0.5.
 
         Returns:
-        -   dict: A dictionary containing top, bottom, left, and right connections of the given annotation.
+        -   dict OR str: A dictionary containing top, bottom, left, and right connections of the given annotation.
                 Each connection is represented by the ID of the connected annotation, or False if no connection exists.
+                If the SKU is not present in the bucket, a string with value 'SKU is absent in annotation bucket' is returned.
 
         Examples:
             Example usages of TBLR:

retailtree/structs/annotation_struct.py

@@ -1,5 +1,5 @@
 
-from typing import Optional, Dict, Any
+from typing import Any
 
 
 class Annotation:
@@ -25,7 +25,8 @@ class Annotation:
 
     """
 
-    def __init__(self, id: int, x_min: float,  y_min: float, x_max: float, y_max: float, label: Optional[Any] = None, metadata: Optional[Dict[Any, Any]] = None):
+    def __init__(self, id, x_min, y_min, x_max, y_max, label=None, metadata=None):
+        # type:(int, float, float, float, float, Any , dict[Any, Any]) -> None
         self.__id = int(id)
         self.__x_min = float(x_min)
         self.__x_max = float(x_max)
@@ -40,10 +41,10 @@ def __init__(self, id: int, x_min: float,  y_min: float, x_max: float, y_max: fl
         self.__width = None
 
         # TODO Check compatibility for older versions of python
-        self.right: "Annotation" = None
-        self.left: "Annotation" = None
-        self.top: "Annotation" = None
-        self.bottom: "Annotation" = None
+        self.right = None  # type: Annotation
+        self.left = None  # type: Annotation
+        self.top = None  # type: Annotation
+        self.bottom = None  # type: Annotation
 
         self.overlap_right = 0
         self.overlap_top = 0
@@ -128,9 +129,3 @@ def get_coords(self):
 
     def __repr__(self) -> str:
         return str(self.id)
-
-
-# ann = Annotation(10, 1, 1, 1, 1)
-
-
-# print(ann.y_mid)

retailtree/utils/dist_func.py

@@ -3,9 +3,11 @@
 
 # Function to calculate euclidean distance
 def euclidean(p1, p2):
+    # type:(tuple[float, float], tuple[float, float]) -> float
     return math.sqrt(pow((p2[0]-p1[0]), 2) + pow((p2[1]-p1[1]), 2))
 
 
 # Function to calculate manhattan distance
 def manhattan(p1, p2):
+    # type:(tuple[float, float], tuple[float, float]) -> float
     return abs(p2[0] - p1[0]) + abs(p2[1] - p1[1])

setup.py

@@ -9,7 +9,7 @@ def read(fname):
 
 setup(
     name="retailtree",
-    version="1.3",
+    version="1.3.1",
     long_description=read("README.md"),
     long_description_content_type='text/markdown',
     packages=find_packages(),