Geo polygons support (#325)

* regenerate client

* regen grpc client

* document build process

* work from Zein in #272

* update rest client with optional field

* code refactor and better naming

* remove shapely as a dependency

* fix typing

---------

Co-authored-by: zzzz-vincent <wenzishen.vincent@hotmail.com>
Co-authored-by: generall <andrey@vasnetsov.com>

Author: 3 people

.gitignore

@@ -11,3 +11,4 @@ htmlcov
 dist
 *.tar.gz
 local_cache/*/*
+.python-version

poetry.lock

@@ -4,6 +4,7 @@
 from qdrant_client._pydantic_compat import update_forward_refs
 from qdrant_client.http.api_client import (  # noqa F401
     ApiClient as ApiClient,
+    AsyncApiClient as AsyncApiClient,
     AsyncApis as AsyncApis,
     SyncApis as SyncApis,
 )

qdrant_client/grpc/points_pb2.py

@@ -89,7 +89,7 @@ def request_sync(self, *, type_: Any, **kwargs: Any) -> Any:  # noqa F811
 
     def send(self, request: Request, type_: Type[T]) -> T:
         response = self.middleware(request, self.send_inner)
-        if response.status_code in [200, 201]:
+        if response.status_code in [200, 201, 202]:
             try:
                 return parse_as_type(response.json(), type_)
             except ValidationError as e:
@@ -161,7 +161,7 @@ def request_sync(self, *, type_: Any, **kwargs: Any) -> Any:  # noqa F811
 
     async def send(self, request: Request, type_: Type[T]) -> T:
         response = await self.middleware(request, self.send_inner)
-        if response.status_code in [200, 201]:
+        if response.status_code in [200, 201, 202]:
             try:
                 return parse_as_type(response.json(), type_)
             except ValidationError as e:

qdrant_client/http/__init__.py

@@ -432,6 +432,9 @@ class FieldCondition(BaseModel, extra="forbid"):
         default=None, description="Check if points geo location lies in a given area"
     )
     geo_radius: Optional["GeoRadius"] = Field(default=None, description="Check if geo point is within a given radius")
+    geo_polygon: Optional["GeoPolygon"] = Field(
+        default=None, description="Check if geo point is within a given polygon"
+    )
     values_count: Optional["ValuesCount"] = Field(default=None, description="Check number of values of the field")
 
 
@@ -462,6 +465,14 @@ class GeoBoundingBox(BaseModel, extra="forbid"):
     )
 
 
+class GeoLineString(BaseModel, extra="forbid"):
+    """
+    Ordered sequence of GeoPoints representing the line
+    """
+
+    points: List["GeoPoint"] = Field(..., description="Ordered sequence of GeoPoints representing the line")
+
+
 class GeoPoint(BaseModel, extra="forbid"):
     """
     Geo point payload schema
@@ -471,6 +482,21 @@ class GeoPoint(BaseModel, extra="forbid"):
     lat: float = Field(..., description="Geo point payload schema")
 
 
+class GeoPolygon(BaseModel, extra="forbid"):
+    """
+    Geo filter request  Matches coordinates inside the polygon, defined by `exterior` and `interiors`
+    """
+
+    exterior: "GeoLineString" = Field(
+        ...,
+        description="Geo filter request  Matches coordinates inside the polygon, defined by `exterior` and `interiors`",
+    )
+    interiors: Optional[List["GeoLineString"]] = Field(
+        default=None,
+        description="Interior lines (if present) bound holes within the surface each GeoLineString must consist of a minimum of 4 points, and the first and last points must be the same.",
+    )
+
+
 class GeoRadius(BaseModel, extra="forbid"):
     """
     Geo filter request  Matches coordinates inside the circle of `radius` and center with coordinates `center`
@@ -1618,7 +1644,7 @@ class UpdateCollection(BaseModel, extra="forbid"):
 
     vectors: Optional["VectorsConfigDiff"] = Field(
         default=None,
-        description="Vector data parameters to update. It is possible to provide one config for single vector mode and list of configs for multiple vectors mode.",
+        description="Map of vector data parameters to update for each named vector. To update parameters in a collection having a single unnamed vector, use an empty string as name.",
     )
     optimizers_config: Optional["OptimizersConfigDiff"] = Field(
         default=None,

qdrant_client/http/api_client.py

@@ -1,4 +1,5 @@
 from math import asin, cos, radians, sin, sqrt
+from typing import List, Tuple
 
 
 def geo_distance(lon1: float, lat1: float, lon2: float, lat2: float) -> float:
@@ -38,3 +39,51 @@ def test_geo_distance() -> None:
     assert geo_distance(moscow["lon"], moscow["lat"], london["lon"], london["lat"]) < 2600 * 1000
     assert geo_distance(moscow["lon"], moscow["lat"], berlin["lon"], berlin["lat"]) > 1600 * 1000
     assert geo_distance(moscow["lon"], moscow["lat"], berlin["lon"], berlin["lat"]) < 1650 * 1000
+
+
+def boolean_point_in_polygon(
+    point: Tuple[float, float],
+    exterior: List[Tuple[float, float]],
+    interiors: List[List[Tuple[float, float]]],
+) -> bool:
+    inside_poly = False
+
+    if in_ring(point, exterior, True):
+        in_hole = False
+        k = 0
+        while k < len(interiors) and not in_hole:
+            if in_ring(point, interiors[k], False):
+                in_hole = True
+            k += 1
+        if not in_hole:
+            inside_poly = True
+
+    return inside_poly
+
+
+def in_ring(
+    pt: Tuple[float, float], ring: List[Tuple[float, float]], ignore_boundary: bool
+) -> bool:
+    is_inside = False
+    if ring[0][0] == ring[len(ring) - 1][0] and ring[0][1] == ring[len(ring) - 1][1]:
+        ring = ring[0 : len(ring) - 1]
+    j = len(ring) - 1
+    for i in range(0, len(ring)):
+        xi = ring[i][0]
+        yi = ring[i][1]
+        xj = ring[j][0]
+        yj = ring[j][1]
+        on_boundary = (
+            (pt[1] * (xi - xj) + yi * (xj - pt[0]) + yj * (pt[0] - xi) == 0)
+            and ((xi - pt[0]) * (xj - pt[0]) <= 0)
+            and ((yi - pt[1]) * (yj - pt[1]) <= 0)
+        )
+        if on_boundary:
+            return not ignore_boundary
+        intersect = ((yi > pt[1]) != (yj > pt[1])) and (
+            pt[0] < (xj - xi) * (pt[1] - yi) / (yj - yi) + xi
+        )
+        if intersect:
+            is_inside = not is_inside
+        j = i
+    return is_inside

qdrant_client/http/models/models.py

@@ -3,7 +3,7 @@
 import numpy as np
 
 from qdrant_client.http import models
-from qdrant_client.local.geo import geo_distance
+from qdrant_client.local.geo import boolean_point_in_polygon, geo_distance
 from qdrant_client.local.payload_value_extractor import value_by_key
 
 
@@ -70,6 +70,22 @@ def check_geo_bounding_box(condition: models.GeoBoundingBox, values: Any) -> boo
     return False
 
 
+def check_geo_polygon(condition: models.GeoPolygon, values: Any) -> bool:
+    if isinstance(values, dict) and "lat" in values and "lon" in values:
+        lat = values["lat"]
+        lon = values["lon"]
+        exterior = [(point.lat, point.lon) for point in condition.exterior.points]
+        interiors = []
+        if condition.interiors is not None:
+            interiors = [
+                [(point.lat, point.lon) for point in interior.points]
+                for interior in condition.interiors
+            ]
+        return boolean_point_in_polygon(point=(lat, lon), exterior=exterior, interiors=interiors)
+
+    return False
+
+
 def check_range(condition: models.Range, value: Any) -> bool:
     if not isinstance(value, (int, float)):
         return False
@@ -141,6 +157,10 @@ def check_condition(
         if condition.values_count is not None:
             values = value_by_key(payload, condition.key, flat=False)
             return check_values_count(condition.values_count, values)
+        if condition.geo_polygon is not None:
+            if values is None:
+                return False
+            return any(check_geo_polygon(condition.geo_polygon, v) for v in values)
     elif isinstance(condition, models.NestedCondition):
         values = value_by_key(payload, condition.nested.key)
         if values is None:

qdrant_client/local/geo.py

@@ -140,3 +140,50 @@ def test_nested_payload_filters():
 
     res = check_filter(query, payload, 0)
     assert res is False
+    
+
+def test_geo_polygon_filter_query():
+    payload = {
+        "location": [
+            {
+                "lon": 70.0,
+                "lat": 70.0,
+            },
+        ]
+    }
+
+    query = models.Filter(
+        **{
+            "must": [
+                {
+                    "key": "location",
+                    "geo_polygon": {
+                        "exterior": {
+                            "points": [
+                                {"lon": 55.455868, "lat": 55.495862},
+                                {"lon": 86.455868, "lat": 55.495862},
+                                {"lon": 86.455868, "lat": 86.495862},
+                                {"lon": 55.455868, "lat": 86.495862},
+                                {"lon": 55.455868, "lat": 55.495862},
+                            ]
+                        },
+                    },
+                }
+            ]
+        }
+    )
+
+    res = check_filter(query, payload, 0)
+    assert res is True
+
+    payload = {
+        "location": [
+            {
+                "lon": 30.693738,
+                "lat": 30.502165,
+            },
+        ]
+    }
+
+    res = check_filter(query, payload, 0)
+    assert res is False

qdrant_client/local/payload_filters.py

@@ -159,7 +159,7 @@ message OptimizersConfigDiff {
   Do not create segments larger this size (in kilobytes).
   Large segments might require disproportionately long indexation times,
   therefore it makes sense to limit the size of segments.
-
+    
   If indexing speed is more important - make this parameter lower.
   If search speed is more important - make this parameter higher.
   Note: 1Kb = 1 vector of size 256
@@ -179,11 +179,11 @@ message OptimizersConfigDiff {
   optional uint64 memmap_threshold = 5;
   /*
   Maximum size (in kilobytes) of vectors allowed for plain index, exceeding this threshold will enable vector indexing
-
+  
   Default value is 20,000, based on <https://github.com/google-research/google-research/blob/master/scann/docs/algorithms.md>.
-
+  
   To disable vector indexing, set to `0`.
-
+  
   Note: 1kB = 1 vector of size 256.
   */
   optional uint64 indexing_threshold = 6;
@@ -412,7 +412,7 @@ message ShardTransferInfo {
 }
 
 message CollectionClusterInfoResponse {
-  uint64 peer_id = 1;  // ID of this peer
+  uint64 peer_id = 1;  // ID of this peer 
   uint64 shard_count = 2; // Total number of shards
   repeated LocalShardInfo local_shards = 3; // Local shards
   repeated RemoteShardInfo remote_shards = 4; // Remote shards