update clients (#880)

* update clients

* new: update conversions

* fix: update inspection cache

* update grpc

* bump fastembed to 0.5.1

* update poetry lock

* update poetry lock

pyproject.toml

@@ -27,10 +27,10 @@ grpcio-tools = ">=1.41.0"
 urllib3 = ">=1.26.14,<3"
 portalocker = "^2.7.0"
 fastembed = [
-    { version = "0.5.0", optional = true },
+    { version = "0.5.1", optional = true },
 ]
 fastembed-gpu = [
-    { version = "0.5.0", optional = true },
+    { version = "0.5.1", optional = true },
 ]
 
 [tool.poetry.group.dev.dependencies]

qdrant_client/conversions/conversion.py

@@ -2056,6 +2056,12 @@ def convert_strict_mode_config(cls, model: grpc.StrictModeConfig) -> rest.Strict
             max_collection_payload_size_bytes=model.max_collection_payload_size_bytes
             if model.HasField("max_collection_payload_size_bytes")
             else None,
+            filter_max_conditions=model.filter_max_conditions
+            if model.HasField("filter_max_conditions")
+            else None,
+            condition_max_size=model.condition_max_size
+            if model.HasField("condition_max_size")
+            else None,
         )
 
 
@@ -3997,4 +4003,6 @@ def convert_strict_mode_config(cls, model: rest.StrictModeConfig) -> grpc.Strict
             read_rate_limit=model.read_rate_limit,
             write_rate_limit=model.write_rate_limit,
             max_collection_payload_size_bytes=model.max_collection_payload_size_bytes,
+            filter_max_conditions=model.filter_max_conditions,
+            condition_max_size=model.condition_max_size,
         )

qdrant_client/embed/_inspection_cache.py

@@ -215,6 +215,9 @@
     "InferenceObject": [],
     "StrictModeConfig": [],
     "HasVectorCondition": [],
+    "AbortReshardingOperation": [],
+    "StartResharding": [],
+    "StartReshardingOperation": [],
 }
 DEFS = {
     "AbortShardTransfer": {
@@ -3048,6 +3051,46 @@
         "title": "UpsertOperation",
         "type": "object",
     },
+    "ReshardingDirectionOneOf": {
+        "description": "Scale up, add a new shard",
+        "enum": ["up"],
+        "title": "ReshardingDirectionOneOf",
+        "type": "string",
+    },
+    "ReshardingDirectionOneOf1": {
+        "description": "Scale down, remove a shard",
+        "enum": ["down"],
+        "title": "ReshardingDirectionOneOf1",
+        "type": "string",
+    },
+    "StartResharding": {
+        "additionalProperties": False,
+        "properties": {
+            "direction": {
+                "anyOf": [
+                    {"$ref": "#/$defs/ReshardingDirectionOneOf"},
+                    {"$ref": "#/$defs/ReshardingDirectionOneOf1"},
+                ],
+                "description": "",
+                "title": "Direction",
+            },
+            "peer_id": {
+                "anyOf": [{"type": "integer"}, {"type": "null"}],
+                "default": None,
+                "description": "",
+                "title": "Peer Id",
+            },
+            "shard_key": {
+                "anyOf": [{"type": "integer"}, {"type": "string"}, {"type": "null"}],
+                "default": None,
+                "description": "",
+                "title": "Shard Key",
+            },
+        },
+        "required": ["direction"],
+        "title": "StartResharding",
+        "type": "object",
+    },
 }
 RECURSIVE_REFS = ["Filter", "MinShould", "Nested", "NestedCondition", "Prefetch"]
 INCLUDED_RECURSIVE_REFS = ["Prefetch"]

qdrant_client/http/api/service_api.py

@@ -164,13 +164,16 @@ def _build_for_root(
     def _build_for_telemetry(
         self,
         anonymize: bool = None,
+        details_level: int = None,
     ):
         """
         Collect telemetry data including app info, system info, collections info, cluster info, configs and statistics
         """
         query_params = {}
         if anonymize is not None:
             query_params["anonymize"] = str(anonymize).lower()
+        if details_level is not None:
+            query_params["details_level"] = str(details_level)
 
         headers = {}
         return self.api_client.request(
@@ -248,12 +251,14 @@ async def root(
     async def telemetry(
         self,
         anonymize: bool = None,
+        details_level: int = None,
     ) -> m.InlineResponse2001:
         """
         Collect telemetry data including app info, system info, collections info, cluster info, configs and statistics
         """
         return await self._build_for_telemetry(
             anonymize=anonymize,
+            details_level=details_level,
         )
 
 
@@ -323,10 +328,12 @@ def root(
     def telemetry(
         self,
         anonymize: bool = None,
+        details_level: int = None,
     ) -> m.InlineResponse2001:
         """
         Collect telemetry data including app info, system info, collections info, cluster info, configs and statistics
         """
         return self._build_for_telemetry(
             anonymize=anonymize,
+            details_level=details_level,
         )

qdrant_client/http/models/models.py

@@ -11,6 +11,10 @@
 VectorsConfigDiff = Dict[str, "VectorParamsDiff"]
 
 
+class AbortReshardingOperation(BaseModel, extra="forbid"):
+    abort_resharding: Any = Field(..., description="")
+
+
 class AbortShardTransfer(BaseModel, extra="forbid"):
     shard_id: int = Field(..., description="")
     to_peer_id: int = Field(..., description="")
@@ -839,6 +843,8 @@ class HardwareUsage(BaseModel):
     """
 
     cpu: int = Field(..., description="Usage of the hardware resources, spent to process the request")
+    io_read: int = Field(..., description="Usage of the hardware resources, spent to process the request")
+    io_write: int = Field(..., description="Usage of the hardware resources, spent to process the request")
 
 
 class HasIdCondition(BaseModel, extra="forbid"):
@@ -2001,6 +2007,8 @@ def __str__(self) -> str:
     LISTENER = "Listener"
     PARTIALSNAPSHOT = "PartialSnapshot"
     RECOVERY = "Recovery"
+    RESHARDING = "Resharding"
+    RESHARDINGSCALEDOWN = "ReshardingScaleDown"
 
 
 class ReplicateShard(BaseModel, extra="forbid"):
@@ -2048,10 +2056,6 @@ class ReshardingInfo(BaseModel):
     shard_id: int = Field(..., description="")
     peer_id: int = Field(..., description="")
     shard_key: Optional["ShardKey"] = Field(default=None, description="")
-    comment: Optional[str] = Field(
-        default=None,
-        description="A human-readable report of the operation progress. Available only on the source peer.",
-    )
 
 
 class RestartTransfer(BaseModel, extra="forbid"):
@@ -2400,6 +2404,10 @@ def __str__(self) -> str:
 
 class ShardTransferInfo(BaseModel):
     shard_id: int = Field(..., description="")
+    to_shard_id: Optional[int] = Field(
+        default=None,
+        description="Target shard ID if different than source shard ID  Used exclusively with `ReshardStreamRecords` transfer method.",
+    )
     from_: int = Field(..., description="Source peer id", alias="from")
     to: int = Field(..., description="Destination peer id")
     sync: bool = Field(
@@ -2445,6 +2453,17 @@ def __str__(self) -> str:
     WAL_DELTA = "wal_delta"
 
 
+class ShardTransferMethodOneOf3(str, Enum):
+    """
+    Shard transfer for resharding: stream all records in batches until all points are transferred.
+    """
+
+    def __str__(self) -> str:
+        return str(self.value)
+
+    RESHARDING_STREAM_RECORDS = "resharding_stream_records"
+
+
 class ShardingMethod(str, Enum):
     AUTO = "auto"
     CUSTOM = "custom"
@@ -2609,6 +2628,16 @@ def __str__(self) -> str:
     MMAP = "mmap"
 
 
+class StartResharding(BaseModel, extra="forbid"):
+    direction: "ReshardingDirection" = Field(..., description="")
+    peer_id: Optional[int] = Field(default=None, description="")
+    shard_key: Optional["ShardKey"] = Field(default=None, description="")
+
+
+class StartReshardingOperation(BaseModel, extra="forbid"):
+    start_resharding: "StartResharding" = Field(..., description="")
+
+
 class StateRole(str, Enum):
     """
     Role of the peer in the consensus
@@ -2653,6 +2682,10 @@ class StrictModeConfig(BaseModel, extra="forbid"):
     max_collection_payload_size_bytes: Optional[int] = Field(
         default=None, description="Max size of a collections payload storage in bytes"
     )
+    filter_max_conditions: Optional[int] = Field(default=None, description="Max conditions a filter can have.")
+    condition_max_size: Optional[int] = Field(
+        default=None, description="Max size of a condition, eg. items in `MatchAny`."
+    )
 
 
 class TelemetryData(BaseModel):
@@ -2993,6 +3026,8 @@ def __str__(self) -> str:
     CreateShardingKeyOperation,
     DropShardingKeyOperation,
     RestartTransferOperation,
+    StartReshardingOperation,
+    AbortReshardingOperation,
 ]
 ClusterStatus = Union[
     ClusterStatusOneOf,
@@ -3139,6 +3174,7 @@ def __str__(self) -> str:
     ShardTransferMethodOneOf,
     ShardTransferMethodOneOf1,
     ShardTransferMethodOneOf2,
+    ShardTransferMethodOneOf3,
 ]
 SparseIndexType = Union[
     SparseIndexTypeOneOf,

qdrant_client/proto/collections.proto

@@ -340,6 +340,8 @@ message StrictModeConfig {
   optional uint32 read_rate_limit = 11; // Max number of read operations per minute per replica
   optional uint32 write_rate_limit = 12; // Max number of write operations per minute per replica
   optional uint64 max_collection_payload_size_bytes = 13;
+  optional uint64 filter_max_conditions = 14;
+  optional uint64 condition_max_size = 15;
 }
 
 message CreateCollection {
@@ -589,6 +591,15 @@ message ReshardingInfo {
   uint32 shard_id = 1;
   uint64 peer_id = 2;
   optional ShardKey shard_key = 3;
+  ReshardingDirection direction = 4;
+}
+
+/*
+  Resharding direction, scale up or down in number of shards
+*/
+enum ReshardingDirection {
+  Up = 0; // Scale up, add a new shard
+  Down = 1; // Scale down, remove a shard
 }
 
 message CollectionClusterInfoResponse {
@@ -597,8 +608,7 @@ message CollectionClusterInfoResponse {
   repeated LocalShardInfo local_shards = 3; // Local shards
   repeated RemoteShardInfo remote_shards = 4; // Remote shards
   repeated ShardTransferInfo shard_transfers = 5; // Shard transfers
-  // TODO(resharding): enable on release:
-  // repeated ReshardingInfo resharding_operations = 6; // Resharding operations
+  repeated ReshardingInfo resharding_operations = 6; // Resharding operations
 }
 
 message MoveShard {

qdrant_client/proto/points.proto

@@ -1098,4 +1098,6 @@ message GeoPoint {
 
 message HardwareUsage {
   uint64 cpu = 1;
+  uint64 io_read = 2;
+  uint64 io_write = 3;
 }

tests/conversions/fixtures.py

@@ -286,6 +286,8 @@
     # read_rate_limit=model.read_rate_limit, test empty field
     write_rate_limit=2000,
     max_collection_payload_size_bytes=10 * 1024 * 1024 * 1024,
+    filter_max_conditions=100,
+    condition_max_size=5,
 )
 
 collection_config = grpc.CollectionConfig(