Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 34 additions & 3 deletions weaviate_cli/commands/create.py
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@
CreateDataDefaults,
CreateRoleDefaults,
PERMISSION_HELP_STRING,
MAX_WORKERS,
)


Expand Down Expand Up @@ -89,7 +90,7 @@ def create() -> None:
@click.option(
"--training_limit",
default=CreateCollectionDefaults.training_limit,
help="Training limit for PQ and SQ (default: 10000).",
help=f"Training limit for PQ and SQ (default: {CreateCollectionDefaults.training_limit}).",
)
@click.option(
"--multitenant", is_flag=True, help="Enable multitenancy (default: False)."
Expand Down Expand Up @@ -228,7 +229,7 @@ def create_collection_cli(
@click.option(
"--number_tenants",
default=CreateTenantsDefaults.number_tenants,
help="Number of tenants to create (default: 100).",
help=f"Number of tenants to create (default: {CreateTenantsDefaults.number_tenants}).",
)
@click.option(
"--tenant_batch_size",
Expand Down Expand Up @@ -336,7 +337,8 @@ def create_backup_cli(ctx, backend, backup_id, include, exclude, wait, cpu_for_b
@click.option(
"--limit",
default=CreateDataDefaults.limit,
help="Number of objects to import (default: 1000).",
help=f"Number of objects to import (default: {CreateDataDefaults.limit}).",
type=int,
)
@click.option(
"--consistency_level",
Expand Down Expand Up @@ -390,6 +392,23 @@ def create_backup_cli(ctx, backend, backup_id, include, exclude, wait, cpu_for_b
is_flag=True,
help="Enable multi-vector (default: False).",
)
@click.option(
"--dynamic_batch",
is_flag=True,
help="Enable dynamic batching (default: False).",
)
@click.option(
"--batch_size",
default=CreateDataDefaults.batch_size,
help=f"Number of objects to ingest in each batch (default: {CreateDataDefaults.batch_size}).",
type=int,
)
@click.option(
"--concurrent_requests",
default=MAX_WORKERS,
type=int,
help=f"Number of concurrent requests to send to the server (default: {MAX_WORKERS}).",
)
@click.pass_context
def create_data_cli(
ctx,
Expand All @@ -406,6 +425,9 @@ def create_data_cli(
wait_for_indexing,
verbose,
multi_vector,
dynamic_batch,
batch_size,
concurrent_requests,
):
"""Ingest data into a collection in Weaviate."""

Expand All @@ -423,6 +445,12 @@ def create_data_cli(
click.echo("Error: --uuid has no effect unless --limit=1 is enabled.")
sys.exit(1)

if dynamic_batch and not randomize:
click.echo(
"Error: --dynamic_batch has no effect unless --randomize is enabled."
)
sys.exit(1)

client: Optional[WeaviateClient] = None
try:
client = get_client_from_context(ctx)
Expand All @@ -442,6 +470,9 @@ def create_data_cli(
wait_for_indexing=wait_for_indexing,
verbose=verbose,
multi_vector=multi_vector,
dynamic_batch=dynamic_batch,
batch_size=batch_size,
concurrent_requests=concurrent_requests,
)
except Exception as e:
click.echo(f"Error: {e}")
Expand Down
11 changes: 11 additions & 0 deletions weaviate_cli/defaults.py
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
from dataclasses import dataclass, field
import multiprocessing
from typing import Optional, List, Dict


Expand Down Expand Up @@ -50,6 +51,14 @@
QUERY_MAXIMUM_RESULTS = 10000
MAX_OBJECTS_PER_BATCH = 5000

try:
_CPU_COUNT = multiprocessing.cpu_count()
except (NotImplementedError, OSError):
_CPU_COUNT = None
# Fallback to 1 worker if CPU count is unavailable or invalid
_SAFE_CPU_COUNT = _CPU_COUNT if isinstance(_CPU_COUNT, int) and _CPU_COUNT > 0 else 1
MAX_WORKERS = min(32, _SAFE_CPU_COUNT + 4)


@dataclass
class CreateCollectionDefaults:
Expand Down Expand Up @@ -102,6 +111,8 @@ class CreateDataDefaults:
wait_for_indexing: bool = False
verbose: bool = False
multi_vector: bool = False
batch_size: int = 1000
dynamic_batch: bool = False


@dataclass
Expand Down
24 changes: 24 additions & 0 deletions weaviate_cli/managers/config_manager.py
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@
from weaviate.exceptions import WeaviateGRPCUnavailableError
from pathlib import Path
from typing import Dict, Optional, Union
from weaviate.config import AdditionalConfig, Timeout as WeaviateTimeout


class ConfigManager:
Expand Down Expand Up @@ -107,6 +108,9 @@ def _create_client(
) -> Union[weaviate.WeaviateClient, weaviate.WeaviateAsyncClient]:
"""Internal method to create client based on async flag"""
auth_config: Optional[weaviate.auth.AuthCredentials] = None
additional_config: Optional[AdditionalConfig] = (
self.__additional_config_if_slow()
)

if "auth" in self.config:
if self.config["auth"].get("type") == "user":
Expand All @@ -133,6 +137,7 @@ def _create_client(
grpc_port=self.config["grpc_port"],
auth_credentials=auth_config,
headers=headers,
additional_config=additional_config,
)
if async_client:
return weaviate.use_async_with_local(**common_kwargs)
Expand All @@ -144,6 +149,7 @@ def _create_client(
cluster_url=self.config["host"],
auth_credentials=auth_config,
headers=headers,
additional_config=additional_config,
)
if async_client:
try:
Expand Down Expand Up @@ -189,6 +195,7 @@ def _create_client(
grpc_port=grpc_port,
auth_credentials=auth_config,
headers=headers,
additional_config=additional_config,
)
else:
common_kwargs = dict(
Expand All @@ -200,8 +207,25 @@ def _create_client(
grpc_port=self.config["grpc_port"],
auth_credentials=auth_config,
headers=headers,
additional_config=additional_config,
)

if async_client:
return weaviate.use_async_with_custom(**common_kwargs)
return weaviate.connect_to_custom(**common_kwargs)

def __additional_config_if_slow(self) -> Optional[AdditionalConfig]:
"""Return AdditionalConfig with doubled timeouts when SLOW_CONNECTION is set."""
env_flag = os.getenv("SLOW_CONNECTION")
if not env_flag:
return None
if str(env_flag).strip().lower() not in {"1", "true", "yes", "on"}:
return None
defaults = AdditionalConfig().timeout
return AdditionalConfig(
timeout=WeaviateTimeout(
query=defaults.query * 2,
insert=defaults.insert * 2,
init=defaults.init * 2,
)
)
Loading