Overview
Transports define where Speckle objects are stored and retrieved from. They provide a consistent interface for operations.send() and operations.receive().
from specklepy.transports.server import ServerTransport
transport = ServerTransport(stream_id="project_id", client=client)
ServerTransport
Sends and receives objects from a Speckle server. This is the primary transport for production use.
Constructor
ServerTransport(
stream_id: str,
client: Optional[SpeckleClient] = None,
account: Optional[Account] = None,
token: Optional[str] = None,
url: Optional[str] = None,
name: str = "RemoteTransport"
)
client
SpeckleClient
default:"None"
Authenticated client
Token (use with url parameter)
Server URL (use with token parameter)
name
str
default:"\"RemoteTransport\""
Transport identifier
Provide one authentication method: client, account, or token + url.
Usage
from specklepy.api import operations
from specklepy.transports.server import ServerTransport
# Create transport
transport = ServerTransport(stream_id="project_id", client=client)
# Send
object_id = operations.send(my_object, transports=[transport])
# Receive
obj = operations.receive(object_id, remote_transport=transport)
Authentication Options
Using Client (Recommended):
transport = ServerTransport(stream_id="project_id", client=client)
transport = ServerTransport(
stream_id="project_id",
token="your_token",
url="https://app.speckle.systems"
)
from specklepy.api.credentials import get_default_account
account = get_default_account()
transport = ServerTransport(stream_id="project_id", account=account)
MemoryTransport
Stores objects in memory. Useful for testing and temporary operations.
from specklepy.transports.memory import MemoryTransport
transport = MemoryTransport()
objects (Dict[str, str]) - Dictionary of stored objectssaved_object_count (int) - Number of objects saved
Data is lost when the application exits. Not suitable for production.
SQLiteTransport
Stores objects in a local SQLite database for persistent local storage.
from specklepy.transports.sqlite import SQLiteTransport
transport = SQLiteTransport(
app_name="MyApp",
scope="ProjectData"
)
Custom directory for database
- Windows:
%APPDATA%\{app_name}\{scope}.db
- macOS:
~/Library/Application Support/{app_name}/{scope}.db
- Linux:
~/.config/{app_name}/{scope}.db
Use cases:
- Local caching of server data
- Offline workflows
- Persistent local storage
Multiple Transports
Send to multiple locations simultaneously:
from specklepy.transports.server import ServerTransport
from specklepy.transports.sqlite import SQLiteTransport
server = ServerTransport(stream_id="project_id", client=client)
local = SQLiteTransport(app_name="MyApp")
# Send to both server and local cache
object_id = operations.send(my_object, transports=[server, local])
# Receive with local cache (checks local first)
obj = operations.receive(
object_id,
remote_transport=server,
local_transport=local
)
Custom Transports
Create custom transports by extending AbstractTransport:
from specklepy.transports.abstract_transport import AbstractTransport
class CustomTransport(AbstractTransport):
def save_object(self, id: str, serialized_object: str) -> None:
# Implement storage logic
pass
def get_object(self, id: str) -> Optional[str]:
# Implement retrieval logic
pass
# Implement other required methods...
Comparison
| Transport | Persistence | Network | Primary Use |
|---|
| ServerTransport | Server | Required | Production |
| MemoryTransport | None | No | Testing |
| SQLiteTransport | Local disk | No | Caching, offline |
