trustgraph/trustgraph-base/trustgraph/api/config.py
cybermaggedon d35473f7f7
feat: workspace-based multi-tenancy, replacing user as tenancy axis (#840)
Introduces `workspace` as the isolation boundary for config, flows,
library, and knowledge data. Removes `user` as a schema-level field
throughout the code, API specs, and tests; workspace provides the
same separation more cleanly at the trusted flow.workspace layer
rather than through client-supplied message fields.

Design
------
- IAM tech spec (docs/tech-specs/iam.md) documents current state,
  proposed auth/access model, and migration direction.
- Data ownership model (docs/tech-specs/data-ownership-model.md)
  captures the workspace/collection/flow hierarchy.

Schema + messaging
------------------
- Drop `user` field from AgentRequest/Step, GraphRagQuery,
  DocumentRagQuery, Triples/Graph/Document/Row EmbeddingsRequest,
  Sparql/Rows/Structured QueryRequest, ToolServiceRequest.
- Keep collection/workspace routing via flow.workspace at the
  service layer.
- Translators updated to not serialise/deserialise user.

API specs
---------
- OpenAPI schemas and path examples cleaned of user fields.
- Websocket async-api messages updated.
- Removed the unused parameters/User.yaml.

Services + base
---------------
- Librarian, collection manager, knowledge, config: all operations
  scoped by workspace. Config client API takes workspace as first
  positional arg.
- `flow.workspace` set at flow start time by the infrastructure;
  no longer pass-through from clients.
- Tool service drops user-personalisation passthrough.

CLI + SDK
---------
- tg-init-workspace and workspace-aware import/export.
- All tg-* commands drop user args; accept --workspace.
- Python API/SDK (flow, socket_client, async_*, explainability,
  library) drop user kwargs from every method signature.

MCP server
----------
- All tool endpoints drop user parameters; socket_manager no longer
  keyed per user.

Flow service
------------
- Closure-based topic cleanup on flow stop: only delete topics
  whose blueprint template was parameterised AND no remaining
  live flow (across all workspaces) still resolves to that topic.
  Three scopes fall out naturally from template analysis:
    * {id} -> per-flow, deleted on stop
    * {blueprint} -> per-blueprint, kept while any flow of the
      same blueprint exists
    * {workspace} -> per-workspace, kept while any flow in the
      workspace exists
    * literal -> global, never deleted (e.g. tg.request.librarian)
  Fixes a bug where stopping a flow silently destroyed the global
  librarian exchange, wedging all library operations until manual
  restart.

RabbitMQ backend
----------------
- heartbeat=60, blocked_connection_timeout=300. Catches silently
  dead connections (broker restart, orphaned channels, network
  partitions) within ~2 heartbeat windows, so the consumer
  reconnects and re-binds its queue rather than sitting forever
  on a zombie connection.

Tests
-----
- Full test refresh: unit, integration, contract, provenance.
- Dropped user-field assertions and constructor kwargs across
  ~100 test files.
- Renamed user-collection isolation tests to workspace-collection.
2026-04-21 23:23:01 +01:00

335 lines
9 KiB
Python

"""
TrustGraph Configuration Management
This module provides interfaces for managing TrustGraph configuration settings,
including retrieving, updating, and deleting configuration values.
"""
import logging
from . exceptions import *
from . types import ConfigValue
logger = logging.getLogger(__name__)
class Config:
"""
Configuration management client.
Provides methods for managing TrustGraph configuration settings across
different types (llm, embedding, etc.), with support for get, put, delete,
and list operations.
"""
def __init__(self, api, workspace="default"):
"""
Initialize Config client.
Args:
api: Parent Api instance for making requests
workspace: Workspace to scope all config operations to
"""
self.api = api
self.workspace = workspace
def request(self, request):
"""
Make a configuration-scoped API request.
Args:
request: Request payload dictionary
Returns:
dict: Response object
"""
return self.api.request("config", request)
def get(self, keys):
"""
Get configuration values for specified keys.
Retrieves the configuration values for one or more configuration keys.
Args:
keys: List of ConfigKey objects specifying which values to retrieve
Returns:
list[ConfigValue]: List of configuration values
Raises:
ProtocolException: If response format is invalid
Example:
```python
from trustgraph.api import ConfigKey
config = api.config()
# Get specific configuration values
values = config.get([
ConfigKey(type="llm", key="model"),
ConfigKey(type="llm", key="temperature"),
ConfigKey(type="embedding", key="model")
])
for val in values:
print(f"{val.type}.{val.key} = {val.value}")
```
"""
input = {
"operation": "get",
"workspace": self.workspace,
"keys": [
{ "type": k.type, "key": k.key }
for k in keys
]
}
object = self.request(input)
try:
return [
ConfigValue(
type = v["type"],
key = v["key"],
value = v["value"]
)
for v in object["values"]
]
except Exception as e:
logger.error("Failed to parse config get response", exc_info=True)
raise ProtocolException("Response not formatted correctly")
def put(self, values):
"""
Set configuration values.
Updates or creates configuration values for the specified keys.
Args:
values: List of ConfigValue objects with type, key, and value
Example:
```python
from trustgraph.api import ConfigValue
config = api.config()
# Set configuration values
config.put([
ConfigValue(type="llm", key="model", value="gpt-4"),
ConfigValue(type="llm", key="temperature", value="0.7"),
ConfigValue(type="embedding", key="model", value="text-embedding-3-small")
])
```
"""
input = {
"operation": "put",
"workspace": self.workspace,
"values": [
{ "type": v.type, "key": v.key, "value": v.value }
for v in values
]
}
self.request(input)
def delete(self, keys):
"""
Delete configuration values.
Removes configuration values for the specified keys.
Args:
keys: List of ConfigKey objects specifying which values to delete
Example:
```python
from trustgraph.api import ConfigKey
config = api.config()
# Delete configuration values
config.delete([
ConfigKey(type="llm", key="old-setting"),
ConfigKey(type="embedding", key="deprecated")
])
```
"""
input = {
"operation": "delete",
"workspace": self.workspace,
"keys": [
{ "type": v.type, "key": v.key }
for v in keys
]
}
self.request(input)
def list(self, type):
"""
List all configuration keys for a given type.
Retrieves a list of all configuration key names within a specific
configuration type.
Args:
type: Configuration type (e.g., "llm", "embedding", "storage")
Returns:
list[str]: List of configuration key names
Example:
```python
config = api.config()
# List all LLM configuration keys
llm_keys = config.list(type="llm")
print(f"LLM configuration keys: {llm_keys}")
# List all embedding configuration keys
embedding_keys = config.list(type="embedding")
print(f"Embedding configuration keys: {embedding_keys}")
```
"""
input = {
"operation": "list",
"workspace": self.workspace,
"type": type,
}
return self.request(input)["directory"]
def get_values(self, type):
"""
Get all configuration values for a given type.
Retrieves all configuration key-value pairs within a specific
configuration type.
Args:
type: Configuration type (e.g., "llm", "embedding", "storage")
Returns:
list[ConfigValue]: List of all configuration values for the type
Raises:
ProtocolException: If response format is invalid
Example:
```python
config = api.config()
# Get all LLM configuration
llm_config = config.get_values(type="llm")
for val in llm_config:
print(f"{val.key} = {val.value}")
# Get all embedding configuration
embedding_config = config.get_values(type="embedding")
for val in embedding_config:
print(f"{val.key} = {val.value}")
```
"""
input = {
"operation": "getvalues",
"workspace": self.workspace,
"type": type,
}
object = self.request(input)
try:
return [
ConfigValue(
type = v["type"],
key = v["key"],
value = v["value"]
)
for v in object["values"]
]
except:
raise ProtocolException(f"Response not formatted correctly")
def get_values_all_workspaces(self, type):
"""
Get all configuration values of a given type across all workspaces.
Unlike get_values(), this is not scoped to a single workspace —
it returns every entry of the given type in the system. Each
returned ConfigValue includes its workspace field. Used by
shared processors to load type-scoped config at startup.
Args:
type: Configuration type (e.g. "prompt", "schema")
Returns:
list[ConfigValue]: Values across all workspaces; each has
its workspace field populated.
Raises:
ProtocolException: If response format is invalid
"""
input = {
"operation": "getvalues-all-ws",
"type": type,
}
object = self.request(input)
try:
return [
ConfigValue(
type = v["type"],
key = v["key"],
value = v["value"],
workspace = v.get("workspace", ""),
)
for v in object["values"]
]
except Exception:
raise ProtocolException("Response not formatted correctly")
def all(self):
"""
Get complete configuration and version.
Retrieves the entire configuration object along with its version number.
Returns:
tuple: (config_dict, version_string) - Complete configuration and version
Raises:
ProtocolException: If response format is invalid
Example:
```python
config = api.config()
# Get complete configuration
config_data, version = config.all()
print(f"Configuration version: {version}")
print(f"Configuration: {config_data}")
```
"""
input = {
"operation": "config",
"workspace": self.workspace,
}
object = self.request(input)
try:
return object["config"], object["version"]
except:
raise ProtocolException(f"Response not formatted correctly")