robosystems-client 0.2.16__tar.gz → 0.2.17__tar.gz

{robosystems_client-0.2.16 → robosystems_client-0.2.17}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: robosystems-client
-Version: 0.2.16
+Version: 0.2.17
 Summary: Python Client for RoboSystems financial graph database API
 Author: RFS LLC
 License: MIT

{robosystems_client-0.2.16 → robosystems_client-0.2.17}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "robosystems-client"
-version = "0.2.16"
+version = "0.2.17"
 description = "Python Client for RoboSystems financial graph database API"
 readme = "README.md"
 requires-python = ">=3.10"

{robosystems_client-0.2.16 → robosystems_client-0.2.17}/robosystems_client/extensions/__init__.py

@@ -28,12 +28,22 @@ from .operation_client import (
   OperationProgress,
   OperationResult,
 )
-from .table_ingest_client import (
-  TableIngestClient,
-  UploadOptions,
-  IngestOptions,
-  UploadResult,
+from .file_client import (
+  FileClient,
+  FileUploadOptions,
+  FileUploadResult,
+  FileInfo,
+)
+from .materialization_client import (
+  MaterializationClient,
+  MaterializationOptions,
+  MaterializationResult,
+  MaterializationStatus,
+)
+from .table_client import (
+  TableClient,
   TableInfo,
+  QueryResult as TableQueryResult,
 )
 from .graph_client import (
   GraphClient,
@@ -177,12 +187,20 @@ __all__ = [
   "OperationStatus",
   "OperationProgress",
   "OperationResult",
-  # Table Ingest Client
-  "TableIngestClient",
-  "UploadOptions",
-  "IngestOptions",
-  "UploadResult",
+  # File Client
+  "FileClient",
+  "FileUploadOptions",
+  "FileUploadResult",
+  "FileInfo",
+  # Materialization Client
+  "MaterializationClient",
+  "MaterializationOptions",
+  "MaterializationResult",
+  "MaterializationStatus",
+  # Table Client
+  "TableClient",
   "TableInfo",
+  "TableQueryResult",
   # Graph Client
   "GraphClient",
   "GraphMetadata",

{robosystems_client-0.2.16 → robosystems_client-0.2.17}/robosystems_client/extensions/extensions.py

@@ -9,7 +9,9 @@ from typing import Dict, Any, Optional, Callable
 from .query_client import QueryClient
 from .agent_client import AgentClient
 from .operation_client import OperationClient
-from .table_ingest_client import TableIngestClient
+from .file_client import FileClient
+from .materialization_client import MaterializationClient
+from .table_client import TableClient
 from .graph_client import GraphClient
 from .sse_client import SSEClient
 
@@ -61,7 +63,9 @@ class RoboSystemsExtensions:
     self.query = QueryClient(self.config)
     self.agent = AgentClient(self.config)
     self.operations = OperationClient(self.config)
-    self.tables = TableIngestClient(self.config)
+    self.files = FileClient(self.config)
+    self.materialization = MaterializationClient(self.config)
+    self.tables = TableClient(self.config)
     self.graphs = GraphClient(self.config)
 
   def monitor_operation(
@@ -92,7 +96,12 @@ class RoboSystemsExtensions:
     self.query.close()
     self.agent.close()
     self.operations.close_all()
-    self.tables.close()
+    if hasattr(self.files, "close"):
+      self.files.close()
+    if hasattr(self.materialization, "close"):
+      self.materialization.close()
+    if hasattr(self.tables, "close"):
+      self.tables.close()
     self.graphs.close()
 
   # Convenience methods that delegate to the appropriate clients

robosystems_client-0.2.17/robosystems_client/extensions/file_client.py

@@ -0,0 +1,380 @@
+"""File Client for RoboSystems API
+
+Manages file operations as first-class resources with multi-layer status tracking.
+Files are independent entities with their own lifecycle (S3 → DuckDB → Graph).
+"""
+
+from dataclasses import dataclass
+from io import BytesIO
+from pathlib import Path
+from typing import Dict, Any, Optional, Callable, Union, BinaryIO
+import logging
+import httpx
+
+from ..api.files.create_file_upload import (
+  sync_detailed as create_file_upload,
+)
+from ..api.files.update_file import (
+  sync_detailed as update_file,
+)
+from ..api.files.list_files import (
+  sync_detailed as list_files,
+)
+from ..api.files.get_file import (
+  sync_detailed as get_file,
+)
+from ..api.files.delete_file import (
+  sync_detailed as delete_file,
+)
+from ..models.file_upload_request import FileUploadRequest
+from ..models.file_status_update import FileStatusUpdate
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class FileUploadOptions:
+  """Options for file upload operations"""
+
+  on_progress: Optional[Callable[[str], None]] = None
+  fix_localstack_url: bool = True
+  ingest_to_graph: bool = False
+
+
+@dataclass
+class FileUploadResult:
+  """Result from file upload operation"""
+
+  file_id: str
+  file_size: int
+  row_count: int
+  table_name: str
+  file_name: str
+  success: bool = True
+  error: Optional[str] = None
+
+
+@dataclass
+class FileInfo:
+  """Information about a file"""
+
+  file_id: str
+  file_name: str
+  file_format: str
+  size_bytes: int
+  row_count: Optional[int]
+  upload_status: str
+  table_name: str
+  created_at: Optional[str]
+  uploaded_at: Optional[str]
+  layers: Optional[Dict[str, Any]] = None
+
+
+class FileClient:
+  """Client for managing files as first-class resources"""
+
+  def __init__(self, config: Dict[str, Any]):
+    self.config = config
+    self.base_url = config["base_url"]
+    self.headers = config.get("headers", {})
+    self.token = config.get("token")
+    self._http_client = httpx.Client(timeout=120.0)
+
+  def upload(
+    self,
+    graph_id: str,
+    table_name: str,
+    file_or_buffer: Union[Path, str, BytesIO, BinaryIO],
+    options: Optional[FileUploadOptions] = None,
+  ) -> FileUploadResult:
+    """
+    Upload a file to a table.
+
+    This handles the complete 3-step upload process:
+    1. Get presigned upload URL
+    2. Upload file to S3
+    3. Mark file as 'uploaded' (triggers DuckDB staging)
+
+    Args:
+        graph_id: Graph database identifier
+        table_name: Table to associate file with
+        file_or_buffer: File path, Path object, BytesIO, or file-like object
+        options: Upload options (progress callback, LocalStack URL fix, auto-ingest)
+
+    Returns:
+        FileUploadResult with file metadata and status
+    """
+    options = options or FileUploadOptions()
+
+    try:
+      # Determine file name and read content
+      if isinstance(file_or_buffer, (str, Path)):
+        file_path = Path(file_or_buffer)
+        file_name = file_path.name
+        with open(file_path, "rb") as f:
+          file_content = f.read()
+      elif isinstance(file_or_buffer, BytesIO):
+        file_name = "data.parquet"
+        file_content = file_or_buffer.getvalue()
+      elif hasattr(file_or_buffer, "read"):
+        file_name = getattr(file_or_buffer, "name", "data.parquet")
+        file_content = file_or_buffer.read()
+      else:
+        raise ValueError(f"Unsupported file type: {type(file_or_buffer)}")
+
+      # Step 1: Get presigned upload URL
+      if options.on_progress:
+        options.on_progress(
+          f"Getting upload URL for {file_name} → table '{table_name}'..."
+        )
+
+      upload_request = FileUploadRequest(
+        file_name=file_name,
+        content_type="application/x-parquet",
+        table_name=table_name,
+      )
+
+      kwargs = {
+        "graph_id": graph_id,
+        "client": self.config.get("client"),
+        "body": upload_request,
+      }
+
+      response = create_file_upload(**kwargs)
+
+      if response.status_code != 200 or not response.parsed:
+        error_msg = f"Failed to get upload URL: {response.status_code}"
+        return FileUploadResult(
+          file_id="",
+          file_size=0,
+          row_count=0,
+          table_name=table_name,
+          file_name=file_name,
+          success=False,
+          error=error_msg,
+        )
+
+      upload_data = response.parsed
+      upload_url = upload_data.upload_url
+      file_id = upload_data.file_id
+
+      # Fix LocalStack URL if needed
+      if options.fix_localstack_url and "localstack:4566" in upload_url:
+        upload_url = upload_url.replace("localstack:4566", "localhost:4566")
+
+      # Step 2: Upload file to S3
+      if options.on_progress:
+        options.on_progress(f"Uploading {file_name} to S3...")
+
+      s3_response = self._http_client.put(
+        upload_url,
+        content=file_content,
+        headers={"Content-Type": "application/x-parquet"},
+      )
+
+      if s3_response.status_code not in [200, 204]:
+        return FileUploadResult(
+          file_id=file_id,
+          file_size=len(file_content),
+          row_count=0,
+          table_name=table_name,
+          file_name=file_name,
+          success=False,
+          error=f"S3 upload failed: {s3_response.status_code}",
+        )
+
+      # Step 3: Mark file as uploaded
+      if options.on_progress:
+        options.on_progress(f"Marking {file_name} as uploaded...")
+
+      status_update = FileStatusUpdate(
+        status="uploaded",
+        ingest_to_graph=options.ingest_to_graph,
+      )
+
+      update_kwargs = {
+        "graph_id": graph_id,
+        "file_id": file_id,
+        "client": self.config.get("client"),
+        "body": status_update,
+      }
+
+      update_response = update_file(**update_kwargs)
+
+      if update_response.status_code != 200 or not update_response.parsed:
+        return FileUploadResult(
+          file_id=file_id,
+          file_size=len(file_content),
+          row_count=0,
+          table_name=table_name,
+          file_name=file_name,
+          success=False,
+          error="Failed to complete file upload",
+        )
+
+      # Extract metadata from response
+      response_data = update_response.parsed
+      actual_file_size = getattr(response_data, "file_size_bytes", len(file_content))
+      actual_row_count = getattr(response_data, "row_count", 0)
+
+      if options.on_progress:
+        options.on_progress(
+          f"✅ Uploaded {file_name} ({actual_file_size:,} bytes, {actual_row_count:,} rows)"
+        )
+
+      return FileUploadResult(
+        file_id=file_id,
+        file_size=actual_file_size,
+        row_count=actual_row_count,
+        table_name=table_name,
+        file_name=file_name,
+        success=True,
+      )
+
+    except Exception as e:
+      logger.error(f"File upload failed: {e}")
+      return FileUploadResult(
+        file_id="",
+        file_size=0,
+        row_count=0,
+        table_name=table_name,
+        file_name=getattr(file_or_buffer, "name", "unknown"),
+        success=False,
+        error=str(e),
+      )
+
+  def list(
+    self,
+    graph_id: str,
+    table_name: Optional[str] = None,
+    status: Optional[str] = None,
+  ) -> list[FileInfo]:
+    """
+    List files in a graph with optional filtering.
+
+    Args:
+        graph_id: Graph database identifier
+        table_name: Optional table name filter
+        status: Optional upload status filter (uploaded, pending, etc.)
+
+    Returns:
+        List of FileInfo objects
+    """
+    try:
+      kwargs = {
+        "graph_id": graph_id,
+        "client": self.config.get("client"),
+      }
+
+      if table_name:
+        kwargs["table_name"] = table_name
+      if status:
+        kwargs["status"] = status
+
+      response = list_files(**kwargs)
+
+      if response.status_code != 200 or not response.parsed:
+        logger.error(f"Failed to list files: {response.status_code}")
+        return []
+
+      files_data = response.parsed
+      files = getattr(files_data, "files", [])
+
+      return [
+        FileInfo(
+          file_id=f.file_id,
+          file_name=f.file_name,
+          file_format=f.file_format,
+          size_bytes=f.size_bytes or 0,
+          row_count=f.row_count,
+          upload_status=f.upload_status,
+          table_name=getattr(f, "table_name", ""),
+          created_at=f.created_at,
+          uploaded_at=f.uploaded_at,
+        )
+        for f in files
+      ]
+
+    except Exception as e:
+      logger.error(f"Failed to list files: {e}")
+      return []
+
+  def get(self, graph_id: str, file_id: str) -> Optional[FileInfo]:
+    """
+    Get detailed information about a specific file.
+
+    Args:
+        graph_id: Graph database identifier
+        file_id: File ID
+
+    Returns:
+        FileInfo with multi-layer status tracking, or None if not found
+    """
+    try:
+      kwargs = {
+        "graph_id": graph_id,
+        "file_id": file_id,
+        "client": self.config.get("client"),
+      }
+
+      response = get_file(**kwargs)
+
+      if response.status_code != 200 or not response.parsed:
+        logger.error(f"Failed to get file {file_id}: {response.status_code}")
+        return None
+
+      file_data = response.parsed
+
+      return FileInfo(
+        file_id=file_data.file_id,
+        file_name=file_data.file_name,
+        file_format=file_data.file_format,
+        size_bytes=file_data.size_bytes or 0,
+        row_count=file_data.row_count,
+        upload_status=file_data.upload_status,
+        table_name=file_data.table_name or "",
+        created_at=file_data.created_at,
+        uploaded_at=file_data.uploaded_at,
+        layers=getattr(file_data, "layers", None),
+      )
+
+    except Exception as e:
+      logger.error(f"Failed to get file {file_id}: {e}")
+      return None
+
+  def delete(self, graph_id: str, file_id: str, cascade: bool = False) -> bool:
+    """
+    Delete a file from all layers.
+
+    Args:
+        graph_id: Graph database identifier
+        file_id: File ID to delete
+        cascade: If True, delete from all layers including DuckDB and graph
+
+    Returns:
+        True if deletion succeeded, False otherwise
+    """
+    try:
+      kwargs = {
+        "graph_id": graph_id,
+        "file_id": file_id,
+        "client": self.config.get("client"),
+        "cascade": cascade,
+      }
+
+      response = delete_file(**kwargs)
+
+      if response.status_code not in [200, 204]:
+        logger.error(f"Failed to delete file {file_id}: {response.status_code}")
+        return False
+
+      return True
+
+    except Exception as e:
+      logger.error(f"Failed to delete file {file_id}: {e}")
+      return False
+
+  def __del__(self):
+    """Cleanup HTTP client on deletion"""
+    if hasattr(self, "_http_client"):
+      self._http_client.close()

robosystems_client-0.2.17/robosystems_client/extensions/materialization_client.py

@@ -0,0 +1,211 @@
+"""Materialization Client for RoboSystems API
+
+Manages graph materialization from DuckDB staging tables.
+Treats the graph database as a materialized view of the mutable DuckDB data lake.
+"""
+
+from dataclasses import dataclass
+from typing import Dict, Any, Optional, Callable
+import logging
+
+from ..api.materialization.materialize_graph import (
+  sync_detailed as materialize_graph,
+)
+from ..api.materialization.get_materialization_status import (
+  sync_detailed as get_materialization_status,
+)
+from ..models.materialize_request import MaterializeRequest
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class MaterializationOptions:
+  """Options for graph materialization operations"""
+
+  ignore_errors: bool = True
+  rebuild: bool = False
+  force: bool = False
+  on_progress: Optional[Callable[[str], None]] = None
+
+
+@dataclass
+class MaterializationResult:
+  """Result from materialization operation"""
+
+  status: str
+  was_stale: bool
+  stale_reason: Optional[str]
+  tables_materialized: list[str]
+  total_rows: int
+  execution_time_ms: float
+  message: str
+  success: bool = True
+  error: Optional[str] = None
+
+
+@dataclass
+class MaterializationStatus:
+  """Status information about graph materialization"""
+
+  graph_id: str
+  is_stale: bool
+  stale_reason: Optional[str]
+  stale_since: Optional[str]
+  last_materialized_at: Optional[str]
+  materialization_count: int
+  hours_since_materialization: Optional[float]
+  message: str
+
+
+class MaterializationClient:
+  """Client for managing graph materialization operations"""
+
+  def __init__(self, config: Dict[str, Any]):
+    self.config = config
+    self.base_url = config["base_url"]
+    self.headers = config.get("headers", {})
+    self.token = config.get("token")
+
+  def materialize(
+    self,
+    graph_id: str,
+    options: Optional[MaterializationOptions] = None,
+  ) -> MaterializationResult:
+    """
+    Materialize graph from DuckDB staging tables.
+
+    Rebuilds the complete graph database from the current state of DuckDB
+    staging tables. Automatically discovers all tables, materializes them in
+    the correct order (nodes before relationships), and clears the staleness flag.
+
+    Args:
+        graph_id: Graph database identifier
+        options: Materialization options (ignore_errors, rebuild, force)
+
+    Returns:
+        MaterializationResult with detailed execution information
+
+    When to use:
+        - After batch uploads (files uploaded with ingest_to_graph=false)
+        - After cascade file deletions (graph marked stale)
+        - Periodic full refresh to ensure consistency
+        - Recovery from partial materialization failures
+    """
+    options = options or MaterializationOptions()
+
+    try:
+      if options.on_progress:
+        options.on_progress("Starting graph materialization...")
+
+      request = MaterializeRequest(
+        ignore_errors=options.ignore_errors,
+        rebuild=options.rebuild,
+        force=options.force,
+      )
+
+      kwargs = {
+        "graph_id": graph_id,
+        "client": self.config.get("client"),
+        "body": request,
+      }
+
+      response = materialize_graph(**kwargs)
+
+      if response.status_code != 200 or not response.parsed:
+        error_msg = f"Materialization failed: {response.status_code}"
+        if hasattr(response, "content"):
+          try:
+            import json
+
+            error_data = json.loads(response.content)
+            error_msg = error_data.get("detail", error_msg)
+          except Exception:
+            pass
+
+        return MaterializationResult(
+          status="failed",
+          was_stale=False,
+          stale_reason=None,
+          tables_materialized=[],
+          total_rows=0,
+          execution_time_ms=0,
+          message=error_msg,
+          success=False,
+          error=error_msg,
+        )
+
+      result_data = response.parsed
+
+      if options.on_progress:
+        options.on_progress(
+          f"✅ Materialization complete: {len(result_data.tables_materialized)} tables, "
+          f"{result_data.total_rows:,} rows in {result_data.execution_time_ms:.2f}ms"
+        )
+
+      return MaterializationResult(
+        status=result_data.status,
+        was_stale=result_data.was_stale,
+        stale_reason=result_data.stale_reason,
+        tables_materialized=result_data.tables_materialized,
+        total_rows=result_data.total_rows,
+        execution_time_ms=result_data.execution_time_ms,
+        message=result_data.message,
+        success=True,
+      )
+
+    except Exception as e:
+      logger.error(f"Materialization failed: {e}")
+      return MaterializationResult(
+        status="failed",
+        was_stale=False,
+        stale_reason=None,
+        tables_materialized=[],
+        total_rows=0,
+        execution_time_ms=0,
+        message=str(e),
+        success=False,
+        error=str(e),
+      )
+
+  def status(self, graph_id: str) -> Optional[MaterializationStatus]:
+    """
+    Get current materialization status for the graph.
+
+    Shows whether the graph is stale (DuckDB has changes not yet in graph database),
+    when it was last materialized, and how long since last materialization.
+
+    Args:
+        graph_id: Graph database identifier
+
+    Returns:
+        MaterializationStatus with staleness and timing information
+    """
+    try:
+      kwargs = {
+        "graph_id": graph_id,
+        "client": self.config.get("client"),
+      }
+
+      response = get_materialization_status(**kwargs)
+
+      if response.status_code != 200 or not response.parsed:
+        logger.error(f"Failed to get materialization status: {response.status_code}")
+        return None
+
+      status_data = response.parsed
+
+      return MaterializationStatus(
+        graph_id=status_data.graph_id,
+        is_stale=status_data.is_stale,
+        stale_reason=status_data.stale_reason,
+        stale_since=status_data.stale_since,
+        last_materialized_at=status_data.last_materialized_at,
+        materialization_count=status_data.materialization_count,
+        hours_since_materialization=status_data.hours_since_materialization,
+        message=status_data.message,
+      )
+
+    except Exception as e:
+      logger.error(f"Failed to get materialization status: {e}")
+      return None