CSV Editor

Instructions

Implementation Reference

• src/csv_editor/tools/io_operations.py:116-185 (handler)

  Core handler function that downloads CSV from URL using pandas.read_csv, validates URL, loads into session manager, and returns operation result with data preview.

  async def load_csv_from_url(
      url: str,
      encoding: str = "utf-8",
      delimiter: str = ",",
      session_id: Optional[str] = None,
      ctx: Context = None
  ) -> Dict[str, Any]:
      """Load a CSV file from a URL.
      
      Args:
          url: URL of the CSV file
          encoding: File encoding
          delimiter: Column delimiter
          session_id: Optional existing session ID
          ctx: FastMCP context
      
      Returns:
          Operation result with session ID and data info
      """
      try:
          # Validate URL
          is_valid, validated_url = validate_url(url)
          if not is_valid:
              return OperationResult(
                  success=False,
                  message=f"Invalid URL: {validated_url}",
                  error=validated_url
              ).model_dump()
          
          if ctx:
              await ctx.info(f"Loading CSV from URL: {url}")
              await ctx.report_progress(0.1, "Downloading file...")
          
          # Download CSV using pandas (it handles URLs directly)
          df = pd.read_csv(url, encoding=encoding, delimiter=delimiter)
          
          if ctx:
              await ctx.report_progress(0.8, "Processing data...")
          
          # Get or create session
          session_manager = get_session_manager()
          session = session_manager.get_or_create_session(session_id)
          session.load_data(df, url)
          
          if ctx:
              await ctx.report_progress(1.0, "Complete!")
              await ctx.info(f"Loaded {len(df)} rows and {len(df.columns)} columns")
          
          return OperationResult(
              success=True,
              message=f"Successfully loaded CSV from URL",
              session_id=session.session_id,
              rows_affected=len(df),
              columns_affected=df.columns.tolist(),
              data={
                  "shape": df.shape,
                  "source_url": url,
                  "preview": df.head(5).to_dict('records')
              }
          ).model_dump()
          
      except Exception as e:
          if ctx:
              await ctx.error(f"Failed to load CSV from URL: {str(e)}")
          return OperationResult(
              success=False,
              message="Failed to load CSV from URL",
              error=str(e)
          ).model_dump()

• src/csv_editor/server.py:121-131 (registration)

  MCP tool registration with @mcp.tool decorator, which delegates to the imported handler _load_csv_from_url from io_operations.

  @mcp.tool
  async def load_csv_from_url(
      url: str,
      encoding: str = "utf-8",
      delimiter: str = ",",
      session_id: Optional[str] = None,
      ctx: Context = None
  ) -> Dict[str, Any]:
      """Load a CSV file from a URL."""
      return await _load_csv_from_url(url, encoding, delimiter, session_id, ctx)

• src/csv_editor/server.py:98-106 (registration)

  Import of the actual handler function aliased as _load_csv_from_url for use in the registered tool stub.

  from .tools.io_operations import (
      load_csv as _load_csv,
      load_csv_from_url as _load_csv_from_url,
      load_csv_from_content as _load_csv_from_content,
      export_csv as _export_csv,
      get_session_info as _get_session_info,
      list_sessions as _list_sessions,
      close_session as _close_session
  )

• src/csv_editor/tools/data_operations.py:6-6 (helper)

  Re-export of load_csv_from_url from io_operations for compatibility in data_operations module.

  load_csv_from_url,