Deephaven MCP

"""Markdown formatters for PyArrow tables.""" import pyarrow as pa def format_markdown_table(arrow_table: pa.Table) -> str: """ Format Arrow table as markdown table. Args: arrow_table (pa.Table): PyArrow Table to format Returns: str: Markdown table string with header and separator rows. Example: | id | name | age | | --- | --- | --- | | 1 | Alice | 30 | | 2 | Bob | 25 | """ # Get column names columns = arrow_table.column_names # Build header row header = "| " + " | ".join(columns) + " |" # Build separator row separator = "| " + " | ".join(["---"] * len(columns)) + " |" # Build data rows (use iterator to avoid loading entire table into memory) rows = [] for batch in arrow_table.to_batches(): for row_dict in batch.to_pylist(): # Escape pipe characters in cell values cells = [str(row_dict[col]).replace("|", "\\|") for col in columns] row_str = "| " + " | ".join(cells) + " |" rows.append(row_str) # Combine all parts parts = [header, separator] + rows return "\n".join(parts) def format_markdown_kv(arrow_table: pa.Table) -> str: """ Format Arrow table as markdown key-value pairs. Highest accuracy format for LLM consumption (60.7% per research). Args: arrow_table (pa.Table): PyArrow Table to format Returns: str: Markdown with record headers and key-value pairs. Example: ## Record 1 id: 1 name: Alice age: 30 ## Record 2 id: 2 name: Bob age: 25 """ columns = arrow_table.column_names records = [] idx = 0 # Use iterator to avoid loading entire table into memory for batch in arrow_table.to_batches(): for row_dict in batch.to_pylist(): idx += 1 # Record header record_lines = [f"## Record {idx}"] # Key-value pairs for col in columns: value = row_dict[col] # Escape colons in values if needed value_str = str(value).replace(":", "\\:") record_lines.append(f"{col}: {value_str}") records.append("\n".join(record_lines)) # Join records with blank line separator return "\n\n".join(records)