qalita 2.6.2__py3-none-any.whl → 2.8.0__py3-none-any.whl

qalita/internal/data_preview.py

@@ -0,0 +1,565 @@
+"""
+# QALITA (c) COPYRIGHT 2025 - ALL RIGHTS RESERVED -
+Data Preview module for Studio integration.
+
+This module provides data preview functionality for various source types,
+used by the gRPC worker to respond to DataPreviewRequest from the platform.
+"""
+
+import base64
+import json
+import mimetypes
+import os
+from dataclasses import dataclass, field
+from typing import Optional
+
+import pandas as pd
+
+from qalita.internal.utils import logger
+
+
+@dataclass
+class DataPreviewResult:
+    """Result of a data preview operation."""
+    
+    ok: bool
+    data_type: str  # table, image, pdf, text, json, error
+    error: Optional[str] = None
+    
+    # For table data
+    headers: list[str] = field(default_factory=list)
+    rows: list[list[str]] = field(default_factory=list)
+    total_rows: Optional[int] = None
+    
+    # For text/json content
+    content: Optional[str] = None
+    
+    # For binary content (image, pdf)
+    binary_base64: Optional[str] = None
+    mime_type: Optional[str] = None
+
+
+# Maximum size for binary content (5MB base64 encoded)
+MAX_BINARY_SIZE = 5 * 1024 * 1024
+# Maximum content size for text/json (1MB)
+MAX_TEXT_SIZE = 1 * 1024 * 1024
+# Default row limit
+DEFAULT_ROW_LIMIT = 1000
+
+
+def _error_result(message: str) -> DataPreviewResult:
+    """Create an error result."""
+    return DataPreviewResult(ok=False, data_type="error", error=message)
+
+
+def _dataframe_to_preview(
+    df: pd.DataFrame,
+    limit: int = DEFAULT_ROW_LIMIT,
+    total_rows: Optional[int] = None,
+) -> DataPreviewResult:
+    """Convert a pandas DataFrame to a preview result."""
+    try:
+        # Get headers
+        headers = [str(col) for col in df.columns.tolist()]
+        
+        # Limit rows
+        df_limited = df.head(limit)
+        
+        # Convert all values to strings for transport
+        rows = []
+        for _, row in df_limited.iterrows():
+            row_values = []
+            for val in row:
+                if pd.isna(val):
+                    row_values.append("")
+                else:
+                    row_values.append(str(val))
+            rows.append(row_values)
+        
+        return DataPreviewResult(
+            ok=True,
+            data_type="table",
+            headers=headers,
+            rows=rows,
+            total_rows=total_rows if total_rows is not None else len(df),
+        )
+    except Exception as e:
+        logger.error(f"Error converting DataFrame to preview: {e}")
+        return _error_result(f"Failed to convert data: {str(e)}")
+
+
+def preview_csv(
+    file_path: str,
+    limit: int = DEFAULT_ROW_LIMIT,
+    encoding: str = "utf-8",
+) -> DataPreviewResult:
+    """Preview a CSV file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        # Count total rows (without loading all data)
+        total_rows = sum(1 for _ in open(file_path, encoding=encoding, errors="ignore")) - 1
+        total_rows = max(0, total_rows)
+        
+        # Read only the needed rows
+        df = pd.read_csv(
+            file_path,
+            nrows=limit,
+            low_memory=False,
+            encoding=encoding,
+            on_bad_lines="warn",
+        )
+        
+        return _dataframe_to_preview(df, limit, total_rows)
+    except Exception as e:
+        logger.error(f"Error previewing CSV file {file_path}: {e}")
+        return _error_result(f"Failed to read CSV: {str(e)}")
+
+
+def preview_excel(
+    file_path: str,
+    limit: int = DEFAULT_ROW_LIMIT,
+    sheet_name: Optional[str] = None,
+) -> DataPreviewResult:
+    """Preview an Excel file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        # Read the Excel file
+        df = pd.read_excel(
+            file_path,
+            sheet_name=sheet_name or 0,
+            nrows=limit,
+            engine="openpyxl",
+        )
+        
+        # Get total rows - need to read without limit for count
+        df_full = pd.read_excel(
+            file_path,
+            sheet_name=sheet_name or 0,
+            engine="openpyxl",
+            usecols=[0],  # Only read first column for counting
+        )
+        total_rows = len(df_full)
+        
+        return _dataframe_to_preview(df, limit, total_rows)
+    except Exception as e:
+        logger.error(f"Error previewing Excel file {file_path}: {e}")
+        return _error_result(f"Failed to read Excel: {str(e)}")
+
+
+def preview_parquet(
+    file_path: str,
+    limit: int = DEFAULT_ROW_LIMIT,
+) -> DataPreviewResult:
+    """Preview a Parquet file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        # Read parquet with row limit
+        df = pd.read_parquet(file_path)
+        total_rows = len(df)
+        
+        return _dataframe_to_preview(df, limit, total_rows)
+    except Exception as e:
+        logger.error(f"Error previewing Parquet file {file_path}: {e}")
+        return _error_result(f"Failed to read Parquet: {str(e)}")
+
+
+def preview_json(file_path: str) -> DataPreviewResult:
+    """Preview a JSON file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        file_size = os.path.getsize(file_path)
+        if file_size > MAX_TEXT_SIZE:
+            # Read first part and indicate truncation
+            with open(file_path, "r", encoding="utf-8") as f:
+                content = f.read(MAX_TEXT_SIZE)
+            content += "\n\n... [truncated - file too large] ..."
+        else:
+            with open(file_path, "r", encoding="utf-8") as f:
+                content = f.read()
+        
+        # Validate JSON and pretty print
+        try:
+            parsed = json.loads(content.split("... [truncated")[0] if "... [truncated" in content else content)
+            content = json.dumps(parsed, indent=2, ensure_ascii=False)
+        except json.JSONDecodeError:
+            pass  # Keep raw content if not valid JSON
+        
+        return DataPreviewResult(
+            ok=True,
+            data_type="json",
+            content=content,
+        )
+    except Exception as e:
+        logger.error(f"Error previewing JSON file {file_path}: {e}")
+        return _error_result(f"Failed to read JSON: {str(e)}")
+
+
+def preview_text(file_path: str) -> DataPreviewResult:
+    """Preview a text file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        file_size = os.path.getsize(file_path)
+        if file_size > MAX_TEXT_SIZE:
+            with open(file_path, "r", encoding="utf-8", errors="replace") as f:
+                content = f.read(MAX_TEXT_SIZE)
+            content += "\n\n... [truncated - file too large] ..."
+        else:
+            with open(file_path, "r", encoding="utf-8", errors="replace") as f:
+                content = f.read()
+        
+        return DataPreviewResult(
+            ok=True,
+            data_type="text",
+            content=content,
+        )
+    except Exception as e:
+        logger.error(f"Error previewing text file {file_path}: {e}")
+        return _error_result(f"Failed to read text file: {str(e)}")
+
+
+def preview_image(file_path: str) -> DataPreviewResult:
+    """Preview an image file (PNG, JPG, GIF, WebP)."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        file_size = os.path.getsize(file_path)
+        if file_size > MAX_BINARY_SIZE:
+            return _error_result(f"Image too large for preview ({file_size} bytes)")
+        
+        # Detect MIME type
+        mime_type, _ = mimetypes.guess_type(file_path)
+        if not mime_type:
+            ext = os.path.splitext(file_path)[1].lower()
+            mime_map = {
+                ".png": "image/png",
+                ".jpg": "image/jpeg",
+                ".jpeg": "image/jpeg",
+                ".gif": "image/gif",
+                ".webp": "image/webp",
+            }
+            mime_type = mime_map.get(ext, "application/octet-stream")
+        
+        with open(file_path, "rb") as f:
+            binary_data = f.read()
+        
+        binary_base64 = base64.b64encode(binary_data).decode("utf-8")
+        
+        return DataPreviewResult(
+            ok=True,
+            data_type="image",
+            binary_base64=binary_base64,
+            mime_type=mime_type,
+        )
+    except Exception as e:
+        logger.error(f"Error previewing image file {file_path}: {e}")
+        return _error_result(f"Failed to read image: {str(e)}")
+
+
+def preview_pdf(file_path: str) -> DataPreviewResult:
+    """Preview a PDF file."""
+    try:
+        if not os.path.exists(file_path):
+            return _error_result(f"File not found: {file_path}")
+        
+        file_size = os.path.getsize(file_path)
+        if file_size > MAX_BINARY_SIZE:
+            return _error_result(f"PDF too large for preview ({file_size} bytes)")
+        
+        with open(file_path, "rb") as f:
+            binary_data = f.read()
+        
+        binary_base64 = base64.b64encode(binary_data).decode("utf-8")
+        
+        return DataPreviewResult(
+            ok=True,
+            data_type="pdf",
+            binary_base64=binary_base64,
+            mime_type="application/pdf",
+        )
+    except Exception as e:
+        logger.error(f"Error previewing PDF file {file_path}: {e}")
+        return _error_result(f"Failed to read PDF: {str(e)}")
+
+
+def preview_database(
+    config: dict,
+    limit: int = DEFAULT_ROW_LIMIT,
+    query: Optional[str] = None,
+    table: Optional[str] = None,
+) -> DataPreviewResult:
+    """Preview data from a database source."""
+    try:
+        from sqlalchemy import create_engine, text
+        
+        # Build connection string from config
+        db_type = config.get("type", "").lower()
+        
+        connection_string = config.get("connection_string")
+        if not connection_string:
+            # Build connection string from components
+            db_type_map = {
+                "postgresql": "postgresql",
+                "mysql": "mysql",
+                "mssql": "mssql+pymssql",
+                "oracle": "oracle+oracledb",
+                "sqlite": "sqlite",
+            }
+            
+            dialect = db_type_map.get(db_type)
+            if not dialect:
+                return _error_result(f"Unsupported database type: {db_type}")
+            
+            if db_type == "sqlite":
+                database_path = config.get("database", ":memory:")
+                connection_string = f"sqlite:///{database_path}"
+            elif db_type == "oracle":
+                connection_string = (
+                    f"{dialect}://{config['username']}:{config['password']}"
+                    f"@{config['host']}:{config['port']}/?service_name={config['database']}"
+                )
+            else:
+                connection_string = (
+                    f"{dialect}://{config['username']}:{config['password']}"
+                    f"@{config['host']}:{config['port']}/{config['database']}"
+                )
+        
+        engine = create_engine(connection_string)
+        
+        # Determine what to query
+        if query:
+            # Custom query provided
+            sql = query
+        elif table:
+            # Specific table
+            schema = config.get("schema")
+            qualified_table = f"{schema}.{table}" if schema else table
+            sql = f"SELECT * FROM {qualified_table}"
+        else:
+            return _error_result("No table or query specified for database preview")
+        
+        # Add LIMIT clause if not present
+        sql_lower = sql.strip().lower()
+        if "limit" not in sql_lower:
+            sql = f"{sql.rstrip(';')} LIMIT {limit}"
+        
+        # Execute query
+        with engine.connect() as conn:
+            # Get total count (approximate)
+            try:
+                if table and not query:
+                    schema = config.get("schema")
+                    qualified_table = f"{schema}.{table}" if schema else table
+                    count_result = conn.execute(text(f"SELECT COUNT(*) FROM {qualified_table}"))
+                    total_rows = count_result.scalar()
+                else:
+                    total_rows = None
+            except Exception:
+                total_rows = None
+            
+            # Get data
+            df = pd.read_sql(sql, conn)
+        
+        return _dataframe_to_preview(df, limit, total_rows)
+    except Exception as e:
+        logger.error(f"Error previewing database: {e}")
+        return _error_result(f"Failed to query database: {str(e)}")
+
+
+def preview_s3(
+    config: dict,
+    limit: int = DEFAULT_ROW_LIMIT,
+) -> DataPreviewResult:
+    """Preview data from S3 source."""
+    try:
+        # Build path
+        path = config.get("path")
+        if not path:
+            bucket = config.get("bucket")
+            key = config.get("key")
+            if bucket and key:
+                path = f"s3://{bucket}/{key}"
+        
+        if not path:
+            return _error_result("S3 path not configured")
+        
+        # Build storage options
+        storage_options = {}
+        for opt_key in ["key", "secret", "token", "client_kwargs"]:
+            if opt_key in config:
+                storage_options[opt_key] = config[opt_key]
+        
+        # Determine file type from path
+        ext = os.path.splitext(path)[1].lower()
+        
+        if ext == ".csv":
+            df = pd.read_csv(
+                path,
+                storage_options=storage_options or None,
+                nrows=limit,
+            )
+            return _dataframe_to_preview(df, limit)
+        elif ext == ".parquet":
+            df = pd.read_parquet(path, storage_options=storage_options or None)
+            total_rows = len(df)
+            return _dataframe_to_preview(df.head(limit), limit, total_rows)
+        elif ext in (".xlsx", ".xls"):
+            df = pd.read_excel(path, nrows=limit, engine="openpyxl")
+            return _dataframe_to_preview(df, limit)
+        elif ext == ".json":
+            df = pd.read_json(path, storage_options=storage_options or None)
+            return _dataframe_to_preview(df.head(limit), limit, len(df))
+        else:
+            return _error_result(f"Unsupported file type in S3: {ext}")
+    except Exception as e:
+        logger.error(f"Error previewing S3 source: {e}")
+        return _error_result(f"Failed to read S3: {str(e)}")
+
+
+def preview_source(
+    source_config: dict,
+    limit: int = DEFAULT_ROW_LIMIT,
+    query: Optional[str] = None,
+) -> DataPreviewResult:
+    """
+    Preview data from a source configuration.
+    
+    This is the main entry point for data preview.
+    
+    Args:
+        source_config: Source configuration dict with 'type' and 'config' keys
+        limit: Maximum number of rows to return
+        query: Optional SQL query for database sources
+    
+    Returns:
+        DataPreviewResult with the preview data
+    """
+    source_type = source_config.get("type", "").lower()
+    config = source_config.get("config", {})
+    
+    logger.info(f"Previewing source type: {source_type}")
+    
+    # File-based sources
+    if source_type in ("file", "csv"):
+        path = config.get("path")
+        if not path:
+            return _error_result("File path not configured")
+        return preview_csv(path, limit)
+    
+    elif source_type == "excel":
+        path = config.get("path")
+        if not path:
+            return _error_result("Excel path not configured")
+        return preview_excel(path, limit, config.get("sheet_name"))
+    
+    elif source_type == "parquet":
+        path = config.get("path")
+        if not path:
+            return _error_result("Parquet path not configured")
+        return preview_parquet(path, limit)
+    
+    elif source_type == "json":
+        path = config.get("path")
+        if not path:
+            return _error_result("JSON path not configured")
+        return preview_json(path)
+    
+    elif source_type == "text":
+        path = config.get("path")
+        if not path:
+            return _error_result("Text file path not configured")
+        return preview_text(path)
+    
+    elif source_type == "image":
+        path = config.get("path")
+        if not path:
+            return _error_result("Image path not configured")
+        return preview_image(path)
+    
+    elif source_type == "pdf":
+        path = config.get("path")
+        if not path:
+            return _error_result("PDF path not configured")
+        return preview_pdf(path)
+    
+    # Database sources
+    elif source_type in ("postgresql", "mysql", "mssql", "oracle", "sqlite"):
+        db_config = {**config, "type": source_type}
+        table = config.get("table") or config.get("default_table")
+        return preview_database(db_config, limit, query, table)
+    
+    # Cloud storage sources
+    elif source_type == "s3":
+        return preview_s3(config, limit)
+    
+    elif source_type in ("gcs", "azure_blob"):
+        return _error_result(f"Preview for {source_type} not yet implemented")
+    
+    # Folder - try to preview first file
+    elif source_type == "folder":
+        path = config.get("path")
+        if not path or not os.path.isdir(path):
+            return _error_result("Folder path not configured or not accessible")
+        
+        # Find first data file
+        for ext in ("*.csv", "*.xlsx", "*.parquet", "*.json"):
+            import glob
+            files = glob.glob(os.path.join(path, ext))
+            if files:
+                # Preview first file found
+                file_ext = os.path.splitext(files[0])[1].lower()
+                if file_ext == ".csv":
+                    return preview_csv(files[0], limit)
+                elif file_ext == ".xlsx":
+                    return preview_excel(files[0], limit)
+                elif file_ext == ".parquet":
+                    return preview_parquet(files[0], limit)
+                elif file_ext == ".json":
+                    return preview_json(files[0])
+        
+        return _error_result("No supported data files found in folder")
+    
+    else:
+        return _error_result(f"Unsupported source type: {source_type}")
+
+
+def detect_preview_type(file_path: str) -> str:
+    """
+    Detect the type of preview based on file extension.
+    
+    Returns one of: csv, excel, parquet, json, text, image, pdf, unknown
+    """
+    ext = os.path.splitext(file_path)[1].lower()
+    
+    ext_map = {
+        ".csv": "csv",
+        ".xlsx": "excel",
+        ".xls": "excel",
+        ".parquet": "parquet",
+        ".pq": "parquet",
+        ".json": "json",
+        ".txt": "text",
+        ".log": "text",
+        ".md": "text",
+        ".yml": "text",
+        ".yaml": "text",
+        ".xml": "text",
+        ".png": "image",
+        ".jpg": "image",
+        ".jpeg": "image",
+        ".gif": "image",
+        ".webp": "image",
+        ".pdf": "pdf",
+    }
+    
+    return ext_map.get(ext, "unknown")

qalita/internal/request.py

@@ -171,7 +171,8 @@ def send_request(
             f"Retrying {current_retry+1}/{total_retry} in {grace_period} seconds..."
         )
         time.sleep(grace_period)
-        r = send_request(
+        # Use __wrapped__ to bypass the @pass_config decorator and pass config directly
+        r = send_request.__wrapped__(
             config,
             request,
             mode,

qalita/internal/utils.py

@@ -224,7 +224,7 @@ def safe_path_check(path: str) -> str:  # lgtm[py/path-injection]
 
 
 def get_version():
-    return "2.6.2"
+    return "2.8.0"
 
 
 def make_tarfile(output_filename, source_dir):

qalita/web/app.py

@@ -25,7 +25,13 @@ def create_app(config_obj) -> Flask:
         import sys
         if sys.version_info < (3, 13):
             # Only try gevent on Python < 3.13 due to zope.interface compatibility issues
+            # But first verify gevent is actually installed and working
+            import gevent  # noqa: F401
+            from gevent import monkey  # noqa: F401
             async_mode = "gevent"
+    except ImportError:
+        # gevent not installed, use threading
+        pass
     except Exception:
         pass
 
@@ -112,8 +118,6 @@ def get_socketio():
 def run_dashboard_ui(config_obj, host: str = "localhost", port: int = 7070):
     app = create_app(config_obj)
     socketio = get_socketio()
-    url = f"http://{host}:{port}"
-    print(f"QALITA CLI UI is running. Open {url}")
 
     # Use gevent for production WebSocket support
     if socketio: