parqv 0.1.0__py3-none-any.whl → 0.2.1__py3-none-any.whl

parqv/views/utils/stats_formatters.py

@@ -0,0 +1,160 @@
+"""
+Statistics formatting utilities for parqv views.
+"""
+
+from typing import Any, Dict, List, Union
+
+from rich.text import Text
+
+
+def format_stats_for_display(stats_data: Dict[str, Any]) -> List[Union[str, Text]]:
+    """
+    Format statistics dictionary for display as lines of rich text.
+    
+    Args:
+        stats_data: Raw statistics dictionary from handler
+        
+    Returns:
+        List of formatted lines ready for display
+    """
+    if not stats_data:
+        return [Text.from_markup("[red]No statistics data available.[/red]")]
+
+    lines: List[Union[str, Text]] = []
+    
+    # Extract basic column information
+    col_name = stats_data.get("column", "N/A")
+    col_type = stats_data.get("type", "Unknown")
+    nullable_val = stats_data.get("nullable")
+
+    # Format column header
+    lines.extend(_format_column_header(col_name, col_type, nullable_val))
+    
+    # Handle calculation errors
+    calc_error = stats_data.get("error")
+    if calc_error:
+        lines.extend(_format_error_section(calc_error))
+    
+    # Add informational messages
+    message = stats_data.get("message")
+    if message:
+        lines.extend(_format_message_section(message))
+    
+    # Format calculated statistics
+    calculated = stats_data.get("calculated")
+    if calculated:
+        lines.extend(_format_calculated_stats(calculated, has_error=bool(calc_error)))
+    
+    return lines
+
+
+def format_column_info(column_name: str, column_type: str, nullable: Any) -> List[Union[str, Text]]:
+    """
+    Format basic column information for display.
+    
+    Args:
+        column_name: Name of the column
+        column_type: Type of the column
+        nullable: Nullability information
+        
+    Returns:
+        List of formatted lines for column info
+    """
+    return _format_column_header(column_name, column_type, nullable)
+
+
+def _format_column_header(col_name: str, col_type: str, nullable_val: Any) -> List[Union[str, Text]]:
+    """Format the column header section."""
+    # Determine nullability display
+    if nullable_val is True:
+        nullable_str = "Nullable"
+    elif nullable_val is False:
+        nullable_str = "Required"
+    else:
+        nullable_str = "Unknown Nullability"
+    
+    lines = [
+        Text.assemble(("Column: ", "bold"), f"`{col_name}`"),
+        Text.assemble(("Type:   ", "bold"), f"{col_type} ({nullable_str})"),
+        "─" * (len(col_name) + len(col_type) + 20)
+    ]
+    
+    return lines
+
+
+def _format_error_section(calc_error: str) -> List[Union[str, Text]]:
+    """Format the error section."""
+    return [
+        Text("Calculation Error:", style="bold red"),
+        f"```\n{calc_error}\n```",
+        ""
+    ]
+
+
+def _format_message_section(message: str) -> List[Union[str, Text]]:
+    """Format the informational message section."""
+    return [
+        Text(f"Info: {message}", style="italic cyan"),
+        ""
+    ]
+
+
+def _format_calculated_stats(calculated: Dict[str, Any], has_error: bool = False) -> List[Union[str, Text]]:
+    """Format the calculated statistics section."""
+    lines = [Text("Calculated Statistics:", style="bold")]
+    
+    # Define the order of statistics to display
+    stats_order = [
+        "Total Count", "Valid Count", "Null Count", "Null Percentage",
+        "Distinct Count", "Distinct Values (Approx)",
+        "Min", "Max", "Mean", "Median (50%)", "StdDev", "Variance",
+        "True Count", "False Count",
+        "Value Counts"
+    ]
+    
+    found_stats = False
+    
+    for key in stats_order:
+        if key in calculated:
+            found_stats = True
+            value = calculated[key]
+            lines.extend(_format_single_stat(key, value))
+    
+    # Add any additional stats not in the predefined order
+    for key, value in calculated.items():
+        if key not in stats_order:
+            found_stats = True
+            lines.extend(_format_single_stat(key, value))
+    
+    # Handle case where no stats were found
+    if not found_stats and not has_error:
+        lines.append(Text("  (No specific stats calculated for this type)", style="dim"))
+    
+    return lines
+
+
+def _format_single_stat(key: str, value: Any) -> List[Union[str, Text]]:
+    """Format a single statistic entry."""
+    lines = []
+    
+    if key == "Value Counts" and isinstance(value, dict):
+        lines.append(f"  - {key}:")
+        for sub_key, sub_val in value.items():
+            sub_val_str = _format_stat_value(sub_val)
+            lines.append(f"    - {sub_key}: {sub_val_str}")
+    else:
+        formatted_value = _format_stat_value(value)
+        lines.append(f"  - {key}: {formatted_value}")
+    
+    return lines
+
+
+def _format_stat_value(value: Any) -> str:
+    """Format a single statistic value."""
+    if isinstance(value, (int, float)):
+        if isinstance(value, int):
+            return f"{value:,}"
+        else:
+            return f"{value:,.4f}"
+    else:
+        return str(value) 

parqv-0.2.1.dist-info/METADATA

@@ -0,0 +1,104 @@
+Metadata-Version: 2.4
+Name: parqv
+Version: 0.2.1
+Summary: An interactive Python TUI for visualizing, exploring, and analyzing files directly in your terminal.
+Author-email: Sangmin Yoon <sanspareilsmyn@gmail.com>
+License-Expression: Apache-2.0
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: textual>=1.0.0
+Requires-Dist: pyarrow>=16.0.0
+Requires-Dist: pandas>=2.0.0
+Requires-Dist: numpy>=1.20.0
+Requires-Dist: duckdb>=1.2.0
+Dynamic: license-file
+
+# parqv
+
+[![Python Version](https://img.shields.io/badge/Python-3.10+-blue.svg)](https://www.python.org/)
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE)
+[![PyPI version](https://badge.fury.io/py/parqv.svg)](https://badge.fury.io/py/parqv) <!-- TODO: Link after first PyPI release -->
+[![Built with Textual](https://img.shields.io/badge/Built%20with-Textual-blueviolet.svg)](https://textual.textualize.io/)
+
+---
+
+**Supported File Formats:** ✅ **Parquet** | ✅ **JSON** / **JSON Lines (ndjson)** | *(More planned!)*
+
+---
+
+**`parqv` is a Python-based interactive TUI (Text User Interface) tool designed to explore, analyze, and understand various data file formats directly within your terminal.** Initially supporting Parquet and JSON, `parqv` aims to provide a unified, visual experience for quick data inspection without leaving your console.
+
+## 💻 Demo
+
+![parqv.gif](assets/parqv.gif)
+*(Demo shows Parquet features; UI adapts for other formats)*
+
+## 🤔 Why `parqv`?
+1.  **Unified Interface:** Launch `parqv <your_data_file>` to access **metadata, schema, data preview, and column statistics** all within a single, navigable terminal window. No more juggling different commands for different file types.
+2.  **Interactive Exploration:**
+    *   **🖱️ Keyboard & Mouse Driven:** Navigate using familiar keys (arrows, `hjkl`, Tab) or even your mouse (thanks to `Textual`).
+    *   **📜 Scrollable Views:** Easily scroll through large schemas, data tables, or column lists.
+    *   **🌲 Clear Schema View:** Understand column names, data types, and nullability at a glance. (Complex nested structures visualization might vary by format).
+    *   **📊 Dynamic Stats:** Select a column and instantly see its detailed statistics (counts, nulls, min/max, mean, distinct values, etc.).
+3.  **Cross-Format Consistency:**
+    *   **🎨 Rich Display:** Leverages `rich` and `Textual` for colorful, readable tables and text across supported formats.
+    *   **📈 Quick Stats:** Get key statistical insights consistently, regardless of the underlying file type.
+    *   **🔌 Extensible:** Designed with a handler interface to easily add support for more file formats in the future (like CSV, Arrow IPC, etc.).
+
+## ✨ Features (TUI Mode)
+*   **Multi-Format Support:** Currently supports **Parquet** (`.parquet`) and **JSON/JSON Lines** (`.json`, `.ndjson`). Run `parqv <your_file.{parquet,json,ndjson}>`.
+*   **Metadata Panel:** Displays key file information (path, format, size, total rows, column count, etc.). *Fields may vary slightly depending on the file format.*
+*   **Schema Explorer:**
+    *   Interactive list view of columns.
+    *   Clearly shows column names, data types, and nullability.
+*   **Data Table Viewer:**
+    *   Scrollable table preview of the file's data.
+    *   Attempts to preserve data types for better representation.
+*   **Column Statistics Viewer:**
+    *   Select a column in the Schema tab to view detailed statistics.
+    *   Shows counts (total, valid, null), percentages, and type-specific stats (min/max, mean, stddev, distinct counts, length stats, boolean value counts where applicable).
+*   **Row Group Inspector (Parquet Specific):**
+    *   *This panel only appears when viewing Parquet files.*
+    *   Lists row groups with stats (row count, compressed/uncompressed size).
+    *   (Planned) Select a row group for more details.
+
+## 🚀 Getting Started
+
+**1. Prerequisites:**
+*   **Python:** Version 3.10 or higher.
+*   **pip:** The Python package installer.
+
+**2. Install `parqv`:**
+*   Open your terminal and run:
+    ```bash
+    pip install parqv
+    ```
+    *(This will also install dependencies like `textual`, `pyarrow`, `pandas`, and `duckdb`)*
+*   **Updating `parqv`:**
+    ```bash
+    pip install --upgrade parqv
+    ```
+
+**3. Run `parqv`:**
+*   Point `parqv` to your data file:
+    ```bash
+    #parquet
+    parqv /path/to/your/data.parquet
+    
+    # json
+    parqv /path/to/your/data.json
+*   The interactive TUI will launch. Use your keyboard (and mouse, if supported by your terminal) to navigate:
+    *   **Arrow Keys / `j`,`k` (in lists):** Move selection up/down.
+    *   **`Tab` / `Shift+Tab`:** Cycle focus between the main tab content and potentially other areas. (Focus handling might evolve).
+    *   **`Enter` (in column list):** Select a column to view statistics.
+    *   **View Switching:** Use `Ctrl+N` (Next Tab) and `Ctrl+P` (Previous Tab) or click on the tabs (Metadata, Schema, Data Preview).
+    *   **Scrolling:** Use `PageUp` / `PageDown` / `Home` / `End` or arrow keys/mouse wheel within scrollable areas (like Schema stats or Data Preview).
+    *   **`q` / `Ctrl+C`:** Quit `parqv`.
+    *   *(Help Screen `?` is planned)*
+
+---
+
+## 📄 License
+
+Licensed under the Apache License, Version 2.0. See [LICENSE](LICENSE) for the full license text.

parqv-0.2.1.dist-info/RECORD

@@ -0,0 +1,34 @@
+parqv/__init__.py,sha256=CcARikIb8knQqd3bGu6Y9exgSbzdywjdORz15VKKxmU,611
+parqv/app.py,sha256=KXfL-RfMOOngzxn0uEeWa5UQgHRISBBl1IHK0ffQlzI,5556
+parqv/cli.py,sha256=9KPOYywA53vUVp_5RI2lFgKtZay2EHosOsTo-sz4rOU,3242
+parqv/parqv.css,sha256=C42ZXUwMX1ZXfGo0AmixbHxz0CWKzWBHZ_hkhq5aehg,2920
+parqv/core/__init__.py,sha256=C8P-wqP72hk54qcjRNjDQ9X4hTu9eRttAyVynCAeDQw,791
+parqv/core/config.py,sha256=4dAyTkqOnqoin8OdB7QGl7JYlWx0EGV_q8JgY0TnMuI,542
+parqv/core/file_utils.py,sha256=m7d8wD9nxXCmfUU1b6IGBqjW0mBL2Nr40r9M8NI67FY,2374
+parqv/core/handler_factory.py,sha256=xUj5vNOwLkEReL0x8CJbEbsqJ6XXm44w4NEx--TT8d4,3035
+parqv/core/logging.py,sha256=fom0zPykBgn7zpW_vlvaCDlCERI1lccdjtBRMES94-A,1173
+parqv/data_sources/__init__.py,sha256=7l7rA1Q4-gqPFnYRCpGnC4aZq7layQhXTAIitnUmQfM,873
+parqv/data_sources/base/__init__.py,sha256=Alo4tGZJdUsfvGvDgpRodur3xn-SO7hgeODmhx6tA6U,524
+parqv/data_sources/base/exceptions.py,sha256=QjTQqPw_LgEYchN2OjWCOkUuRV-Sn9I1J3gmoiS4ekA,802
+parqv/data_sources/base/handler.py,sha256=I0Hmsf-WYhWyLGUHEV902iPS7Gg4dDkRPgyXOfGZSTQ,4429
+parqv/data_sources/formats/__init__.py,sha256=XRaoyDCh03UIC7YwHJJiPcd5XZw_Yh6w3obD5CQ1Eog,308
+parqv/data_sources/formats/json.py,sha256=bqspvndHTHIQevvXckfdeEnDCSjtl7eyv-PXJfXmaFg,19998
+parqv/data_sources/formats/parquet.py,sha256=L8JA3pxjinNTZY3lMPKs9CwVO59fdwnkeSui5tQxrew,31580
+parqv/views/__init__.py,sha256=aQAsH9akPMBVf7F0KNiELAtb-yoThQsR41nXk3G_mbQ,943
+parqv/views/base.py,sha256=9kRsUjTXdxEoiy9gSlNmU1ksZsN7FkyXWepX4C0vBFQ,2948
+parqv/views/data_view.py,sha256=Xk7XdXCN82AmH3hyrmjDAIGtc212zc1lrqE_2ijWKeY,4344
+parqv/views/metadata_view.py,sha256=6qdLEyO0te1TMiVtTxq-L3PWTOxCtFkkrDH_4yaXPaw,2058
+parqv/views/schema_view.py,sha256=r1ZHJZ7g20KT_EfGrh71AU0-gNj1Z4Z-WUDQCwOu3BY,8966
+parqv/views/components/__init__.py,sha256=rBsX9UH67GCLn275LhG2Xyf84gRtrm-otuj6zW5JAiU,267
+parqv/views/components/enhanced_data_table.py,sha256=57pJD9rT-kiy2llw4p5pZOReUtv5CZxN-2tD6e7QDaI,4952
+parqv/views/components/error_display.py,sha256=Ak1AbT9dikkF3izVtVFxdRB0nas6jqjb5tPG3l-_JjQ,2444
+parqv/views/components/loading_display.py,sha256=1lLxvt5oMBGDiojODTVQ6O_hVNFHCWqpWamk863BMkY,1440
+parqv/views/utils/__init__.py,sha256=3PVlUMR-317CJmgvvT1iFk-mYZEETZz3H88g2e-7HAo,343
+parqv/views/utils/data_formatters.py,sha256=3Kp77Ue3z-8M3BGetwxY7H3CocUeBoWaXr8tTXus_lg,4386
+parqv/views/utils/stats_formatters.py,sha256=vbsVI-mi7ITyS0CGPIBx1k_IrGgZYvmMhbcN1JHkm-8,4890
+parqv-0.2.1.dist-info/licenses/LICENSE,sha256=Ewl2wCa8r6ncxHlpf-ZZXb77c82zdfxHuEeKzBbm6nM,11324
+parqv-0.2.1.dist-info/METADATA,sha256=q2ngRgL0aUrRrgwb45HqRuYAHJtgS--5PNlJVPCS3JA,5369
+parqv-0.2.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+parqv-0.2.1.dist-info/entry_points.txt,sha256=8Tm8rTiIB-tbVItoOA4M7seEmFnrtK25BMH9UKzqfXg,44
+parqv-0.2.1.dist-info/top_level.txt,sha256=_t3_49ZluJbvl0QU_P3GNVuXxCffqiTp37dzZIa2GEw,6
+parqv-0.2.1.dist-info/RECORD,,

{parqv-0.1.0.dist-info → parqv-0.2.1.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (78.1.0)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any