Gridka

A fast, native macOS CSV viewer powered by DuckDB. Opens files of any size with instant scrolling, sorting, filtering, and search.

Why Gridka?

Most CSV tools either choke on large files (Excel, Numbers), consume absurd memory (Electron apps), or lack a GUI entirely (csvkit, xsv). Gridka uses DuckDB's analytical engine behind AppKit's NSTableView to handle 100M+ row files while staying under 500 MB RAM.

Features

• Open any CSV/TSV — auto-detects delimiter, encoding, headers, and column types
• Virtual scrolling — 60 fps at any file size; only visible rows are in memory
• Column sorting — click to sort, Shift+click for multi-column sort
• Filtering — type-aware filters per column (text, numeric, date, boolean operators)
• Global search — Cmd+F to search across all columns with match highlighting
• Column management — resize, reorder, hide/show, rename, delete, auto-fit
• Cell inspection — detail pane for long text, URLs, JSON content
• Cell editing — double-click to edit, save back to CSV
• Multi-tab — open multiple files in tabs within a single window
• Quick Look — preview CSV files in Finder via Quick Look extension

Analysis

• Column sparklines — mini distribution charts in every column header; click to open profiler
• Column profiler — sidebar with stats (min, max, mean, median, stddev), distribution histogram, and top values
• Value frequency — floating panel with histogram and sortable value-count table; click a value to filter
• Computed columns — add formula columns using DuckDB SQL expressions with live preview
• Group By — visual aggregation builder (COUNT, SUM, AVG, MIN, MAX, STDDEV); results open as a new tab

Installation

Mac App Store

Coming soon.

Download

Grab the latest .zip from Releases, unzip, and drag Gridka.app to /Applications.

Building from Source

# 1. Clone
git clone https://github.com/Ceesaxp/gridka.git
cd gridka

# 2. Ensure DuckDB libraries are in Libraries/
#    Download the macOS release from https://github.com/duckdb/duckdb/releases
#    Place duckdb.h, libduckdb.a, and libduckdb.dylib in Libraries/

# 3. Generate Xcode project (requires XcodeGen)
xcodegen generate

# 4. Build
xcodebuild -scheme Gridka -configuration Release build

# Or open in Xcode:
open Gridka.xcodeproj

Requirements: macOS 14.0 (Sonoma)+, Xcode 15.0+, Swift 5.10+

Keyboard Shortcuts

Shortcut	Action
Cmd+O	Open file
Cmd+S	Save
Shift+Cmd+S	Save As
Cmd+T	New tab
Cmd+W	Close tab
Cmd+F	Toggle search bar
Cmd+G / Shift+Cmd+G	Find next / previous
Cmd+C	Copy cell value
Shift+Cmd+C	Copy entire row
Option+Cmd+C	Copy column (visible rows)
Shift+Cmd+D	Toggle detail pane
Shift+Cmd+P	Toggle column profiler
Option+Cmd+T	Show analysis toolbar
Option+Cmd+F	Add computed column
Option+Cmd+G	Group By
Option+Cmd+N	Add column
Option+Cmd+R	Add row
Cmd+Delete	Delete selected row(s)
Click header	Sort ascending/descending/clear
Shift+click header	Multi-column sort
Right-click header	Filter / hide / rename / frequency
Right-click cell	Copy / filter by value / exclude value
Double-click divider	Auto-fit column width
Double-click cell	Inline edit
Drag & drop	Open CSV by dragging onto window

Architecture

Single-process Swift app. DuckDB embedded via C API (static library). No IPC, no Electron, no web views.

User interaction → ViewState mutation → SQL generation → DuckDB query → NSTableView reload

Every sort, filter, and search is a SQL query. DuckDB does the heavy lifting. See CLAUDE.md for full architectural documentation.

Project Structure

Gridka/
├── Sources/
│   ├── App/              # AppDelegate, MainMenu
│   ├── Engine/           # DuckDB wrapper, query coordinator, row cache
│   ├── Model/            # FileSession, ViewState, ColumnFilter
│   ├── UI/               # NSTableView, filter bar, search, status bar, help
│   └── Bridging/         # C API bridging header, Swift type wrappers
├── QuickLookExtension/   # Quick Look preview for CSV files
├── Libraries/            # Vendored DuckDB (duckdb.h + libduckdb)
├── Resources/            # Assets, Info.plist, PrivacyInfo
├── Tests/                # Unit, integration, and regression tests
├── UITests/              # XCUITest UI automation
├── scripts/              # Build and release automation
└── docs/                 # App Store submission guide

Contributing

1. Fork the repo
2. Create a feature branch (git checkout -b feature/my-feature)
3. Commit your changes
4. Push and open a Pull Request

License

MIT