Local, read-only desktop explorer for Parquet, CSV, TSV, JSON, JSON Lines, Excel and SQLite
  • Rust 66.5%
  • TypeScript 17.4%
  • JavaScript 10.7%
  • CSS 4.7%
  • Makefile 0.4%
  • Other 0.3%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-10-24 14:47:09 +02:00
.github/workflows Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
docs Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
experiments/library-checks fix: Fix build problems 2026-10-22 01:42:08 +02:00
scripts chore: Make version read from standard location 2026-10-22 13:30:33 +02:00
src Add HDF5 browsing, slicing and export (#3) 2026-10-24 13:12:11 +02:00
src-tauri Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
tests Add HDF5 browsing, slicing and export (#3) 2026-10-24 13:12:11 +02:00
.gitignore feat: Add update function, and fix some bugs 2026-10-22 13:09:16 +02:00
CHANGELOG.md Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
icon.svg Sivro 0.1.0: local, read-only desktop data explorer 2026-10-22 00:45:55 +02:00
index.html feat: Add update function, and fix some bugs 2026-10-22 13:09:16 +02:00
LICENSE Sivro 0.1.0: local, read-only desktop data explorer 2026-10-22 00:45:55 +02:00
Makefile Add HDF5 browsing, slicing and export (#3) 2026-10-24 13:12:11 +02:00
package-lock.json Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
package.json Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
README.md Prepare v0.1.2 release (#4) 2026-10-24 14:47:09 +02:00
tsconfig.json Sivro 0.1.0: local, read-only desktop data explorer 2026-10-22 00:45:55 +02:00
vite.config.ts Sivro 0.1.0: local, read-only desktop data explorer 2026-10-22 00:45:55 +02:00

Sivro

CI Release License: MIT

A local, read-only desktop explorer for tabular data, built with Rust, Tauri 2 and DuckDB.

  • Opens Parquet, Arrow IPC/Feather V2, HDF5, CSV, TSV, JSON, JSON Lines, Excel (.xlsx, .xlsm, .xlsb, .xls) and SQLite.
  • Exports the current view to Parquet, Arrow IPC files/streams, HDF5, CSV, TSV, a JSON array or JSON Lines.
  • Safe by design: it never modifies a source file, runs no macros or formulas and accepts no SQL. Exploring data works entirely offline.
  • Exact values: large integers and decimals are never rounded through floating point.
  • Appearance: Catppuccin Latte and Mocha themes, a saved Light/Dark/System choice, visible menu hover and keyboard highlights, and clear pending-view indicators on tabs.
  • Manual updates: About → Check for updates contacts GitHub only when requested. Review the release notes and confirm before downloading and installing a signed update. There are no automatic checks or updates.

Sivro is a pre-1.0 beta; see the changelog for known limitations.

Install

Linux and macOS:

curl -fsSL https://raw.githubusercontent.com/MohamedElashri/sivro/main/scripts/install.sh | sh

The script picks the right package (.deb, .rpm or AppImage on Linux, the universal .dmg on macOS), verifies it against SHA256SUMS.txt and installs it. Set SIVRO_VERSION=v0.1.2 to pin a release.

Manual download: get the package for your platform from the releases page.

Platform Package
Windows x86_64 .msi or -setup.exe
macOS (Intel and Apple Silicon) universal .dmg
Linux x86_64 .deb, .rpm or .AppImage

Note

macOS: the app is not signed with an Apple Developer ID or notarized, so macOS blocks a downloaded copy ("Sivro is damaged" or "cannot be opened"). After moving it to Applications, allow it with:

xattr -dr com.apple.quarantine /Applications/Sivro.app

or open it once and choose Open Anyway under System Settings → Privacy & Security. The install script does this for you.

Windows: the installer is not code-signed either, so SmartScreen may warn; choose More info → Run anyway.

Usage

  1. Open a file with Open (Ctrl/⌘+O) or drag it onto the window. Each file opens in its own tab.
  2. Confirm the parsing preview for CSV/TSV, pick a sheet for Excel workbooks, pick a table for SQLite databases, or browse HDF5 groups and choose a dataset and slice.
  3. Search (Ctrl/⌘+K), filter, sort, hide columns, profile a column or inspect a row. Count rows gives exact totals for filtered views.
  4. Export writes the current view (columns, filters and sort) to a new file.

JSON files that aren't tables open as a navigable document tree. Formats and behaviour covers supported readers and limits.

Development

Requires Node.js 24, Rust 1.90+ and the Tauri prerequisites for your platform.

make dev      # run the app with hot reload
make ci       # formatting, clippy and all tests, as CI runs them
make build    # release packages
make clean    # remove build outputs
make help     # list every target

CI lints with the latest stable Rust, so run rustup update stable first. On Windows the Makefile builds DuckDB from source automatically. See development for architecture, tests and releasing.

License

MIT.