Explore tabular data in your terminal.
Parquet, CSV, JSON, Excel, SQLite, logs and more, on disk, in S3, GCS or Azure, over HTTP(S), from a pipe or from Python. Query with SQL, then chart, reshape and export.
datui s3://noaa-ghcn-pds/parquet/by_year/YEAR=2024/ # 38M rows, no login
Linux, macOS and Windows · One binary, built on Polars · MIT license · No telemetry
Get started
Install
macOS
Homebrew
brew tap derekwisong/datui && brew trust derekwisong/datui && brew install datui
Install script
curl -fsSL https://raw.githubusercontent.com/derekwisong/datui/main/scripts/install/install.sh | sh
Linux
Install script
curl -fsSL https://raw.githubusercontent.com/derekwisong/datui/main/scripts/install/install.sh | sh
Debian, Ubuntu (apt)
curl -fsSL https://derekwisong.github.io/datui-apt/public.key | sudo gpg --dearmor -o /usr/share/keyrings/datui-archive-keyring.gpg
echo "deb [signed-by=/usr/share/keyrings/datui-archive-keyring.gpg] https://derekwisong.github.io/datui-apt/ ./" | sudo tee /etc/apt/sources.list.d/datui.list
sudo apt update
sudo apt install datui
Fedora, RHEL (dnf)
sudo curl -fsSL -o /etc/yum.repos.d/datui.repo https://derekwisong.github.io/datui-apt/rpm/datui.repo
sudo dnf install datui
Arch Linux (AUR)
yay -S datui-bin # or: paru -S datui-bin
Homebrew
brew tap derekwisong/datui && brew trust derekwisong/datui && brew install datui
Windows
WinGet
winget install derekwisong.datui
Python
Python (pip)
pip install datui
Rust
Rust (cargo)
cargo install datui --locked
Binaries
Binaries
Linux (x86_64, arm64), macOS (Intel, Apple silicon) and Windows binaries, plus .deb, .rpm and Arch tarballs, are on the latest release.
More ways to install, and how to uninstall: Installation guide · Latest release
What it does
File formats
Parquet, CSV, JSON, Arrow, Excel, SQLite and more. For a format datui does not know, describe its layout in a TOML format spec; any other file opens in the hex view.
Formats → · Format specs →
Local or in the cloud
Open files, directories and globs on disk, or data in S3, GCS, Azure and over HTTP(S). A Hive-partitioned directory opens as one table. The home screen lists your catalogs, the cloud accounts you are logged in to, and Example datasets that need no login.
Home screen → · Cloud storage →
Query and find
: runs a query in SQL or q. / finds text, a regex or letters in order. [ and ] sort by the column under the cursor, + and - keep or drop the rows with its value, and v saves the steps as a view to use on the next file.
Query data →
Chart
c charts the column under the cursor as a line, scatter, bar, histogram, box, KDE or heatmap, over the rows your query and filters keep. e exports the chart to PNG, SVG or PDF.
Charts →
Reshape and retype
p pivots or melts the table, with a live preview. Change a column's type from the Schema tab in Info, or right-click a cell.
Reshaping → · Column types →Also: catalogs of your own datasets, each with a Documentation view (Ctrl+E); themes (datui theme list); mouse support; and copying over SSH through OSC 52. Every key →
night-market and day-market built in, gruvbox-dark and high-contrast from contrib/themes/. Themes →Formats
Each of these opens from disk, HTTP(S) or a cloud bucket. The formats table shows which are read lazily and which open compressed.
- Columnar and JSON
- Parquet · JSON · NDJSON · Arrow IPC · Avro · ORC · Excel
- Delimited text
- CSV · TSV · PSV · plain text
- Model files
- SafeTensors · GGUF
- Signals and logs
- NMEA · GPX · WAV/AIFF audio · MIDI · VCD · FIX · SDF · ELF · ULog · DataFlash · candump · systemd journal
- Databases and arrays
- SQLite · NumPy
- Format specs
- Your own binary formats, described in TOML
How it stays fast
- Written in Rust on Polars. Every table stays a
LazyFrameuntil it is drawn, and only the rows on screen, plus a buffer ahead, are collected. - Sorts, filters and queries add steps to the lazy plan; they do not copy the data.
- Parquet in S3, GCS or Azure is read in place: the footers for the schema and row count, then only the row groups the screen needs.
- Rows are counted in the background, so you can scroll before the count is done.
First rows in milliseconds on local files, under a second from S3. Performance has the numbers.
Try it
In your terminal
datui # the home screen: your files, and Example datasets that need no login
datui https://vincentarelbundock.github.io/Rdatasets/csv/palmerpenguins/penguins.csv
printf 'id,amount\n1,9.50\n2,3.25\n' | datui
Press ? for the keys on any screen; Ctrl+Q quits. Quick start · Keys
From Python
import datui
import polars as pl
url = "https://vincentarelbundock.github.io/Rdatasets/csv/palmerpenguins/penguins.csv"
datui.view(pl.scan_csv(url))
pip install datui installs the command and the module. Use datui from Python
What it is not
- An editor. Cells cannot be changed, and datui never writes to a file it opened unless you export over it. Results leave as an export or a copy.
- A service. No telemetry and no update check: datui connects only to the data and cloud accounts you open or browse.
Documentation
Read the latest release's docs · Keys · Query syntax · Command line
Older releases
- v0.4.10
- v0.4.9
- v0.4.8
- v0.4.7
- v0.4.6
- v0.4.5
- v0.4.4
- v0.4.3
- v0.4.2
- v0.4.1
- v0.4.0
- v0.3.2
- v0.3.1
- v0.3.0
- v0.2.56
- v0.2.55
- v0.2.54
- v0.2.53
- v0.2.52
- v0.2.51
- v0.2.50
- v0.2.49
- v0.2.48
- v0.2.47
- v0.2.46
- v0.2.45
- v0.2.44
- v0.2.43
- v0.2.42
- v0.2.41
- v0.2.40
- v0.2.39
- v0.2.38
- v0.2.37
- v0.2.34
- v0.2.33
- v0.2.32
- v0.2.31
- v0.2.30
- v0.2.29
- v0.2.28
- v0.2.27
- v0.2.26
- v0.2.25
- v0.2.24
- v0.2.23
- v0.2.22
- v0.2.21
- v0.2.20
- v0.2.19
- v0.2.18
- v0.2.17
- v0.2.16
- v0.2.15
- v0.2.14
- v0.2.13
- v0.2.12
- v0.2.11
- v0.2.10
- v0.2.9
- v0.2.8
- v0.2.7
- v0.2.6
- v0.2.5
- v0.2.4
- v0.2.3
- v0.2.2
- v0.2.1