🥢 qo

qo [cue-oh] noun.

1. A minimalist TUI for querying JSON, CSV using SQL.
2. "query" what you need, and get it "out" to the pipeline.

Why qo?

• Muscle Memory: Use the SQL syntax you've known for years.
• Pipeline Native: Reads from stdin, writes to stdout.
• Interactive: Don't guess the query. See the result, then hit Enter.

Install

macOS and Linux (amd64/arm64) are supported.

Homebrew (Package)

brew install kiki-ki/tap/qo

Shell (Binary)

curl -sfL https://raw.githubusercontent.com/kiki-ki/go-qo/main/install.sh | sh

Git (Source)

# The binary will be located at ./bin/qo
git clone --depth 1 https://github.com/kiki-ki/go-qo.git && cd go-qo && make build

Usage

qo reads from both file arguments and standard input (stdin).

# Interactive mode (Open TUI)
cat x.json | qo
qo x.json y.json

# Non-interactive mode (Direct output)
cat x.json | qo -q "SELECT * FROM tmp WHERE id > 100"
qo -q "SELECT * FROM x JOIN y ON x.id = y.x_id" x.json y.json"

Pipe-Friendly TUI

TUI mode works seamlessly with pipes. Explore data interactively, then pass the result to other tools.

# Fetch JSON API > Filter interactively with qo > Format with jq
curl -s https://api.github.com/repos/kiki-ki/go-qo/commits | qo | jq '.[].sha'

# Explore > Filter > Compress
cat large.json | qo | gzip > filtered.json.gz

Query Logs & Aggregate

Use SQL to analyze structured data.

# Filter error logs
cat app.log | qo -q "SELECT timestamp, message FROM tmp WHERE level = 'error'"

# Aggregate sales by region
qo -i csv sales.csv -o csv -q "SELECT region, SUM(amount) FROM sales GROUP BY region"

Convert Formats

Transform between various formats.

qo -o csv data.json -q "SELECT id, name FROM data"             # JSON → CSV
qo -i csv -o json users.csv -q "SELECT * FROM users"           # CSV → JSON
qo -o jsonl data.json -q "SELECT * FROM data"                  # JSON → JSON Lines
qo -i csv --no-header raw.csv -q "SELECT col1, col2 FROM raw"  # Headerless CSV

Options

Flag	Short	Default	Description
--input	-i	json	Input format: json, csv, tsv, psv ("json" includes "jsonl")
--output	-o	json	Output format: json, jsonl, csv, tsv, psv, table
--query	-q		Run SQL query directly (Skip TUI)
--no-header			Treat first row as data, not header (CSV/TSV/PSV only)

UI Controls

Key	Mode	Action
Tab	ALL	Switch between Query/Table mode
Esc	ALL	Output & Quit
Ctrl+C	ALL	Quit (Output nothing)
Enter	QUERY	Execute query
↑ ↓ / j k	TABLE	Scroll rows
← → / h l	TABLE	Scroll columns

SQL Dialect

qo uses SQLite as its SQL engine. All queries follow SQLite syntax and support its built-in functions.

Querying Nested JSON

Use SQLite's json_extract() function to access nested fields in JSON data.

# Sample data: [{"user": {"name": "Alice", "age": 30}}, {"user": {"name": "Bob", "age": 25}}]

# Extract nested fields
qo data.json -q "SELECT json_extract(user, '$.name') AS name FROM data"
# Filter by nested value
qo data.json -q "SELECT * FROM data WHERE json_extract(user, '$.age') > 25"

For more details, see SQLite JSON Functions.

Built With

Category	Library
TUI Framework	Bubble Tea
Styling	Lip Gloss
CLI	Cobra
SQL Engine	modernc.org/sqlite (Pure Go, CGO-free)

License

MIT