Query Command

Interactive shell and CLI for querying data from CSV, TSV, NDJSON files with format conversion.

Usage

# Interactive shell mode
grpctestify query

# CLI mode with query
grpctestify query -q "users status=active" data/users.csv

# With stdin input
cat data.csv | grpctestify query -q "stdin status=active" -

# Output to file (format auto-detected from extension)
grpctestify query -q "users status=active" data/users.csv --output active_users.ndjson

Query Syntax

<source> [col<op>value]...

Where:

• <source> — file name without extension, or stdin
• col<op>value — filter expression

Filter Operators

Operator	Description	Example
=	Equals	status=active
!=	Not equals	status!=pending
>=	Greater or equal	age>=18
<=	Less or equal	age<=65
>	Greater	score>100
<	Less	score<50
~glob	Glob pattern	name~glob"*John*"
~re:	Regex	msg~re:"error|warn"
= with commas	In list (matches any value)	status=active,pending

Examples

# Simple equality
grpctestify query -q "users status=active" users.csv

# Multiple filters (AND)
grpctestify query -q "users status=active age>=18" users.csv

# Glob pattern
grpctestify query -q 'users name~glob"*John*"' users.csv

# Regex
grpctestify query -q 'logs msg~re:"error|warn|debug"' logs.ndjson

# IN list
grpctestify query -q "users status=active,pending,suspended" users.csv

CLI Options

Option	Short	Description	Default
--query	-q	Query expression	-
--shell	-s	Force interactive shell	false
--indexed-by	-i	Index column	-
--format	-f	Output format (json, csv, table, line, tsv)	table
--limit	-n	Max rows to return	100
--offset	-o	Skip N rows	0
--columns	-c	Output columns (comma-separated)	all
--order-by		Sort column (prefix - for DESC)	-
--output		Output file (format from extension)	stdout
--no-header		Skip header row in output	false

Output Formats

# Table (default)
grpctestify query -q "users id=1" users.csv
# ┌─────┬───────┬──────────┐
# │ id  │ name  │ status   │
# ├─────┼───────┼──────────┤
# │ 1   │ alice │ active   │
# └─────┴───────┴──────────┘

# JSON
grpctestify query -q "users id=1" users.csv -f json
# {"id":"1","name":"alice","status":"active"}

# CSV
grpctestify query -q "users id=1" users.csv -f csv
# id,name,status
# 1,alice,active

# Line (key=value pairs, space-separated, one row per line)
grpctestify query -q "users id=1" users.csv -f line
# id=1 name=alice status=active

Interactive Shell

grpctestify query users.csv another.tsv

Shell Commands

Command	Description
.help	Show help
.quit	Exit shell
.tables	List loaded sources
.schema <name>	Show source columns
.indexes <name>	Show index info
.info	Show all sources summary
.count <name>	Count rows
.sample <name> [n]	Show sample rows (default: 5)
.load <file.gctf>	Load sources from .gctf file
.add <name> <file>	Add file as source
.remove <name>	Remove source
.mode <format>	Set output format
.headers <on|off>	Toggle headers
.output <file>	Set output file

Shell Example

$ grpctestify query users.csv
grpctestify query shell
Type '.help' for available commands

query> .tables
  users

query> .schema users
Schema for 'users':
  id
  name
  status
  age

query> .info
Sources: 1
  users: 4 columns

query> status=active
┌─────┬───────┬──────────┐
│ id  │ name  │ status   │
├─────┼───────┼──────────┤
│ 1   │ alice │ active   │
│ 3   │ charlie │ active │
└─────┴───────┴──────────┘

query> .quit

Supported Formats

Format	Extensions	Auto-detect
CSV	.csv	By content (comma-separated, no tabs)
TSV	.tsv	By content (tab-separated)
NDJSON	.ndjson, .jsonl	By content (lines starting with {)
JSON	.json	By extension only

Format Detection

When reading from stdin (-), format is auto-detected:

• Lines contain tabs but no commas → TSV
• Lines start with { → NDJSON
• Lines contain commas → CSV

File Loading

Files are loaded by extension:

• .gctf — parsed as gRPC test file, sources extracted
• .csv, .tsv, .ndjson, .json — data file, source name = file stem

Directories can be specified — all .gctf files are loaded.

# Load directory of gctf files
grpctestify query ./tests/data/

# Multiple files
grpctestify query users.csv orders.csv products.ndjson