# dadbod-grip.nvim task runner

# Install just: https://github.com/casey/just

# Default recipe: run tests

default: test

# Run all unit tests (328 specs across 14 modules)

test:

nvim --headless -u tests/minimal_init.lua -l tests/run_specs.lua

# Run a single spec file by name (e.g., just spec data)

spec name:

nvim --headless -u tests/minimal_init.lua -l tests/spec/{{name}}_spec.lua

# Lint with luacheck (if installed)

lint:

luacheck lua/ --no-unused-args --no-max-line-length

# Seed PostgreSQL test database

seed-pg:

createdb grip_test 2>/dev/null || true

psql grip_test < tests/seed_pg.sql

# Seed SQLite test database

seed-sqlite:

sqlite3 tests/seed_sqlite.db < tests/seed_sqlite.sql

# Seed MySQL test database

seed-mysql:

mysql -u root -e "CREATE DATABASE IF NOT EXISTS grip_test"

mysql -u root grip_test < tests/seed_mysql.sql

# Seed DuckDB test database

seed-duckdb:

rm -f tests/seed_duckdb.duckdb

duckdb tests/seed_duckdb.duckdb < tests/seed_duckdb.sql

# Seed httpfs demo: DuckDB connection + saved queries for remote URLs

seed-httpfs:

mkdir -p .grip/queries

echo '[{"name":"DuckDB (memory)","url":"duckdb::memory:"}]' > .grip/connections.json

echo "SELECT species, island, avg(body_mass_g) as avg_mass FROM 'https://blobs.duckdb.org/data/penguins.csv' GROUP BY species, island ORDER BY avg_mass DESC" > .grip/queries/penguins-csv.sql

echo "SELECT Pclass, Sex, count(*) as n, sum(Survived) as survived, round(100.0*sum(Survived)/count(*),1) as pct FROM 'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv' GROUP BY Pclass, Sex ORDER BY Pclass, Sex" > .grip/queries/titanic-csv.sql

echo "SELECT symbol, min(price), max(price), round(max(price)-min(price),2) as range FROM 'https://vega.github.io/vega-datasets/data/stocks.csv' GROUP BY symbol ORDER BY range DESC" > .grip/queries/stocks-csv.sql

echo "SELECT * FROM 'https://duckdb.org/data/prices.parquet' LIMIT 100" > .grip/queries/prices-parquet.sql

echo "SELECT Origin, round(avg(Miles_per_Gallon),1) as mpg, round(avg(Horsepower),0) as hp, count(*) as n FROM 'https://vega.github.io/vega-datasets/data/cars.json' GROUP BY Origin" > .grip/queries/cars-json.sql

echo "SELECT Species, round(avg(SepalLengthCm),2) as sepal, round(avg(PetalLengthCm),2) as petal, count(*) as n FROM 'https://huggingface.co/api/datasets/scikit-learn/iris/parquet/default/train/0.parquet' GROUP BY Species" > .grip/queries/iris-parquet.sql

echo "SELECT userId, count(*) as total, sum(case when completed then 1 else 0 end) as done FROM 'https://duckdb.org/data/json/todos.json' GROUP BY userId ORDER BY done DESC" > .grip/queries/todos-json.sql

# Seed all test databases

seed-all: seed-pg seed-sqlite seed-mysql seed-duckdb

# Regenerate test databases from seed SQL

reseed: seed-sqlite seed-duckdb

# Open Neovim connected to the Softrear Analyst Portal (seeds on first run)

start:

nvim --cmd "set rtp^=." -c "GripStart"

# Open Neovim with the plugin loaded from this directory

dev:

nvim --cmd "set rtp^=."

# Open Neovim connected to DuckDB for httpfs testing

dev-httpfs: seed-httpfs

nvim --cmd "set rtp^=." -c "let g:db='duckdb::memory:'"

# Open Neovim and immediately connect to a SQLite test DB

dev-sqlite: seed-sqlite

nvim --cmd "set rtp^=." -c "let g:db='sqlite:tests/seed_sqlite.db'"

# ── Watch / Write mode test fixtures ─────────────────────────────────────────

# Create a local CSV that grows by one row every 4 seconds (for --watch testing).

# Run this in a separate terminal, then in Neovim: :Grip /tmp/grip_watch.csv --watch

watch-fixture:

#!/usr/bin/env bash

FILE=/tmp/grip_watch.csv

echo "id,name,score,ts" > "$FILE"

echo "1,alice,90,$(date -u +%H:%M:%S)" >> "$FILE"

echo "2,bob,75,$(date -u +%H:%M:%S)" >> "$FILE"

echo "Watching: appending to $FILE every 4s: Ctrl-C to stop"

N=3

while true; do

sleep 4

echo "$N,user_$N,$((RANDOM % 100)),$(date -u +%H:%M:%S)" >> "$FILE"

echo " row $N appended ($(wc -l < "$FILE") rows total)"

N=$((N + 1))

done

# Create a writable CSV for --write testing.

# Then in Neovim: :Grip /tmp/grip_write.csv --write

write-fixture:

#!/usr/bin/env bash

FILE=/tmp/grip_write.csv

printf 'id,first_name,last_name,score,active\n' > "$FILE"

printf '1,alice,smith,88,true\n' >> "$FILE"

printf '2,bob,jones,74,true\n' >> "$FILE"

printf '3,carol,white,95,false\n' >> "$FILE"

printf '4,dave,black,61,true\n' >> "$FILE"

printf '5,eve,grey,82,false\n' >> "$FILE"

echo "Write fixture ready: $FILE"

echo "In Neovim: :Grip /tmp/grip_write.csv --write"

# Create a Parquet fixture for --write testing (requires DuckDB).

write-fixture-parquet:

duckdb -c "COPY (SELECT i AS id, 'user_'||i AS name, (random()*100)::int AS score FROM range(1,11) t(i)) TO '/tmp/grip_write.parquet' (FORMAT PARQUET)"

echo "Parquet fixture ready: /tmp/grip_write.parquet"

echo "In Neovim: :Grip /tmp/grip_write.parquet --write"

# Open Neovim ready to test --watch (fixture must be running in another terminal)

dev-watch: seed-httpfs

nvim --cmd "set rtp^=." -c "Grip /tmp/grip_watch.csv --watch"

# Open Neovim ready to test --write on CSV

dev-write: write-fixture

nvim --cmd "set rtp^=." -c "Grip /tmp/grip_write.csv --write"

# Open Neovim ready to test --write on Parquet

dev-write-parquet: write-fixture-parquet

nvim --cmd "set rtp^=." -c "Grip /tmp/grip_write.parquet --write"

# Show git log for the current feature branch

log:

git log --oneline --graph -20

# Count lines of Lua source (excluding tests)

loc:

find lua/ -name '*.lua' | xargs wc -l | tail -1