maddytae/pytae

README 'Quick Start: CLI' example 2: pytae penguins.parquet -filter/-mutate/-arrange/-select -o heavy_adelie.csv (run against the pinned source installed into a venv); output wrote 92 rows and CSV contents were verified by head.

cd <host-path> && git log --oneline -1 2>/dev/null; ls -la; cat README* 2>/dev/null | head -200
cat <host-path> | head -150; ls -la <host-path> 2>/dev/null
cd <host-path> && cat readme.md
cd <host-path> && python3 --version && (python3 -m venv <host-path> 2>&1 || (apt-get update -qq && apt-get install -y -qq python3-venv >/dev/null 2>&1 && python3 -m venv <host-path>)) && <host-path> install -q --upgrade pip && <host-path> install -q "./[plot]" 2>&1 | tail -20; echo rc=$?; ls <host-path> | grep -i pytae
sudo -n true && echo SUDO_OK; sudo apt-get update -qq && sudo apt-get install -y -qq python3.12-venv >/dev/null 2>&1; echo apt_rc=$?; rm -rf <host-path>; python3 -m venv <host-path> && <host-path> install -q --upgrade pip && cd <host-path> && <host-path> install -q "./[plot]" 2>&1 | tail -20; echo rc=$?; ls <host-path> | grep -i pytae
cd /tmp && <host-path> --help 2>&1 | head -40; <host-path> -c "import pytae as pt; print(pt.__file__); df=pt.sample('penguins'); print(type(df))" 2>&1 | tail -3
mkdir -p <host-path> && cd <host-path> && <host-path> -c "import pytae as pt; pt.sample('penguins').to_polars().write_parquet('penguins.parquet')" && rm -f heavy_adelie.csv && <host-path> penguins.parquet -shape -head 5; echo rc=$?; <host-path> penguins.parquet \
  -filter "species = 'Adelie', body_mass_g > 3500" \
  -mutate "mass_kg = body_mass_g / 1000" \
  -arrange "mass_kg desc" \
  -select "species,island,mass_kg" \
  -o heavy_adelie.csv; echo rc=$?; ls -la; head -3 heavy_adelie.csv
cd <host-path> && rm -f heavy_adelie.csv && mkdir -p <host-path> && cat > <host-path> <<'EOF'
{"kind":"cli","argv":["<host-path>","-filter","species = 'Adelie', body_mass_g > 3500","-mutate","mass_kg = body_mass_g / 1000","-arrange","mass_kg desc","-select","species,island,mass_kg","-o","<host-path>","expected_text":"Wrote 92 rows to <host-path>"}
EOF
cat > <host-path> <<'EOF'
{"outcome":"working","operation":"README 'Quick Start: CLI' example 2: pytae penguins.parquet -filter/-mutate/-arrange/-select -o heavy_adelie.csv (run against the pinned source installed into a venv); output wrote 92 rows and CSV contents were verified by head.","operation_scope":"CLI pipeline only. Input penguins.parquet was generated from the bundled pt.sample('penguins') because the README's sample file is not in the repo. Visualization (-plot, needs matplotlib, installed via [plot] extra) and MCP are not exercised; pytae is not an MCP server.","steps":["Read readme.md and pyproject.toml","apt-get install python3.12-venv (via sudo)","python3 -m venv <host-path>","pip install ./[plot] from <host-path> (installs pytae CLI)","Generated <host-path> via pt.sample('penguins')","Ran README CLI example 2 with -o output: 'Wrote 92 rows to heavy_adelie.csv'"],"deviations":["README CLI example 1 'pytae penguins.parquet -shape -head 5' fails with: 'pytae: error: -shape does not return a DataFrame/Series, so no flag may follow it'. This is a product behavior, not an environment issue; I did not use it as the probe.","README assumes penguins.parquet exists; not shipped in repo, generated it from pt.sample('penguins') outside the workspace.","python3-venv was missing; installed python3.12-venv via sudo apt-get."],"stuck_step":null,"main_blocker":null,"suggested_fix":null}
EOF
cd <host-path> && <env-var>=<host-path> python3 -I <host-path> --validate; echo rc=$?