PolyPress

Downloads · built from tagged source

Get it, or
build it.

This page lists whatever is actually published, read from the release feed at deploy time. If a build is not there, it is because it does not exist.

Both apps are in one public repository, potJim80/PolyPressApp. Every version below is built from a tag in it, so the download and the source you can read are the same code.

01

Polypress

The lossless table compressor. Python 3.9+ and numpy. A C compiler is optional — the accelerator builds itself on first import and falls back to numpy without one.

On PyPI — version 0.2.0

pip install polypress, or pip install 'polypress[parquet]' to read and write .parquet. Listed at pypi.org/project/polypress.

Or run it from source

# install into your environment
git clone https://github.com/potJim80/PolyPressApp
cd PolyPressApp/work && pip install .

# or run it without installing anything at all
python3 work/tzip.py compress data.csv

The developer path. Both forms run the same code as the published package.

02

Handrail

The macOS app that writes R into your RStudio script. Requires macOS 14 or newer. No R packages, no installer, no account.

No published build yet

There is no .dmg and no icon. Build it from source with the commands below; this section fills in automatically once a handrail-v* tag is released.

Build it from source

git clone https://github.com/potJim80/PolyPressApp
cd PolyPressApp/handrail

./build.sh                          # builds, installs to /Applications, opens it
swift run handrail-test             # 225 assertions, 16 of them running real R
swift run -c release handrail-bench <file.csv>   # re-measure the speed claims yourself

The benchmark is the honest one to run: it prints size, seconds, bytes actually read, and peak memory, on your own file rather than on ours.

Release data read at deploy time from the GitHub API. All releases