dkapitan

show-project-size:
  uv pip list --format freeze | \
  awk -F = {'print $1'} | \
  xargs uv pip show | \
  grep -E 'Location:|Name:' | \
  cut -d ' ' -f 2 | \
  paste -d ' ' - - | \
  awk '{print $2 "/" tolower($1)}' | \
  xargs du -sh 2> /dev/null | \
  sort -hr

dkapitan

def get_binary(url):
    """Download binary file in memory without saving to disk."""    
    from io import BytesIO
    import requests
    

    try:
        response = requests.get(url)

        # Raise an exception if the request was unsuccessful (e.g., 404 Not Found)

dkapitan

python3 -m pip install \
                --config-settings="--global-option=build_ext" \
                --config-settings="--global-option=-I$(brew --prefix graphviz)/include/" \
                --config-settings="--global-option=-L$(brew --prefix graphviz)/lib/" \
                pygraphviz

dkapitan

uv_poetry_install () {
    uv venv
    uv pip install --no-deps -r <(POETRY_WARNINGS_EXPORT=false poetry export --without-hashes --with dev -f requirements.txt)
    poetry install --only-root
}

dkapitan

runtime: python
env: flex
instance_class: F2
runtime_config:
  operating_system: "ubuntu22"
  runtime_version: "3.12"

# name of app
service: some-name

dkapitan

dkapitan

import re
import requests
import urllib


PDOK = "https://api.pdok.nl/bzk/locatieserver/search/v3_1/free"


def query_pdok(query):
    query_params = {

dkapitan

import ndjson


def redact(patient: dict) -> dict:
    for key in ["name", "telecom"]:
        patient.pop(key, None)
    return patient


with open("patient.ndjson") as f:

dkapitan

Objectives

• Get to know Debian with very minimal install to run RoonServer
• Use Debian as Linux Docker Engine platform for dev/test

Debian 12 setup

• Download Debian live standard (no desktop manager, link)
• add user to sudo
• su root

dkapitan

Upcoming Python libraries with a Rust runtime that are useful for the composable data stack:

• greptime: an open-source, cloud-native, distributed time-series database with PromQL/SQL/Python supported
• tokenizers: fast state-of-the-art tokenizers from Huggingface optimized for research and production
• pixi: package management tool for developers, build on top of the conda packaging ecosystem
• polars: lightning-fast DataFrame library for Rust and Python
• pydantic: most widely used data validation library for Python, implemented in Rust since v2
• robyn: fast, innovator friendly, and community driven Python-Rust web framework
• RustPython: RustPython is a Python interpreter written in Rust. RustPython can be embedded into Rust programs to use Python as a scripting language for your application