A genuine interface to the telefonliste

telefonliste.py

#! /usr/bin/env nix-shell
#! nix-shell -i python3
#! nix-shell -p python3Packages.requests
#! nix-shell -p python3Packages.pandas
#! nix-shell -p python3Packages.openpyxl
#! nix-shell -p python3Packages.click
#! nix-shell -p python3Packages.unidecode
#! nix-shell -p python3Packages.thefuzz

import os
import json
import requests
import pandas as pd
import click
import unidecode

from thefuzz import fuzz

from email.utils import formatdate


URL = "https://intranet.hs-fulda.de/fileadmin/RZ/Dokumente/Telefonliste.xlsx"

XDG_CACHE = os.environ.get("XDG_CACHE_HOME", os.path.expanduser("~/.cache"))
CACHE_DIR = os.path.join(XDG_CACHE, "phonebook-cli")
XLSX_FILE = os.path.join(CACHE_DIR, "phonebook.xlsx")
PARSED_FILE = os.path.join(CACHE_DIR, "phonebook.json")

os.makedirs(CACHE_DIR, exist_ok=True)


def download(force=False):
    headers = {}

    if not force:
        try:
            mtime = os.path.getmtime(XLSX_FILE)
        except FileNotFoundError:
            pass
        else:
            headers["If-Modified-Since"] = formatdate(mtime, usegmt=True)

    response = requests.get(URL, headers=headers, stream=True)
    if response.status_code == 304:
        return force  # Not modified – if forced, we update anyways

    response.raise_for_status()

    with open(XLSX_FILE, "wb") as f:
        for chunk in response.iter_content(chunk_size=8192):
            f.write(chunk)

    return True


def parse_and_cache():
    records = pd.read_excel(XLSX_FILE, engine="openpyxl")
    records = records.to_dict(orient="records")

    with open(PARSED_FILE, "w") as f:
        json.dump(records, f)

    return records


def load_cached():
    try:
        with open(PARSED_FILE, "r") as f:
            return json.load(f)

    except FileNotFoundError:
        return parse_and_cache()


def load(force):
    if download(force):
        return parse_and_cache()

    else:
        return load_cached()


def normalize(text):
    return unidecode.unidecode(str(text).lower())


def search_name(name_query, force_update=False):
    name_query_norm = normalize(name_query)

    records = load(force_update)

    matches = []
    for rec in records:
        full_name = f"{rec.get("Vorname", "")} {rec.get("Name", "")}"
        full_name = normalize(full_name)
        score = fuzz.partial_ratio(name_query_norm, full_name)
        if score >= 80:
            matches.append(rec)

    return matches


@click.command()
@click.argument("name")
@click.option("--update/--no-update",
              default=False,
              help="Force update before searching")
def main(name, update):
    """Search for a NAME in the phonebook"""
    results = search_name(name, force_update=update)
    if results:
        for rec in results:
            gn = rec.get("Vorname", "")
            sn = rec.get("Name", "")

            tel = rec.get("Telefon")
            if tel:
                tel = float(tel)

            if tel.is_integer():
                tel = str(int(tel))
            else:
                tel = ""

            click.echo(f"{gn} {sn} - {tel}")

    else:
        click.echo("No results found.")


if __name__ == "__main__":
    main()

As a derivation:

let
  script = pkgs.fetchurl {
    url = "https://gist.githubusercontent.com/fooker/db6948cb3c3551e5516dd6c875f6de05/raw/6dd568aa3494712c94f6f7a2976fc1248541fc8e/telefonliste.py";
  };
in
pkgs.writers.writePython3Bin "telefonliste" {
  flakeIgnore = [ "E265" ];
  libraries = with pkgs.python3Packages; [
    requests
    pandas
    openpyxl
    click
    unidecode
    thefuzz
  ];
} script;