cz-region-boundaries snippets

refresh-data.sh

#!/bin/bash
# GPS data passed through https://github.com/tyrasd/osmtogeojson
curl -X POST 'https://lz4.overpass-api.de/api/interpreter' -d@overpass-query.txt | osmtogeojson > src.geojson
# utility script from https://github.com/JirkaChadima/cz-region-boundaries/blob/master/scripts/prepare-gps-data.js
node prepare-gps-data.js

# ZIP codes processing
PSCSOURCE='https://www.ceskaposta.cz/documents/10180/3738087/xls_pcobc.zip/50617e56-6e9a-4335-9608-96fec214e6ef'
# https://github.com/JirkaChadima/cz-region-boundaries/blob/master/data/zip/county-region.csv, collected from https://cs.wikipedia.org/wiki/Okresy_v_%C4%8Cesku#Okresy_podle_samospr%C3%A1vn%C3%BDch_kraj%C5%AF
OKRESFILE="../data/zip/county-region.csv"
ZIPFILE="src.zip"
FILE='zv_pcobc'
PSCFILE="../data/zip/zip-county.csv"
RESULT="../data/zip/zip-region.csv"

wget "$PSCSOURCE" -O src.zip

unzip -n "$ZIPFILE"
libreoffice --headless --convert-to csv --infilter=csv:44,34,76 "$FILE.xls" --outdir . > /dev/null
cut -d, -f2,5 "$FILE.csv" | tail -n +2 | sort | uniq > "$PSCFILE"
# This bash magic is basically an Excel contingency table
join -t ',' -1 2 -2 1 -o 1.1,2.2 <(sort -t ',' -k2,2 "$PSCFILE") <(sort -t ',' -k1,1 "$OKRESFILE") | sort | uniq > "$RESULT"
rm "$ZIPFILE"
rm "$FILE.xls"
rm "$FILE.csv"