Last active
March 11, 2022 10:49
-
-
Save arbakker/e158d992558f23edd6512b382088987f to your computer and use it in GitHub Desktop.
Bash script for scraping WFS services in a responsible way
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env bash | |
# Bash script for scraping WFS services in a responsible way. Script divides in area of interest in list of bounding boxes, which are requested sequentially. The HTTP get request is configured to retry, using the default CURL exponential backoff algorithm. | |
# Only use when no bulk download (like ATOM) service is available. | |
set -eu | |
TARGET_GPKG=output.gpkg | |
WFS_URL="https://service.pdok.nl/prorail/spoorwegen/wfs/v1_0" | |
FT_NAME=spoorwegen:kilometrering | |
LAYER_NAME=$(cut -d":" -f2 <<<$FT_NAME) | |
SLEEP=1 | |
# extent nl (vasteland): 6500,310000,290000,620000 - RD coordinaten in meters | |
NL_MINX=6500 | |
NL_MINY=310000 | |
NL_MAXX=290000 | |
NL_MAXY=620000 | |
GRID_SIZE=10000 # 10 KM > resulteert in 928 bboxen | |
function get_features() { | |
bbox="$1" | |
page_size=1000 | |
i=0 | |
while true; do # loop until no more features are returned by WFS (wfs hit count is not reliable AFAIK) | |
uuid=$(uuidgen) | |
start_index=$(bc <<<"$page_size*$i") | |
url="${WFS_URL}?SERVICE=WFS&VERSION=2.0.0&bbox=${bbox}&REQUEST=GetFeature&TYPENAMES=${FT_NAME}&COUNT=${page_size}&STARTINDEX=${start_index}" | |
tmp_gml="/tmp/${uuid}.gml" | |
echo "requesting ${url}" | |
curl -A "pdok-wfs-download-script" -sv --retry-max-time 60 --retry 10 --retry-connrefused --connect-timeout 10 "$url" >"$tmp_gml" | |
ft_count=$(ogrinfo "$tmp_gml" "$LAYER_NAME" -so | grep "Feature Count" | cut -d" " -f 3) | |
if [[ $ft_count -eq 0 ]]; then | |
rm -f "$tmp_gml" | |
rm -f "${uuid}.gfs" # ogr generates gfs file when opening gml | |
break | |
else | |
append="" | |
if [[ -f "$TARGET_GPKG" ]]; then | |
append="-append" | |
fi | |
ogr2ogr -f GPKG $append "$TARGET_GPKG" "$tmp_gml" "$LAYER_NAME" | |
rm -f "$tmp_gml" | |
fi | |
i=$(bc <<<"$i+1") | |
done | |
} | |
for minx in $(# overshoot van 1 grid cell in het geval er geen remainder is in (nl_maxx-nl_minx)/GRID_SIZE | |
seq $NL_MINX $GRID_SIZE $NL_MAXX | |
); do | |
for miny in $(seq $NL_MINY $GRID_SIZE $NL_MAXY); do | |
maxx=$(bc <<<"$minx+$GRID_SIZE") | |
maxy=$(bc <<<"$miny+$GRID_SIZE") | |
bbox="$minx,$miny,$maxx,$maxy" | |
ft_count=0 | |
if [[ -f "$TARGET_GPKG" ]]; then | |
ft_count=$(ogrinfo "$TARGET_GPKG" "$LAYER_NAME" -spat $(tr "," " " <<<"$bbox") -so | grep "Feature | |
Count" | cut -d" " -f 3) | |
fi | |
if [[ $ft_count -eq 0 ]]; then | |
get_features "$bbox" | |
sleep $SLEEP | |
else | |
echo "FEATURE COUNT: ${ft_count}, BBOX: ${bbox}. SKIPPING WFS REQUEST." | |
fi | |
done | |
done |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment