Download, unzip and parse Census TIGER files for U.S. places, return a pipe-delimited text file with lat/lng.

place_getter.sh

#!/bin/bash

wget -O fips_ref.txt http://www2.census.gov/geo/docs/reference/state.txt
wget --mirror --continue --no-directories ftp://ftp2.census.gov/geo/tiger/TIGER2015/PLACE/

for f in *.zip
do
    unzip $f
done

echo "state_fips,place_fips,place,lat,lng" >> places.txt

REC_COUNTER=0
for f in *.dbf
do
    in2csv -f dbf -v $f | tail -n +2 | csvcut -c 1,2,5,15,16 >> places.txt
    DBF="$(file -b $f)"
    DB_SIZE="$(echo $DBF | grep -Eo "[0-9]+ records" | sed 's/ records//g')"
    let REC_COUNTER+=$DB_SIZE
done

# compare record count
echo "Database records: $REC_COUNTER"
wc -l places.txt

# The DBF files are ASCII encoded, and a few had non-ASCII characters that borked on in2csv conversion. Here, we pause to manually convert those files to UTF-8 and add to places.txt. 

# join on FIPS file to get state names
csvcut -c 1,3 -d "|" fips_ref.txt > states.txt
csvjoin -c "state_fips,STATE" places.txt states.txt | csvcut -c 1,2,3,4,5,7 | csvsort -c 1 | csvformat -D "|" > us_places.txt

# clean up
rm .listing fips_ref.txt places.txt

for f in *.dbf *.shp *.xml *.cpg *.prj *.shx
do
    rm $f
done