sort.sh - osm-zipcodes - Extract (dutch) addresses from OpenStreetMap OSM XML 
 (HTM) git clone git://git.codemadness.org/osm-zipcodes
 (DIR) Log
 (DIR) Files
 (DIR) Refs
 (DIR) README
 (DIR) LICENSE
       ---
       sort.sh (469B)
       ---
            1 #!/bin/sh
            2 # override malloc options for sort, else it is too slow.
            3 export MALLOC_OPTIONS="scfgju"
            4 # simple/binary collation.
            5 export LC_ALL=C
            6 
            7 # sort on city, street, housenumber
            8 sort \
            9         --mmap \
           10         -k 7,7 -k 5,5 -k 6,6n \
           11         -t "$(printf '\t')" \
           12         -S 1G \
           13         -T "$(pwd)" \
           14         -o addr_sort.csv addr.csv
           15 
           16 # unique entries, ignore latitude, longitude.
           17 uniq -f 3 addr_sort.csv > addr_sort_uniq.csv
           18 
           19 # DEBUG: show duplicate entries.
           20 #uniq -d -f 3 addr_sort.csv > addr_sort_uniq.csv