dbpedia-snippets.sh

dbpedia-snippet-01.sh

# some examples of quering dbpedia from the command line
# knb 20201218
#
# show serialization formats that dbpedia can return
# (other than text/turtle)
# 
# curl -sL: silent, follow redirects
# curl -I : HTTP HEAD request
# RDF triples: all facts about Pudding

# Hypernyms (~subtypes of) Pudding. For uni2utf8.pl see Stackoverflow
curl -sL  --header "Accept: text/turtle" "http://dbpedia.org/resource/Pudding"  \
    | rapper -q -i turtle - http://dbpedia.org/resource \
    | grep  hypernym  | sort | uni2utf8.pl

# write it to a local file and pretty-print RDF
rapper  -i turtle ./pudding.ttl

# returns 917 facts in 2022
rapper --count -i turtle ./pudding.ttl


# HTTP HEAD request to DBpedia, Pudding. (/resource -> ...3 redirects... -> /data)
curl -sL --no-progress-meter   --head --header "Accept: text/turtle" "http://dbpedia.org/resource/Pudding"



## shorter version of previous command:
## --no-progress-meter and -s are the same, --head and -I are the same.
## same command as before (HTTP HEAD request),  but grab the "Alternates: " response-header, 
## and make it more humanreable.     
# HTTP HEAD request to DBpedia - show available formats
curl -sL -I --header "Accept: text/turtle" "http://dbpedia.org/resource/Pudding"  \
 | awk 'BEGIN {FS=": "}/^link:/{print $2}' \
 | perl -pE "s/, ?/\n/g; s/0{3,}//g;" \
 | sort -n -k2 -k1 
 
# HTTP GET
# count different @prefixes used in result  
 curl -sL -H "Accept: text/turtle" "http://dbpedia.org/resource/Pudding"   | grep @prefix | wc -l
 # (n = 44)
 
 # same request, to .../data/... URL not .../resource/... link
 curl -sL  "http://dbpedia.org/data/Pudding.ttl"  | grep @prefix | wc -l
 # (n = 24)
 
 # have a look at the actual triles returned - 
 # for uni2utf8.pl, see below
 curl -sL -H "Accept: text/turtle" "http://dbpedia.org/resource/Pudding" | uni2utf8.pl

dbpedia-snippet-02.sh

##!/usr/bin/env bash
# return json for 1 URL, return URLs with related resources 
curl -sL -H "Accept: application/json" https://dbpedia.org/resource/Potsdam \
  | jq '. | keys_unsorted ' | jq .  
  | perl -pE "s#http://dbpedia.org/resource/##g"  | sort | cat -n

Search-for-people-emigrants-example.sparql

# from someone's Twitter timeline, Dec 2022
# Wikidata: search for emigrants during Nazi era
SELECT DISTINCT ?category ?person ?description  WHERE { 
  ?category a skos:Concept . 
  ?category skos:prefLabel ?label . 
  FILTER (CONTAINS(?label, "Zeit des Nationalsozialismus")) 
?person dct:subject ?category. 
?person rdfs:comment ?description . 
      FILTER (LANG(?description) = 'de') . }

uni2utf8.pl: helper script for converting \u1234 sequences to characters the terminal can display.
example: turn Tavuk_g\u00F6\u011Fs\u00FC into Tavuk_göğsü
example call: curl -sL -H "Accept: text/turtle" "http://dbpedia.org/resource/Pudding" | uni2utf8.pl
source: stackoverflow

#!/usr/bin/perl
# uni2ascii.pl
# helper script
# convert \u1234 sequences to characters the terminal can display.
# save this as ~/bin/uni2utf8.pl
# run example:
# curl -sL -H "Accept: text/turtle" "http://dbpedia.org/resource/Pudding"   | uni2utf8.pl
use strict;
use warnings;

binmode(STDOUT, ':utf8');

while (<>) {
    s/\\u([0-9a-fA-F]{4})/chr(hex($1))/eg;
    print;
}