import java.util.ArrayList;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class Query {
	public static void main(String[] args) {
		String sparql1 = "prefix value: <http://dbpedia.org/resource/>\n"
				+ "prefix foaf: <http://xmlns.com/foaf/0.1/>\n"
				+ "prefix : <http://dbpedia.org/>\n"
				+ "select ?v5 ?v6 ?v9 ?v4 ?v8 where { \n"
				+ "{ value:AbboT ?v5 ?v6. ?v6 foaf:name ?v8. } \n"
				+ "UNION { ?v9 ?v5 value:AbboT; foaf:name ?v4.}\n";

		String sparql2 = "prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>"
				+ "SELECT ?v1 "
				+ "WHERE {"
				+ "{ ?v1 rdfs:label  %%v%% } UNION { ?v1 rdfs:label  %%v%% }."
				+ "FILTER (regex(str(?v1),'http://dbpedia.org/resource/') ||  regex(str(?v1,'http://dbpedia.org/ontology/') ||  regex(str(?v1),'http://www.w3.org/2002/07/owl') ||  regex(str(?v1),'http://www.w3.org/2001/ XMLSchema') ||  regex(str(?v1),'http://www.w3.org/2000/01/rdf-schema') ||  regex(str(?v1),'http://www.w3.org/1999/02/22-rdf-syntax -ns')) }";

		String sparql3 = "Prefix dbpedia: <http://dbpedia.org/resource/>\n"
				+ "Prefix dbp-owl: <http://dbpedia.org/ontology/>\n"
				+ "Prefix dbp-prop: <http://dbpedia.org/property/>\n"
				+ "Prefix dbp-yago: <http://dbpedia.org/class/yago/>\n"
				+ "Prefix dbp-cat: <http://dbpedia.org/resource/Category/>\n"
				+ "Prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax -ns#>\n"
				+ "Prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>\n"
				+ "Prefix owl: <http://www.w3.org/2002/07/owl#>\n"
				+ "Prefix xsd: <http://www.w3.org/2001/ XMLSchema#>\n"
				+ "Prefix skos: <http://www.w3.org/2004/02/skos/core#>\n"
				+ "Prefix foaf: <http://xmlns.com/foaf/0.1/>\n"
				+ "Prefix georss: <http://www.georss.org/georss/>\n"
				+ "SELECT * WHERE { ?v6 a dbp-owl:PopulatedPlace; dbp-owl:abstract ?v1; rdfs:label ?v2; geo:lat ?v3; geo:long ?v4. {?v6 rdfs:label  %%v%%.}  UNION { ?v5 dbp-prop:redirect ?v6; rdfs:label  %%v%%. } OPTIONAL { ?v6 foaf:depiction ?v8 } OPTIONAL { ?v6 foaf:homepage ?v10 } OPTIONAL { ?v6 dbp-owl:populationTotal ?v12 } OPTIONAL { ?v6 dbp-owl:thumbnail ?v14 } FILTER ( langMatches( lang(?v1), 'de') &&  langMatches( lang(?v2), 'de') )";

		sparqlParser(sparql3);
	}

	public static ArrayList<String[]> sparqlParser(String q) {
		// Replace prefixes
		Pattern prefixPattern = Pattern.compile(
				"prefix\\s+([a-z\\-]*:)\\s+<(.+)>", Pattern.CASE_INSENSITIVE);
		Matcher m = prefixPattern.matcher(q);

		int start = 0;
		while (m.find(start)) {
			start = m.end();
			q = q.replace(m.group(), "");
			q = q.replaceAll("\\s+" + m.group(1), " " + m.group(2));
		}
		// ************ END Replace Prefixes ************** //

		// Some Pre-processing to query //

		// Remplace all brackets, and extra white space to one space
		q = q.replaceAll("[{}\\s+]", " ");

		// Remove the terms union and optional with anycase
		q = q.replaceAll(
				"[uU][nN][iI][oO][nN]|[oO][pP][tT][iI][oO][nN][aA][lL]", "");

		// Place a space before any semi-colon
		q = q.replaceAll(";", " ;");

		// Make sure any extra white space is removed
		q = q.replaceAll("\\s+", " ");

		// Only get the where clause that we need to work on.
		Pattern where = Pattern.compile("where\\s*(.*)",
				Pattern.CASE_INSENSITIVE);
		m = where.matcher(q);
		if (m.find())
			q = m.group(1);

		// Remove any white space between "FILTER" and "("
		q = q.replaceAll("([Ff][Ii][Ll][Tt][Ee][Rr])\\s+(\\()", "$1$2");

		// Remove any white space before any .
		q = q.replaceAll("\\s+\\.", "\\.");

		// ******* End of Pre-Processing Query ******** //

		// Remove Filters //
		Pattern filterPattern = Pattern.compile("FILTER\\(",
				Pattern.CASE_INSENSITIVE);
		m = filterPattern.matcher(q);
		start = 0;
		while (m.find(start)) {
			q = removeFilterFromSparql(q, m.start());
			start = m.end();
		}
		// ********** Remove Filters *********** \\
		// System.out.println(q);

		// Split query to parts
		String[] queryArray = q.split(" ");
		ArrayList<String[]> parts = new ArrayList<>();
		for (int i = 0; i < queryArray.length; i++) {
			if (i % 3 == 0)
				parts.add(new String[3]);
			parts.get(i / 3)[i % 3] = queryArray[i];
			if (queryArray[i].matches(";"))
				parts.get(i / 3)[i % 3] = parts.get(i / 3 - 1)[0];

			if (i % 3 == 2)
				parts.get(i / 3)[2] = parts.get(i / 3)[2].replaceAll(
						"\\s*\\.$", "");
		}
		// **** End Split Query to Parts ****** //

		for (int i = 0; i < parts.size(); i++) {
			for (int j = i + 1; j < parts.size(); j++) {
				if (parts.get(i)[0].equals(parts.get(j)[0])
						&& parts.get(i)[1].equals(parts.get(j)[1])
						&& parts.get(i)[2].equals(parts.get(j)[2])) {
					parts.remove(j);
					j--;
				}
			}
		}

		return parts;
	}

	public static String removeFilterFromSparql(String q, int filterStart) {
		// First character after "filter("
		int i = filterStart + 6;

		int brackets = 1;
		while (brackets > 0 && ++i < q.length()) {
			if (q.charAt(i) == '(')
				brackets++;
			else if (q.charAt(i) == ')')
				brackets--;
		}

		q = q.substring(0, filterStart).concat(
				q.substring(i - 1, q.length() - 1));
		return q;
	}
}