asparagus

#!/bin/php
<?php
/**
 * CLI interface for the Wikidata query service. Build on the Wikimedia Hackathon
 * in Lyon.
 *
 * @author Marius Hoch < hoo@online.de >
 * @license GPL-2.0+
 */

$endPoint = 'https://wdqs-beta.wmflabs.org/bigdata/namespace/wdq/sparql?query=$1';
$defaultPrefixes = 'PREFIX wd: <http://www.wikidata.org/entity/>
PREFIX entity: <http://www.wikidata.org/entity/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wikibase: <http://wikiba.se/ontology#>
PREFIX p: <http://www.wikidata.org/prop/>
PREFIX v: <http://www.wikidata.org/prop/statement/>
PREFIX q: <http://www.wikidata.org/prop/qualifier/>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>';

/**
 * @param string $query
 * @param string $endPoint
 * @param string $responseFormat
 *
 * @return string|bool Result
 */
function doQuery( $query, $endPoint, $responseFormat ) {
	$url = str_replace(
		'$1',
		urlencode( $query ),
		$endPoint
	);

	$opts = array();
	if ( $responseFormat !== null ) {
		$opts['header'] = "Accept: $responseFormat\r\n";
	}

	$context = stream_context_create( array( 'https' => $opts, 'http' => $opts ) );

	// @TODO: Better error handling
	return @file_get_contents( $url, false, $context );
}

/**
 * Reads the user input and returns once a full query has been entered,
 * or the user wants to exit.
 *
 * @return string|bool The query, or false if the user wants to exit
 */
function readInput() {
	$query = '';

	// Whether we saw an initial opening {. This assumes all queries contain at least one {…} section.
	$sawInitialBracket = false;

	// Number of open { sections
	$brackC = 0;

	// Number of single quotes seen
	$singleQuoteC = 0;

	// Number of double quotes seen
	$doubleQuoteC = 0;

	// Number of consecutive backslashes seen
	$consecutiveBackslashC = 0;

	while ( $brackC !== 0 || !$sawInitialBracket ) {
		$input = readline( $query === '' ? '> ' : '… ' );

		if ( $query === '' && ( $input === 'quit' || $input === 'exit' ) ) {
			return false;
		}

		for ( $i = 0; $i < strlen( $input ); $i++ ) {
			// { or }
			if ( $doubleQuoteC % 2 === 0 && $singleQuoteC % 2 === 0 ) {
				if ( $input[$i] === '{' ) {
					$sawInitialBracket = true;
					$brackC--;
					continue;
				} elseif ( $input[$i] === '}' ) {
					$brackC++;
					continue;
				}
			}

			// Backslashes
			if ( $input[$i] === '\\' ) {
				$consecutiveBackslashC++;
				continue;
			} else {
				$consecutiveBackslashC = 0;
			}

			// (Unescaped) quotes
			if ( $input[$i] === '"' && $consecutiveBackslashC % 2 === 0 ) {
				$doubleQuoteC++;
				continue;
			} elseif ( $input[$i] === "'" && $consecutiveBackslashC % 2 === 0 ) {
				$singleQuoteC++;
				continue;
			}
		}

		if ( preg_match( '/\\\\[gG]$/', $input ) ) {
			$input = preg_replace( '/\\\\[gG]$/', '', $input );
			$query .= $input;
			return $query;
		}

		if ( $input !== '' ) {
			$query .= $input . "\n";
		}
	}

	// If the query didn't end on a \g or \G, read on until an empty input line occurs
	while ( true ) {
		$input = readline( '… ' );

		if ( $input === '' ) {
			break;
		}

		if ( preg_match( '/\\\\[gG]$/', $input ) ) {
			$input = preg_replace( '/\\\\[gG]$/', '', $input );
			$query .= $input;
			return $query;
		}

		$query .= $input . "\n";
	}

	return $query;
}

if ( $argc > 1 && ( $argv[1] === '--help' || $argv[1] === '-h' ) ) {
	echo "BlazeGraph cli interface targeted at the Wikidata Query service.\n";
	echo "Runs runs queries interactively.\n\n";
	echo "Usage: " . $argv[0] . " [--response=format-mime-type] [--no-default-prefixes]\n";

	exit( 1 );
}

$longopts = array(
	'response::', // [--response=format]
	'no-default-prefixes', // [--no-default-prefixes]
);
$opts = getopt( '', $longopts );

$response = isset( $opts['response'] ) ? $opts['response'] : 'application/sparql-results+json';

echo "Enter SPARQL queries delimited by either an empty line after the query, a \\g or a \\G.
Quit by typing 'quit' or 'exit'.\n";

while( true ) {
	$query = readInput();
	if ( $query === false ) {
		if ( mt_rand( 0, 10 ) === 8 ) {
			die( "Here, have a kitten: https://commons.wikimedia.org/wiki/File:Cat_deck.PNG\n" );
		}
		die( "cu\n" );
	}

	if ( !isset( $opts['no-default-prefixes'] ) ) {
		$query = $defaultPrefixes . "\n" . $query;
	}

	$result = doQuery( $query, $endPoint, $response );
	if ( $result === false ) {
		echo "Query execution failed.\n";
	} else {
		echo $result . "\n";
	}
}