statistics/byPeriodSlice.php

<?php
/*
 * Generates slice data for statistics of posts by date and sender
 * Copyright (C) 2020  Polyna <https://wandystan.eu/B196>
 *
 * This file is part of LDMW statistical scripts.
 *
 * LDMW statistical scripts are free software: you can redistribute them
 * and/or modify them under the terms of the GNU Affero General Public
 * License as published by the Free Software Foundation, either version 3
 * of the License, or (at your option) any later version.
 *
 * LDMW statistical scripts are distributed in the hope that they will be
 * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public
 * License along with LDMW statistical scripts.  If not, see
 * <https://www.gnu.org/licenses/>.
 */
require_once 'common.php';

// get request data and validate it
$param_date_start = $_GET['date-start'];
$param_date_end = $_GET['date-end'];

if (! $param_date_start) {
	exit_error('No start date specified.');
} elseif (! $param_date_end) {
	exit_error('No end date specified.');
} elseif (! mb_ereg_match('^' . REGEXP_ISO8601_DATE_OR_INTERVAL . '$', $param_date_start)) {
	exit_error('Invalid start date.');
} elseif (! mb_ereg_match('^' . REGEXP_ISO8601_DATE_OR_INTERVAL . '$', $param_date_end)) {
	exit_error('Invalid end date.');
} elseif (starts_with($param_date_start, 'P') && starts_with($param_date_end, 'P')) {
	exit_error('Both start and end dates cannot be periods.');
}

// create start and end date objects
$tz = new DateTimeZone('UTC');

try {
	if (! starts_with($param_date_start, 'P'))
		$date_start = new DateTimeImmutable($param_date_start, $tz);

	if (! starts_with($param_date_end, 'P'))
		$date_end = new DateTimeImmutable($param_date_end, $tz);

	if (starts_with($param_date_start, 'P')) {
		$interval = new DateInterval($param_date_start);
		$date_start = $date_end->sub($interval);
	}

	if (starts_with($param_date_end, 'P')) {
		$interval = new DateInterval($param_date_end);
		$date_end = $date_start->add($interval);
	}
} catch (Exception $e) {
	exit_error('Invalid date specification: ' . $e->getMessage());
}

// get page with the search results for the given period
try {
	$document = load_remote_dom_document('https://wandystan.groups.io/g/wandystan/search?p=Created,,,10000,2,0,0&d=6&startdate=' . $date_start->format('m/d/Y') . '&enddate=' . $date_end->modify('-1 second')->format('m/d/Y'));
} catch (Exception $e) {
	exit_error($e->getMessage());
}

$xpath = new DOMXPath($document);

// find all messages in search results and summarise them
$messages = $xpath->query('//div[@id = "maincontent"]/table//td[span[@class = "subject"]]');
$summary = [];

foreach ($messages as $message) {
	$sender = trim($xpath->evaluate('substring-before(substring-after(normalize-space(div[@class = "hidden-xs"]/following-sibling::text()[1]), "By "), " · #")', $message), '"');
	$timestamp = $xpath->evaluate('substring-before(substring-after(.//script[@class = "timedisp"], "DisplayShortTime("), ",")', $message);

	$date = new DateTime(null, $tz);
	$date->setTimestamp($timestamp / 1000000000);

	$key = serialize([
		'date' => $date->format('Y-m-d'),
		'sender' => $sender
	]);
	$summary[$key]++;
}

// construct a linked data object corresponding to gathered data
$ld = [
	'@context' => 'https://wandystan.eu/statistics/context.jsonld',
	'@type' => 'Slice',
	'key' => 'key/by-period',
	'period' => [
		'@type' => 'Interval',
		'hasBeginning' => [
			'@type' => 'Instant',
			'timestamp' => $date_start->format(DATE_ATOM)
		],
		'hasEnd' => [
			'@type' => 'Instant',
			'timestamp' => $date_end->format(DATE_ATOM)
		]
	]
];

$observations = [];
foreach ($summary as $key => $entry) {
	$key = unserialize($key);
	$nameHash = hash('fnv1a32', $key['sender']);
	$observations []= [
		'@id' => 'by-period/' . $key['date'] . '/P1D#' . $nameHash,
		'@type' => 'Observation',
		'dataset' => "by-period/",
		'date' => $key['date'],
		'sender' => [
			'@id' => 'sender/' . $nameHash,
			'@type' => 'Agent',
			'name' => $key['sender'],
			'name_fnv1a32sum' => $nameHash
		],
		'posts' => $entry
	];
}

$ld['observations'] = $observations;

header('Content-type: application/ld+json');
print json_encode($ld);