A quick command line PHP script to scrape "week by week" pregnancy updates from thebump.com's API into local .html files.

weekbyweek.php

<?php
/**
 * A command line script to grab the "week-by-week" updates from
 * TheBump.com's API and save them to local .html files.
 *
 * Tries to be respectful and cache all results to avoid unecessary HTTP
 * calls. Uses the API key that thebump.com's own website uses. No malice
 * is intended by this.
 *
 * Usage:
 *
 *   $ mkdir weeks
 *   $ cd weeks/
 *   # (Save the script in this folder as `weekbyweek.php`)
 *   $ php weekbyweek.php
 *   # (Files like `week_4.json` and `week_4.html` will be created in this folder.)
 *   # (This process only needs to be run once to generate the .html output files.)
 *
 * Then from another page, to embed a given week's html block:
 *
 *   <?php include 'weeks/week_4.html'; ?>
 *
 * @author [email protected]
 * @version 1.0.0
 */


/**
 * The base URL to make API requests against.
 *
 * There is a `%s` token that will be replaced by the current week in
 * each request.
 *
 * @var string
 */
$urlPattern = 'http://services.thebump.com/core/v1//feeds/pregnant/static?apikey=mtycrccp3w2tefa7gud68jfvzvgt9gw7sv8&week=%s';

/**
 * main()
 *
 * Loops through the available pregenancy weeks and either attempts to
 * query thebump.com's API (saving the JSON results), or uses the
 * already-saved results to extract data and format it into HTML blocks.
 *
 * The first time this process is run, it will create a bunch of files
 * in whatever directory this script exists in.
 */
for ($i = 4; $i <= 42; $i++) {
	echo PHP_EOL . '==============================';
	echo PHP_EOL . "Processing week {$i}...";

	$jsonFile = "week_{$i}.json";
	$htmlFile = "week_{$i}.html";

	// Cache the JSON api responses so we don't annoy the API server.
	if (is_readable($jsonFile)) {
		echo PHP_EOL . "Using existing JSON file `{$jsonFile}`.";
		$json = file_get_contents($jsonFile);
	} else {
		$url = sprintf($urlPattern, $i);
		echo PHP_EOL . "Fetching URL `{$url}`.";
		$json = fetchHtml($url);
		echo PHP_EOL . "Saving JSON to file `{$jsonFile}`.";
		file_put_contents($jsonFile, $json);
	}

	// Cache the final HTML blocks as files for easy inclusion elsewhere.
	if (is_readable($htmlFile)) {
		echo PHP_EOL . "Using existing HTML file `{$htmlFile}`.";
		$html = file_get_contents($htmlFile);
	} else {
		echo PHP_EOL . "Converting JSON and saving HTML to file `{$htmlFile}`.";
		$html = formatBlock(extractBlock($json), $i);
		file_put_contents($htmlFile, $html);
	}

	echo PHP_EOL . "Final HTML for week {$i}:" . PHP_EOL . $html . PHP_EOL . PHP_EOL;
}

echo PHP_EOL . 'Processing complete!' . PHP_EOL . PHP_EOL;


/**
 * Fetch the contents of the provided URL and return as a string.
 *
 * Your copy of PHP must have the CURL extension installed in order for
 * this to work.
 *
 * @param  string $url The remote URL to fetch.
 * @return string      The contents of the page returned from $url. Note
 *                     that this may NOT be HTML if the requests was to a
 *                     JSON api service.
 */
function fetchHtml($url) {
	$c = curl_init();
	curl_setopt($c, CURLOPT_URL, $url);
	curl_setopt($c, CURLOPT_RETURNTRANSFER, 1);
	curl_setopt($c, CURLOPT_CONNECTTIMEOUT, 5);
	$data = curl_exec($c);
	curl_close($c);

	return $data;
}

/**
 * Processes the returned json string and extracts the bits relevant to us.
 *
 * This function is specific to the data returned from thebump.com's api
 * service.
 *
 * @param  string $json The raw JSON string to decode and process.
 * @return array        An array containing the relevant parts of the JSON
 *                      payload, including the title, image, source URL and
 *                      main body html.
 */
function extractBlock($json) {
	$json = json_decode($json);
	$card = $json->cards[0];

	$data = array(
		'title' => $card->title,
		'summary' => $card->dek,
		'body' => $card->content->body,
		'img_url' => $card->media->url,
		'source_url' => $card->content_url,
	);

	return $data;
}

/**
 * Converts a processed entry into a formatted HTML string.
 *
 * Expects as input the result of calling extractBlock(). As usual, you
 * should always be wary of outputing data obtained from an untrusted
 * third party on your own domain.
 *
 * @param  array  $data    The array returned from extractBlock() for a
 *                         given json payload.
 * @param  int    $weekNum The integer week number. Used to hint the
 *                         image's alt tag.
 * @return string          A formatted HTML <div> suitable for displaying
 *                         on a page.
 */
function formatBlock($data, $weekNum) {
	$format = <<<EOD
	<div class="week-by-week">
		<img style="float:right;" src="{$data['img_url']}" alt="Week {$weekNum}" />
		<h2><a href="{$data['source_url']}">{$data['title']}</a></h2>
		{$data['summary']}
		{$data['body']}
	</div>
EOD;

	return $format;
}