(.*?)<\/a> <\/h2>/', '$2
', $pageContents); // Change formatting of article titles to fit on one line and all be an href so that you just have to click on link once $pageContents = preg_replace('/[[:space:]]*(.*?)<\/em><\/span>

(.*?)<\/a> <\/h2>/', '$1: $3', $pageContents); // handle letters section $pageContents = preg_replace('/

(.*?)<\/a> <\/h2>/', '$2', $pageContents); // change h1s into h2s $pageContents = preg_replace('/

(.*?)<\/h1>/', '

$1

', $pageContents); // // write TOC file // $tocFile = $dateDirectory . '/mbp_toc.html'; $fh = fopen($tocFile, 'w'); fwrite($fh, '

The Economist: ' . $dateLong . '

'); fwrite($fh, $pageContents); fwrite($fh, ''); fclose($fh); echo " done!\n"; } /** * Creates an issue's HTML * * @param string $dateDirectory Issue's directory * @param bool $withImages Gather images as well * @param array $urls Article URLs * @param array $ids Article IDs * * @return void * */ function createHTML($dateDirectory, $withImages, $urls, $ids) { $indexFile = $dateDirectory . '/economist.html'; $indexFileH = fopen($indexFile, 'w'); fwrite($indexFileH, ''); // write all urls echo "Creating articles file: economist.html:\n"; for ($i = 0; $i < count($urls); $i++) { $url = (substr($urls[$i], 0, 5) === 'http:') ? $urls[$i] : 'http://www.economist.com/' . $urls[$i]; echo "\t$url\n"; // download article $article = economistGetUrl($url); if ($article === '') { echo "\t\tERROR in downloading!\n"; } $startPos = strpos($article, '

instead if ($startPos === false) { $startPos = strpos($article, ''); $endPos = getArticleEnd($article); $content = substr($article, $startPos, ($endPos - $startPos)); // now that we have got rid of that other stuff, there is only one h1 and it is the new beginning $content = strstr($content, '

'); // some articles lead with a single h1 of Economist header if (substr($content, 0, 16) === '

') { // find the second h1 $content = strstr(substr($content, 16), '

'); } // add TOC link to article $content = ereg_replace('

', '

', $content); } else { $endPos = getArticleEnd($article); $content = substr($article, $startPos, ($endPos - $startPos)); // reformat title and add TOC link to article // turn this:

Obamanomics

Hope and fear

// into this:

Obamanomics: Hope and fear

$content = ereg_replace('

(.*)

', ': ', $content); $content = ereg_replace('

([[:space:]])*', '

', $content); } // get rid of this:

Feb 28th 2008
From The Economist print edition

$content = preg_replace('/

(.*?)<\/p>/', '', $content); // get rid of banner ads $content = preg_replace('/

([[:space:]])*

(.*?)<\/div>([[:space:]])*<\/div>/s', '', $content); // allow inside-links $content = preg_replace('/', '
', $content); // convert title from h1 to h2 $content = ereg_replace('h1', 'h2', $content); // the only
with a space after it is under the subtitle, give an extra line $content = ereg_replace('
[[:space:]]', '

 

', $content); if ($withImages) { $content = handleImages($content, $dateDirectory); } if ($content === '') { echo "\t\tERROR in processing!\n"; } fwrite($indexFileH, "\n\n"); fwrite($indexFileH, $content); //add page break to bottom of each article fwrite($indexFileH, "\n"); } fwrite($indexFileH, ''); fclose($indexFileH); } /** * Grabs an article's images * * @param string $content Article's contents * @param string $dateDirectory Issue's directory * * @return string Content with images * */ function handleImages($content, $dateDirectory) { preg_match_all('/(.*?)<\/div>/', '

$2', $content); //change format of caption to be italics $content = preg_replace('/(.*?)<\/span>/', '
$1', $content); return $content; } /** * Determines an article's end * * @param string $article Article's HTML * * @return int Article's last position (character count) * */ function getArticleEnd($article) { $ends = array(); $end1 = strpos($article, '

'); if ($end1) { $ends[] = $end1; } $end2 = strpos($article, '

'); if ($end2) { $ends[] = $end2; } $end3 = strpos($article, '

'); if ($end3) { $ends[] = $end3; } $end4 = strpos($article, '