presse/epub.php

757 lines
20 KiB
PHP
Raw Permalink Normal View History

2024-10-01 13:46:38 +02:00
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=0">
<title>Revue de presse - Epub</title>
<link rel="icon" type="image/png" href="1F4F0_color.png" /></head>
<body>
<div>
Quick link: <a href="epub.php?mp">All Mediapart home page with renewed authent</a>
</div>
<?php
require 'settings.php';
require 'TPEpubCreator.php';
function get_clean_html($node)
{
// todo check words with "-" (like "sous-traitance")
$innerHTM = '';
foreach ($node->childNodes as $childNode){
$nodeHTML = $childNode->ownerDocument->saveHTML($childNode);
$nodeHTML = str_replace('<aside', '<b><i', $nodeHTML);
$nodeHTML = str_replace('</aside', '</i></b', $nodeHTML);
$nodeHTML = str_replace('<blockquote', '<b><i', $nodeHTML);
$nodeHTML = str_replace('</blockquote', '</i></b', $nodeHTML);
$innerHTML .= $nodeHTML;
}
return '<div>' . strip_tags($innerHTML, '<p><b><h2><i><strong>') . '</div>';
}
function add_mp_article($url, $context, $epub)
{
if (isset($_POST['mptoday']) && $_POST['mptoday'] && !str_contains($url, (new DateTime('today'))->format('dmy')))
{
return;
}
// todo filter on "A la une" ? in rss item dc:subject contains "A la Une"
$article = file_get_contents($url, false, $context);
$doc = new DOMDocument();
$doc->loadHTML($article);
$finder = new DomXPath($doc);
$title = $finder->query('//h1')->item(0)->textContent;
$author = $finder->query('//a[contains(@href, "biographie")]')->item(0)->textContent;
$date = trim($finder->query('//time')->item(1)->textContent);
$summary = $finder->query('//p[contains(@class, "news__heading__top__intro")]')->item(0)->textContent;
$category = trim($finder->query('//p[contains(@class,"news__heading__top__kicker")]')->item(0)->textContent);
if (trim($author) == 'Agence France-Presse')
{
$title = 'AFP - ' . $title;
}
// strip images
$toremove = $finder->query('//svg');
foreach ($toremove as $elt)
{
$elt->parentNode->removeChild($elt);
}
$toremove = $finder->query('//figure');
foreach ($toremove as $elt)
{
$elt->parentNode->removeChild($elt);
}
$toremove = $finder->query('//span[@class="screen-reader-only"]');
foreach ($toremove as $elt)
{
$elt->parentNode->removeChild($elt);
}
$result = '<h1>' . $title . '</h1>';
$result .= '<p><em>' . $category . '</em></p>';
$result .= '<p>' . $author . '</p>';
$result .= '<p>' . $date . '</p>';
$result .= '<p><b>' . $summary . '</b></p>';
$nodes = $finder->query('//div[contains(@class, "paywall-restricted-content")]');
if (!$nodes->length)
{
// articles accès libre
$nodes = $finder->query('//div[contains(@class, "news__body__center__article")]');
}
if (!$nodes->length)
{
echo 'warning: could not get content of "' . $title . '"<br>';
}
else
{
$node = $nodes->item(0);
$nodehtml = get_clean_html($node);
$result .= $nodehtml;
// add boite noire
$nodes = $finder->query('//aside[contains(@class, "_black")]');
if ($nodes->length)
{
$node = $nodes->item(0);
$nodehtml = get_clean_html($node);
$result .= '<div>' . $nodehtml . '</div>';
}
$epub->AddPage($result, false, $title);
}
}
function get_mp_cookie()
{
global $mp_user;
global $mp_password;
$url = 'https://www.mediapart.fr/login_check';
$postfields = 'email=' . $mp_user . '&password=' . $mp_password . '&submitButton=&_target_path=https%3A%2F%2Fwww.mediapart.fr%2F';
$ch = curl_init($url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($ch, CURLOPT_AUTOREFERER, false);
curl_setopt($ch, CURLOPT_HEADER, 1);
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $postfields);
$result = curl_exec($ch);
$sessid = '';
preg_match_all('/^Set-Cookie:\s*([^;]*)/mi', $result, $matches);
foreach($matches[1] as $item)
{
if (str_starts_with($item, 'MPSESSID='))
{
$sessid = str_replace('MPSESSID=', '', $item);
}
}
return $sessid;
}
function write_epub($epub)
{
if (file_exists($epub->epub_file))
{
unlink($epub->epub_file);
}
if (!$epub->error)
{
$epub->CreateEPUB();
if (!$epub->error)
{
echo 'Success: <a href="' . $epub->epub_file . '">' . $epub->epub_file . '</a> created.<br>';
}
else
{
echo $epub->error;
}
}
else
{
echo $epub->error;
}
}
function lm_download_image($base_url, $url_part, $id)
{
$image_url = preg_replace('/GetPublicationContentItems-.*\.json/', $url_part . $id . '.jpg', $base_url);
$temp_img_path = './temp/' . $id . '.jpg';
$temp_content = file_get_contents($image_url);
file_put_contents($temp_img_path, $temp_content);
return $temp_img_path;
}
function get_json($url)
{
// cache json in case url expires
$tempjsonpath = 'temp/' . hash('md5', $url) . '.json';
$json = '';
if (file_exists($tempjsonpath))
{
$json = file_get_contents($tempjsonpath);
}
else
{
$json = file_get_contents($url);
file_put_contents($tempjsonpath, $json);
}
return json_decode($json);
}
// quick scenario
if (isset($_GET['mp']))
{
$_POST['forcempcookies'] = 1;
$_POST['mphome'] = 1;
}
date_default_timezone_set('Europe/Paris');
$today = (new DateTime('today'))->format('Ymd');
if (isset($_POST['emptycache']) && $_POST['emptycache'])
{
$files = glob('temp/*');
foreach($files as $file)
{
if(is_file($file))
{
unlink($file);
}
}
}
// Cafeyn
if (isset($_POST['cafeyn']) && $_POST['cafeyn'])
{
$url = $_POST['cafeyn'];
// extract url from curl command
if (str_starts_with($url, 'curl '))
{
$url = explode("'", $url)[1];
echo '<p>extracted url from curl command:</p><p>' . $url . '</p>';
}
// cache json in case url expires
$publication = get_json($url);
$pubname = $publication->metadata->title;
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/' . $pubname . 'cafeyn.epub';
$epub->title = $pubname . ' Cafeyn - ' . $today ;
$list = (array)$publication->articles;
$page = 1;
foreach ($list as $entry)
{
$articleurl = str_replace('material', 'article/' . $entry->hash . '.json', $url);
$article = get_json($articleurl);
$content = '<h1>' . $article->title . '</h1>';
//$content .= '<b>' . $article->abstract . '</b>';
$content .= '<div>Page ' . $article->page . '</div>';
if (count($article->rubrics))
{
$content .= '<div>Rubriques: ' . implode(',', $article->rubrics) . '</div>';
}
if (count($article->authors))
{
$content .= '<div>Auteurs: ' . implode(',', $article->authors) . '</div>';
}
foreach ($article->content->sections as $section)
{
$dropcap = '';
foreach ($section->items as $item)
{
if ($item->type == 'text')
{
if ($item->class == 'quote')
{
$content .= '<blockquote>' .$item->content . '</blockquote>';
}
else if ($item->class == 'paragraphTitle')
{
$content .= '<h2>' .$item->content . '</h2>';
}
else if ($item->class == 'introduction')
{
$content .= '<p><b>' .$item->content . '</b></p>';
}
else if ($item->class == 'dropcap')
{
$dropcap = $item->content;
}
else
{
if ($dropcap)
{
$item->content = '<b>' . $dropcap . '</b>' . $item->content;
$dropcap = '';
}
$content .= '<p>' . $item->content . '</p>';
}
}
}
}
while ($article->page >= $page)
{
$path = $publication->pages[$page - 1]->tn;
$path = str_replace('material', $path, $url);
$path = str_replace('api.', 'content.', $path);
if ($page == 1)
{
// cover
$epub->AddImage($path, 'image/jpeg', true);
}
else
{
$epub->AddPage('<img style="width: 100%" src="' . $path . '">', false, 'Page ' . $page, true);
}
$page++;
}
$epub->AddPage($content, false, $article->title);
}
write_epub($epub);
}
// Le Monde
if (isset($_POST['lemonde']) && $_POST['lemonde'])
{
$url = $_POST['lemonde'];
// extract url from curl command
if (str_starts_with($url, 'curl '))
{
$url = explode("'", $url)[1];
echo '<p>extracted url from curl command:</p><p>' . $url . '</p>';
}
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/lemonde.epub';
$epub->title = 'Le Monde ' . $today;
$publication = get_json($url);
$content = array_filter($publication->Content, function($item) { return $item->Category == 'Le Monde'; });
usort($content, function ($a, $b) { return $a->PageNumber - $b->PageNumber; });
$pageindex = get_json(str_replace('GetPublicationContentItems', 'GetPublicationPages', $url));
$page = 0;
foreach ($content as $article)
{
if ($article->PageNumber > $page)
{
$page = $article->PageNumber;
$pageobj = array_values(array_filter($pageindex->Page, function($p)
{
global $page;
return $page == $p->PageNumber;
}))[0];
$pageid = $pageobj->PublicationPageID;
$path = lm_download_image($url, 'Preview-MEDIUM-', $pageid);
if ($page == 1)
{
// cover
$epub->AddImage($path, 'image/jpeg', true);
}
else
{
$epub->AddPage('<img style="width: 100%" src="' . $path . '">', false, 'Page ' . $page, true);
}
}
$articlebody = array_filter($article->ContentItem, function($item) { return $item->ContentType == 'text/xml'; });
$articlebody = array_values($articlebody)[0];
if ($articlebody->Title && $articlebody->HtmlText)
{
$pagecontent = '<h1>'. strip_tags($articlebody->Title) . '</h1>';
$pagecontent .= '<p>Page ' . $article->PageNumber . '</p>';
$author = array_filter($article->ContentItem, function($item) { return $item->ContentType == 'author/xml'; });
$author = array_values($author)[0];
if ($author->Author)
{
$pagecontent .= $author->Author;
}
if ($articlebody->Introduction)
{
$pagecontent .= '<b>' . $articlebody->Introduction . '</b>';
}
$pagecontent .= $articlebody->HtmlText;
$epub->AddPage($pagecontent, false, str_replace('&nbsp;', ' ', strip_tags($articlebody->Title)), true);
}
}
write_epub($epub);
}
if (isset($_POST['forcempcookies']) && $_POST['forcempcookies'])
{
$old = $mp_sessionid;
$mp_sessionid = get_mp_cookie();
$settings = file_get_contents('settings.php');
$settings = str_replace($old, $mp_sessionid, $settings);
file_put_contents('settings.php', $settings);
echo '<div>Updated mpsessid in settings.</div>';
}
// Mediapart RSS
if (isset($_POST['mediapart']) && $_POST['mediapart'])
{
$feedurl = 'https://www.mediapart.fr/articles/feed';
$opts = [
'http' => [
'method' => "GET",
'header' => "Accept-language: en\nCookie: MPSESSID=" . $mp_sessionid,
]
];
$context = stream_context_create($opts);
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/mediapart.epub';
$epub->title = 'Mediapart - RSS ' . $today;
// load feeds
$feed = file_get_contents($feedurl);
$xml = new SimpleXMLElement($feed);
$items = $xml->xpath("/rss/channel/item");
foreach (array_reverse($items) as $item)
{
add_mp_article($item->link, $context, $epub);
}
write_epub($epub);
}
// Politis RSS
if (isset($_POST['politis']) && $_POST['politis'])
{
$issuenb = $_POST['politis'];
$feedurl = 'https://www.politis.fr/flux-rss-politis-fr/';
$opts = [
'http' => [
'method' => "GET",
//'header' => "Accept-language: en\nCookie: MPSESSID=" . $mp_sessionid, // todo?
]
];
$context = stream_context_create($opts);
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/politis' . $issuenb . '.epub';
$epub->title = 'Politis ' . $issuenb . ' - RSS ' . $today;
// find cover
$issues = file_get_contents('https://www.politis.fr/boutique/magazines-unite/');
$doc = new DOMDocument();
$doc->loadHTML($issues);
$finder = new DomXPath($doc);
$cover = $finder->query('//img[contains(@src, "' . $issuenb . '_")]');
if ($cover->length)
{
$epub->AddImage($cover->item(0)->getAttribute('src'), 'image/jpeg', true);
}
else
{
echo 'warning: no cover for issue ' . $issuenb . '<br>';
}
// load feeds
$feed = file_get_contents($feedurl);
$xml = new SimpleXMLElement($feed);
$items = $xml->xpath("/rss/channel/item");
foreach (array_reverse($items) as $item)
{
$url = $item->link;
$article = file_get_contents($url, false, $context);
if (!str_contains($article, ' dans lhebdo N° ' . $issuenb))
{
continue;
}
$doc = new DOMDocument();
$doc->loadHTML($article);
$finder = new DomXPath($doc);
$title = $finder->query('//h1')->item(0)->textContent;
$author = $finder->query('//span[@class="nom"]')->item(0)->textContent;
$date = $finder->query('//span[@class="date"]')->item(0)->textContent;
$summary = $finder->query('//div[@class="extrait"]')->item(0)->textContent;
// transform "zoom" divs into aside tags
$zooms = $finder->query('//div[contains(@class,"gt_zoom")]');
foreach ($zooms as $zoom)
{
$aside = $doc->createElement('aside');
while ($zoom->childNodes->length > 0)
{
$aside->appendChild($zoom->childNodes->item(0));
}
$zoom->parentNode->replaceChild($aside, $zoom);
}
$result = '<h1>' . $title . '</h1>';
$result .= '<p>' . $author . '</p>';
$result .= '<p>' . $date . '</p>';
$result .= '<p><b>' . $summary . '</b></p>';
$nodes = $finder->query('//div[@id="content"]');
$node = $nodes->item(0);
$nodehtml = get_clean_html($node);
$result .= $nodehtml;
$epub->AddPage($result, false, $title);
}
write_epub($epub);
}
// Mediapart home page
if (isset($_POST['mphome']) && $_POST['mphome'])
{
$homeurl = 'https://www.mediapart.fr';
$opts = [
'http' => [
'method' => "GET",
'header' => "Accept-language: en\nCookie: MPSESSID=" . $mp_sessionid,
]
];
$context = stream_context_create($opts);
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/mediaparthome.epub';
$epub->title = 'Mediapart - A la une ' . $today;
$home = file_get_contents($homeurl, false, $context);
$doc = new DOMDocument();
$doc->loadHTML($home);
$finder = new DomXPath($doc);
$xpath = '//a[@data-js = "teaser-link" and contains(@href, "/journal/") and not(contains(@href, "/dossier/")) and not(contains(@href, "/studio/")) and not(contains(@href, "fil-")) and (@href != "/journal/series")]/@href';
$links = $finder->query($xpath);
foreach ($links as $link)
{
add_mp_article('https://www.mediapart.fr' . $link->value, $context, $epub);
}
write_epub($epub);
}
// New York Times RSS
if (isset($_POST['nyt']) && $_POST['nyt'])
{
$feedurl = 'https://rss.nytimes.com/services/xml/rss/nyt/HomePage.xml';
$opts = [
'http' => [
'method' => "GET",
'header' => "Accept-language: en\nCookie: NYT-S=" . $nyt_sessionid,
]
];
$context = stream_context_create($opts);
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/newyorktimes.epub';
$epub->title = 'The New York Times ' . $today;
// load feeds
$feed = file_get_contents($feedurl);
$xml = new SimpleXMLElement($feed);
$items = $xml->xpath("/rss/channel/item");
foreach (array_reverse($items) as $item)
{
$title = $item->title;
$author = $item->xpath('dc:creator')[0];
$summary = $item->description;
$article = file_get_contents($item->link, false, $context);
$doc = new DOMDocument();
$doc->loadHTML($article);
$finder = new DomXPath($doc);
$result = '<h1>' . $title . '</h1>';
$result .= '<p>' . $author . '</p>';
$result .= '<p>' . $item->pubDate . '</p>';
$result .= '<p><b>' . $summary . '</b></p>';
$nodes = $finder->query('//section[@name="articleBody"]');
if (!$nodes->length)
{
echo 'warning: could not get content of "' . $title . '"<br>';
}
else
{
$node = $nodes->item(0);
$nodehtml = get_clean_html($node);
$result .= $nodehtml;
$epub->AddPage($result, false, $title);
}
}
write_epub($epub);
}
// New York Times today's paper
if (isset($_POST['nyttoday']) && $_POST['nyttoday'])
{
$todaypage = file_get_contents("https://www.nytimes.com/section/todayspaper");
$opts = [
'http' => [
'method' => "GET",
'header' => "Accept-language: en\nCookie: NYT-S=" . $nyt_sessionid,
]
];
$context = stream_context_create($opts);
$epub = new TPEpubCreator();
$epub->temp_folder = 'temp/';
$epub->epub_file = 'epub/newyorktimestoday.epub';
$epub->title = 'The New York Times today\'s paper ' . $today;
// grab today's paper
$start = strpos($todaypage, "window.__preloadedData = ") + 25;
$end = strpos($todaypage, "};", $start) + 1;
$json = str_replace(":undefined", ":null", substr($todaypage, $start, $end - $start));
$data = json_decode($json);
foreach($data->initialState as $item)
{
if ($item->__typename == "Article")
{
$headline = ($data->initialState->{ $item->headline->id });
$title = $headline->default;
$summary = $item->summary;
$result = '<h1>' . $title . '</h1>';
// authors
$result .= '<p>';
$creators = ($data->initialState->{ $item->bylines[0]->id })->creators;
foreach ($creators as $creator)
{
$author = $data->initialState->{ $creator->id };
$result .= $author->displayName . '&nbsp;';
}
$result .= '</p>';
$result .= '<p>' . $item->lastMajorModification . '</p>';
$result .= '<p><b>' . $summary . '</b></p>';
$article = file_get_contents($item->url, false, $context);
$doc = new DOMDocument();
$doc->loadHTML($article);
$finder = new DomXPath($doc);
$nodes = $finder->query('//section[@name="articleBody"]');
if (!$nodes->length)
{
echo 'warning: could not get content of "' . $title . '"<br>';
}
else
{
$node = $nodes->item(0);
$nodehtml = get_clean_html($node);
$result .= $nodehtml;
$epub->AddPage($result, false, $title);
}
}
}
write_epub($epub);
}
?>
<h1>Generate epub</h1>
<div>
<form method="post">
<input type="submit">
<h2>Mediapart</h2>
<input id="mediapart" name="mediapart" type="checkbox">
<label for="mediapart">From RSS</label>
<br>
<input id="mphome" name="mphome" type="checkbox">
<label for="mphome">From home page</label>
<br>
<input id="forcempcookies" name="forcempcookies" type="checkbox">
<label for="forcempcookies">Force cookies</label>
<br>
<input id="mptoday" name="mptoday" type="checkbox">
<label for="mptoday">Today only</label>
<h2>Le Monde</h2>
<label for="lemonde">"GetPublicationContentItems" json url: </label>
<input id="lemonde" name="lemonde">
<h2>Cafeyn</h2>
<label for="cafeyn">"material" json url: </label>
<input id="cafeyn" name="cafeyn">
<h2>The New York Times</h2>
<input id="nyt" name="nyt" type="checkbox">
<label for="nyt">From RSS</label>
<br>
<input id="nyttoday" name="nyttoday" type="checkbox">
<label for="nyttoday">From today's paper embedded script</label>
<h2>Politis</h2>
<label for="politis">Issue number: </label><input id="politis" name="politis">
<h2>Admin</h2>
<input id="emptycache" name="emptycache" type="checkbox" checked>
<label for="emptycache">Empty cache</label>
<br><br>
<input type="submit">
</form>
</div>
<h1>Available files</h1>
<?php
if ($handle = opendir('epub')) {
echo "<ul>";
while (false !== ($file = readdir($handle))) {
if ($file != "." && $file != "..") {
$filePath = 'epub/' . $file;
$modTime = date("F d Y H:i:s.", filemtime($filePath));
echo "<li><a href='$filePath' download>$file</a> - $modTime</li>";
}
}
echo "</ul>";
closedir($handle);
}
?>
<h1>Mediapart RSS summary</h1>
<?php
$mprss = file_get_contents('https://www.mediapart.fr/articles/feed', false, $context);
$xml = new SimpleXMLElement($mprss);
$items = $xml->xpath("/rss/channel/item");
foreach ($items as $item)
{
echo '<div><a href="'.$item->link.'">'.$item->pubDate. ' '.$item->title.'</a></div>';
}
?>
<h1>Links</h1>
<a href="../epub">Generated files</a><br>
<a href="../">Press review</a><br>
</body>
2024-05-27 13:28:36 +02:00
</html>