446 lines
12 KiB
PHP
446 lines
12 KiB
PHP
<!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>
|
|
<style>
|
|
body {
|
|
font-family: helvetica, sans-serif;
|
|
}
|
|
</style>
|
|
<link rel="icon" type="image/png" href="1F4F0_color.png" /></head>
|
|
<body>
|
|
|
|
<?php
|
|
|
|
require 'settings.php';
|
|
require 'TPEpubCreator.php';
|
|
|
|
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;
|
|
}
|
|
}
|
|
|
|
date_default_timezone_set('Europe/Paris');
|
|
$date = (new DateTime('today'))->format('Ymd');
|
|
|
|
// Le Monde
|
|
if (isset($_POST['lemonde']) && $_POST['lemonde'])
|
|
{
|
|
// parameters hardcoded for now
|
|
$includeimages = true;
|
|
$includepages = true;
|
|
|
|
$url = $_POST['lmurl'];
|
|
|
|
// extract url from curl command
|
|
if (str_starts_with($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 ' . $date ;
|
|
|
|
// 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);
|
|
}
|
|
|
|
$publication = json_decode($json);
|
|
$content = array_filter($publication->Content, function($item) { return $item->Category == 'Le Monde'; });
|
|
usort($content, function ($a, $b) { return $a->PageNumber - $b->PageNumber; });
|
|
|
|
if ($includepages)
|
|
{
|
|
$pageindex = json_decode(file_get_contents(str_replace('GetPublicationContentItems', 'GetPublicationPages', $url)));
|
|
}
|
|
|
|
$page = 0;
|
|
foreach ($content as $article)
|
|
{
|
|
if ($includepages && $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;
|
|
|
|
// SMALL / MEDIUM / XLARGE
|
|
$pageurl = preg_replace('/GetPublicationContentItems-.*\.json/', 'Preview-MEDIUM-' . $pageid . '.jpg', $url);
|
|
$tempimgpath = './temp/' . $pageid . '.jpg';
|
|
$tempcontent = file_get_contents($pageurl);
|
|
file_put_contents($tempimgpath, $tempcontent);
|
|
$epub->AddPage('<img src="' . $tempimgpath . '">', false, 'Page ' . $page, true);
|
|
|
|
// Add page 1 as cover
|
|
if ($page == 1)
|
|
{
|
|
$epub->AddImage($tempimgpath, 'image/jpeg', 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>' . $articlebody->Introduction . '</b>';
|
|
}
|
|
|
|
if ($includeimages)
|
|
{
|
|
$images = array_values(array_filter($article->ContentItem, function($item) { return $item->ContentType == 'graphic/jpeg' || $item->ContentType == 'image/jpeg'; }));
|
|
foreach ($images as $image)
|
|
{
|
|
$imageid = $image->ContentItemId;
|
|
$imageurl = preg_replace('/GetPublicationContentItems-.*\.json/', 'Image-MEDIUM-' . $imageid . '.jpg', $url);
|
|
|
|
// Download image from source, put it in temp and include it into article main html content.
|
|
// Then pass true to last AddPage param to download image to epub.
|
|
$tempimgpath = './temp/' . $imageid . '.jpg';
|
|
$tempcontent = file_get_contents($imageurl);
|
|
file_put_contents($tempimgpath, $tempcontent);
|
|
$pagecontent .= '<p><img src="' . $tempimgpath . '"></p>';
|
|
}
|
|
}
|
|
|
|
$pagecontent .= $articlebody->HtmlText;
|
|
|
|
// debug
|
|
// echo '<pre style="white-space: pre-wrap;">' . $pagecontent . '<br>----------------------</pre>';
|
|
|
|
$epub->AddPage($pagecontent, false, strip_tags($articlebody->Title), true);
|
|
}
|
|
}
|
|
write_epub($epub);
|
|
}
|
|
|
|
// Mediapart
|
|
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 ' . $date;
|
|
|
|
// load feeds
|
|
$feed = file_get_contents($feedurl);
|
|
$xml = new SimpleXMLElement($feed);
|
|
$items = $xml->xpath("/rss/channel/item");
|
|
|
|
foreach ($items as $item)
|
|
{
|
|
$title = $item->title;
|
|
|
|
$category = $item->xpath('dc:subject')[0];
|
|
$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);
|
|
|
|
// 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>' . $author . '</p>';
|
|
$result .= '<p>' . $item->pubDate . '</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);
|
|
|
|
$innerHTML = '';
|
|
foreach ($node->childNodes as $childNode){
|
|
$innerHTML .= $childNode->ownerDocument->saveHTML($childNode);
|
|
}
|
|
$result .= '<div>' . strip_tags($innerHTML, '<p><b><h2><i>') . '</div>';
|
|
|
|
$epub->AddPage($result, false, $title);
|
|
}
|
|
}
|
|
|
|
write_epub($epub);
|
|
}
|
|
|
|
// New York Times
|
|
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 ' . $date;
|
|
|
|
// load feeds
|
|
$feed = file_get_contents($feedurl);
|
|
$xml = new SimpleXMLElement($feed);
|
|
$items = $xml->xpath("/rss/channel/item");
|
|
|
|
foreach ($items as $item)
|
|
{
|
|
$title = $item->title;
|
|
|
|
$category = $item->xpath('dc:subject')[0];
|
|
$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);
|
|
|
|
// 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>' . $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);
|
|
|
|
$innerHTML = '';
|
|
foreach ($node->childNodes as $childNode){
|
|
$innerHTML .= $childNode->ownerDocument->saveHTML($childNode);
|
|
}
|
|
|
|
$result .= '<div>' . strip_tags($innerHTML, '<p><b><h2><i>') . '</div>';
|
|
|
|
$epub->AddPage($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 ' . $date;
|
|
|
|
// 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 . ' ';
|
|
}
|
|
$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);
|
|
|
|
$innerHTML = '';
|
|
foreach ($node->childNodes as $childNode){
|
|
$innerHTML .= $childNode->ownerDocument->saveHTML($childNode);
|
|
}
|
|
|
|
$result .= '<div>' . strip_tags($innerHTML, '<p><b><h2><i>') . '</div>';
|
|
|
|
$epub->AddPage($result, false, $title);
|
|
}
|
|
}
|
|
}
|
|
write_epub($epub);
|
|
}
|
|
?>
|
|
|
|
<h3>Generate epub</h3>
|
|
<div>
|
|
<form method="post">
|
|
|
|
Parameters:
|
|
<br>
|
|
GetPublicationContentItems url: <input name="lmurl"><br>
|
|
<!--Update MPSESSID: <input name="mpsessid"><br>
|
|
Update NYT-S: <input name="nyts"><br>-->
|
|
<br>
|
|
|
|
Newspapers:
|
|
<br>
|
|
<input id="lemonde" name="lemonde" type="checkbox">
|
|
<label for="lemonde">Le Monde</label>
|
|
<br>
|
|
|
|
<input id="mediapart" name="mediapart" type="checkbox">
|
|
<label for="mediapart">Mediapart (RSS)</label>
|
|
<br>
|
|
|
|
<input id="nyt" name="nyt" type="checkbox">
|
|
<label for="nyt">The New York Times (RSS)</label>
|
|
<br>
|
|
|
|
<input id="nyttoday" name="nyttoday" type="checkbox">
|
|
<label for="nyttoday">The New York Times (today's paper)</label>
|
|
<br>
|
|
|
|
<input type="submit">
|
|
</form>
|
|
</div>
|
|
<br>
|
|
<a href="../epub">Fichiers générés</a><br>
|
|
<a href="../">Revue de presse</a>
|
|
</body>
|
|
</html>
|