diff --git a/index.html b/index.html index 36d20f8..17bc439 100644 --- a/index.html +++ b/index.html @@ -17,7 +17,7 @@
Mediapart
- - 10 derniers articles en version texte
+ - Epub (10 derniers)
Les jours
Reporterre
Basta!
diff --git a/mediapart.php b/mediapart.php
index 5888692..c64a638 100644
--- a/mediapart.php
+++ b/mediapart.php
@@ -1,6 +1,7 @@
[
@@ -10,6 +11,11 @@ $opts = [
];
$context = stream_context_create($opts);
+$epub = new TPEpubCreator();
+$epub->temp_folder = 'temp/';
+$epub->epub_file = 'mediapart.epub';
+$epub->title = 'Mediapart';
+
// load feeds
$feed = file_get_contents($feedurl);
$xml = new SimpleXMLElement($feed);
@@ -18,65 +24,66 @@ $items = $xml->xpath("/rss/channel/item");
setlocale(LC_ALL, 'fr_FR.UTF8', 'fr_FR','fr','fr','fra','fr_FR@euro');
$localedate = strftime("%A %d %B %Y");
-$header = "Mediapart - " . $localedate . "\n\nSommaire:\n";
-$result = '';
-
foreach ($items as $item)
{
$title = $item->title;
- //$category = $item->category;
+
$category = $item->xpath('dc:subject')[0];
$author = $item->xpath('dc:creator')[0];
$summary = $item->description;
-
- // Add to content
- $header .= '- ' . $title . "\n";
-
+
$article = file_get_contents($item->link, false, $context);
$doc = new DOMDocument();
$doc->loadHTML($article);
-
$finder = new DomXPath($doc);
-
+
// clean images
$figures = $finder->query('//figure');
foreach ($figures as $figure)
{
$figure->parentNode->removeChild($figure);
}
-
+
// clean "à lire aussi" and inline quotes
$asides = $finder->query('//aside');
foreach ($asides as $aside)
{
$aside->parentNode->removeChild($aside);
}
-
+
// improve section titles
$asides = $finder->query('//h2');
foreach ($asides as $aside)
{
$aside->textContent = "\n*" . $aside->textContent . "*\n";
+ // todo keep h2
}
-
- // Output
- $result .= "\n------\n\n";
- $result .= "Article : " . $item->title . "\n";
- $result .= "Date : " . $item->pubDate . "\n";
- $result .= 'Catégorie : ' . str_replace("\n", '', $category) . "\n";
- $result .= 'Auteur.ice : ' . $author . "\n";
- $result .= "Introduction : " . $summary . "\n";
- $result .= "\n";
- // article main text
$maintext = $finder->query('//div[contains(@class, "paywall-restricted-content")]');
- $result .= $maintext->item(0)->textContent;
+ $result = '
' . $author . '
'; + $result .= '' . $item->pubDate . '
'; + $result .= '' . $summary . '
'; + + // todo use html instead + $result .= '