From 4cf3564d28dc8e4d08a3e64f09ad045ffbde97ae Mon Sep 17 00:00:00 2001
From: ArthurHoaro <arthur@hoa.ro>
Date: Fri, 25 Sep 2020 13:29:36 +0200
Subject: Add a setting to retrieve bookmark metadata asynchrounously

  - There is a new standalone script (metadata.js) which requests
    a new controller to get bookmark metadata and fill the form async
  - This feature is enabled with the new setting: general.enable_async_metadata
    (enabled by default)
  - general.retrieve_description is now enabled by default
  - A small rotating loader animation has a been added to bookmark inputs
    when metadata is being retrieved (default template)
  - Custom JS htmlentities has been removed and  mathiasbynens/he
    library is used instead

Fixes #1563
---
 application/http/MetadataRetriever.php | 68 ++++++++++++++++++++++++++++++++++
 1 file changed, 68 insertions(+)
 create mode 100644 application/http/MetadataRetriever.php

(limited to 'application/http/MetadataRetriever.php')

diff --git a/application/http/MetadataRetriever.php b/application/http/MetadataRetriever.php
new file mode 100644
index 00000000..2ca982e2
--- /dev/null
+++ b/application/http/MetadataRetriever.php
@@ -0,0 +1,68 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Shaarli\Http;
+
+use Shaarli\Config\ConfigManager;
+
+/**
+ * HTTP Tool used to extract metadata from external URL (title, description, etc.).
+ */
+class MetadataRetriever
+{
+    /** @var ConfigManager */
+    protected $conf;
+
+    /** @var HttpAccess */
+    protected $httpAccess;
+
+    public function __construct(ConfigManager $conf, HttpAccess $httpAccess)
+    {
+        $this->conf = $conf;
+        $this->httpAccess = $httpAccess;
+    }
+
+    /**
+     * Retrieve metadata for given URL.
+     *
+     * @return array [
+     *                  'title' => <remote title>,
+     *                  'description' => <remote description>,
+     *                  'tags' => <remote keywords>,
+     *               ]
+     */
+    public function retrieve(string $url): array
+    {
+        $charset = null;
+        $title = null;
+        $description = null;
+        $tags = null;
+        $retrieveDescription = $this->conf->get('general.retrieve_description');
+
+        // Short timeout to keep the application responsive
+        // The callback will fill $charset and $title with data from the downloaded page.
+        $this->httpAccess->getHttpResponse(
+            $url,
+            $this->conf->get('general.download_timeout', 30),
+            $this->conf->get('general.download_max_size', 4194304),
+            $this->httpAccess->getCurlDownloadCallback(
+                $charset,
+                $title,
+                $description,
+                $tags,
+                $retrieveDescription
+            )
+        );
+
+        if (!empty($title) && strtolower($charset) !== 'utf-8') {
+            $title = mb_convert_encoding($title, 'utf-8', $charset);
+        }
+
+        return [
+            'title' => $title,
+            'description' => $description,
+            'tags' => $tags,
+        ];
+    }
+}
-- 
cgit v1.2.3


From 5334090be04e66da5cb5c3ad487604b3733c5cac Mon Sep 17 00:00:00 2001
From: ArthurHoaro <arthur@hoa.ro>
Date: Thu, 15 Oct 2020 11:20:33 +0200
Subject: Improve metadata retrieval (performances and accuracy)

  - Use dedicated function to download headers to avoid apply multiple regexps on headers
  - Also try to extract title from meta tags
---
 application/http/MetadataRetriever.php | 1 +
 1 file changed, 1 insertion(+)

(limited to 'application/http/MetadataRetriever.php')

diff --git a/application/http/MetadataRetriever.php b/application/http/MetadataRetriever.php
index 2ca982e2..ba9bd40c 100644
--- a/application/http/MetadataRetriever.php
+++ b/application/http/MetadataRetriever.php
@@ -46,6 +46,7 @@ class MetadataRetriever
             $url,
             $this->conf->get('general.download_timeout', 30),
             $this->conf->get('general.download_max_size', 4194304),
+            $this->httpAccess->getCurlHeaderCallback($charset),
             $this->httpAccess->getCurlDownloadCallback(
                 $charset,
                 $title,
-- 
cgit v1.2.3