Merge pull request 'Update for FW switching to sending Binary instead of raw html' (#3) from Fix-fw-binary into main

Reviewed-on: #3

.gitea/workflows/release.yml

@@ -0,0 +1,38 @@
+name: Create Release
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          username: ${{ secrets.CI_USER }}
+          password: ${{ secrets.CI_TOKEN }}
+
+      - name: Get version
+        id: get_version
+        run: |
+          if [ "${{ github.ref_type }}" = "tag" ]; then
+            echo "version=${GITHUB_REF#refs/tags/}" >> $GITHUB_OUTPUT
+          else
+            echo "version=$(date +'%Y.%m.%d-%H%M')" >> $GITHUB_OUTPUT
+          fi
+
+      - name: Create ZIP archive
+        run: |
+          zip -r fourthwall-store-embed.zip . -x ".git/*" ".gitea/*"
+
+      - name: Create Release
+        uses: softprops/action-gh-release@v2
+        with:
+          token: "${{ secrets.REPO_TOKEN }}"
+          title: Fourthwall-store-embed Release ${{ steps.get_version.outputs.version }}
+          tag_name: ${{ steps.get_version.outputs.version }}
+          files: |
+            fourthwall-store-embed.zip

libs/shortcode.php

@@ -1,12 +1,53 @@
 <?php
 
 function fwembed_parse_html($url = null) {
-  if ($url === null) {
+    if ($url === null) {
     throw new ValueError("Missing URL");
   }
+
+
+  $ch = curl_init();
+
+  // More complete browser-like headers
+  $headers = [
+    'Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
+    'Accept-Language: en-US,en;q=0.5',
+    'Connection: keep-alive',
+    'Upgrade-Insecure-Requests: 1',
+    'Cache-Control: max-age=0'
+  ];
+
+  curl_setopt($ch, CURLOPT_URL, $url);
+  curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+  curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
+  curl_setopt($ch, CURLOPT_ENCODING, "");
+  curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
+  curl_setopt($ch, CURLOPT_USERAGENT, 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/115.0');
+  curl_setopt($ch, CURLOPT_TIMEOUT, 30);
+  curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false); // Only if necessary for testing
+  curl_setopt($ch, CURLOPT_COOKIEJAR, '/tmp/cookies.txt'); // Store cookies
+  curl_setopt($ch, CURLOPT_COOKIEFILE, '/tmp/cookies.txt'); // Use cookies
+
+  $html_content = curl_exec($ch);
+
+  if (curl_errno($ch)) {
+    $error = curl_error($ch);
+    curl_close($ch);
+    return "Error fetching URL: " . $error;
+  }
+
+  $status_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
+  if ($status_code == 403) {
+    curl_close($ch);
+    return "Access forbidden (403). The website may be blocking automated requests.";
+  }
+
+  curl_close($ch);
+
   $html = null;
+  libxml_use_internal_errors(true);
   $dom = new DOMDocument();
-  @$dom->loadHTML(file_get_contents($url));
+  @$dom->loadHTML(loadHTML5($html_content), LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);
   $dom->documentURI = $url;
   $divs = $dom->getElementsByTagName('div');
   foreach ($divs as $div) {
@@ -43,9 +84,14 @@ function fwembed_parse_html($url = null) {
       $html = $html . '<div class="product-tile"><a class="product-link" target="_blank" href="' . $url . $linkHref . '">' . $productHTML . '</a></div>';
     }
   }
+  libxml_clear_errors();
   return $html;
 }
 
+function loadHTML5($html) {
+  return '<!DOCTYPE html><html><body>' . $html . '</body></html>';
+}
+
 function fwembed_shortcode( $atts ) {
   $options = get_option( 'fourthwall_settings_name' );
   $value = isset( $options['fourth_url'] ) ? $options['fourth_url'] : 'https://fourthwall.com';
@@ -53,4 +99,4 @@ function fwembed_shortcode( $atts ) {
   $store_render = '<div class="fw-store-parent">' . PHP_EOL . $store_html . PHP_EOL . '</div>';
   return $store_render;
 }
-add_shortcode( 'fourthwall', 'fwembed_shortcode' );
+add_shortcode( 'fourthwall', 'fwembed_shortcode' );