Refactor CampaignProcessorMessageHandler

composer.json

@@ -79,7 +79,9 @@
         "ext-imap": "*",
         "tatevikgr/rss-feed": "dev-main",
         "ext-pdo": "*",
-        "ezyang/htmlpurifier": "^4.19"
+        "ezyang/htmlpurifier": "^4.19",
+        "ext-libxml": "*",
+        "ext-gd": "*"
     },
     "require-dev": {
         "phpunit/phpunit": "^9.5",

config/parameters.yml.dist

@@ -89,3 +89,12 @@ parameters:
     env(MESSAGING_MAX_PROCESS_TIME): '600'
     messaging.max_mail_size: '%%env(MAX_MAILSIZE)%%'
     env(MAX_MAILSIZE): '209715200'
+    messaging.default_message_age: '%%env(DEFAULT_MESSAGEAGE)%%'
+    env(DEFAULT_MESSAGEAGE): '691200'
+    messaging.use_manual_text_part : '%%env(USE_MANUAL_TEXT_PART)%%'
+    env(USE_MANUAL_TEXT_PART): 0
+
+    phplist.upload_images_dir: '%%env(PHPLIST_UPLOADIMAGES_DIR)%%'
+    env(PHPLIST_UPLOADIMAGES_DIR): 'images'
+    phplist.public_schema: '%%env(PUBLIC_SCHEMA)%%'
+    env(PUBLIC_SCHEMA): 'http'

src/Domain/Common/HtmlToText.php

@@ -0,0 +1,85 @@
+<?php
+
+declare(strict_types=1);
+
+namespace PhpList\Core\Domain\Common;
+
+use PhpList\Core\Domain\Configuration\Model\ConfigOption;
+use PhpList\Core\Domain\Configuration\Service\Provider\ConfigProvider;
+
+class HtmlToText
+{
+    private const WORD_WRAP = 70;
+
+    public function __construct(private readonly ConfigProvider $configProvider)
+    {
+    }
+
+    public function __invoke(string $html): string
+    {
+        $text = preg_replace("/\r/", '', $html);
+
+        $text = preg_replace("/<script[^>]*>(.*?)<\/script\s*>/is", '', $text);
+        $text = preg_replace("/<style[^>]*>(.*?)<\/style\s*>/is", '', $text);
+
+        $text = preg_replace(
+            "/<a[^>]*href=([\"\'])(.*)\\1[^>]*>(.*)<\/a>/Umis",
+            "[URLTEXT]\\3[ENDURLTEXT][LINK]\\2[ENDLINK]\n",
+            $text
+        );
+        $text = preg_replace("/<b>(.*?)<\/b\s*>/is", '*\\1*', $text);
+        $text = preg_replace("/<h[\d]>(.*?)<\/h[\d]\s*>/is", "**\\1**\n", $text);
+        $text = preg_replace("/<i>(.*?)<\/i\s*>/is", '/\\1/', $text);
+        $text = preg_replace("/<\/tr\s*?>/i", "<\/tr>\n\n", $text);
+        $text = preg_replace("/<\/p\s*?>/i", "<\/p>\n\n", $text);
+        $text = preg_replace('/<br[^>]*?>/i', "<br>\n", $text);
+        $text = preg_replace("/<br[^>]*?\/>/i", "<br\/>\n", $text);
+        $text = preg_replace('/<table/i', "\n\n<table", $text);
+        $text = strip_tags($text);
+
+        // find all URLs and replace them back
+        preg_match_all('~\[URLTEXT\](.*)\[ENDURLTEXT\]\[LINK\](.*)\[ENDLINK\]~Umis', $text, $links);
+        foreach ($links[0] as $matchindex => $fullmatch) {
+            $linktext = $links[1][$matchindex];
+            $linkurl = $links[2][$matchindex];
+            // check if the text linked is a repetition of the URL
+            if (trim($linktext) == trim($linkurl) ||
+                'https://'.trim($linktext) == trim($linkurl) ||
+                'http://'.trim($linktext) == trim($linkurl)
+            ) {
+                $linkreplace = $linkurl;
+            } else {
+                //# if link is an anchor only, take it out
+                if (strpos($linkurl, '#') === 0) {
+                    $linkreplace = $linktext;
+                } else {
+                    $linkreplace = $linktext.' <'.$linkurl.'>';
+                }
+            }
+            $text = str_replace($fullmatch, $linkreplace, $text);
+        }
+        $text = preg_replace(
+            "/<a href=[\"\'](.*?)[\"\'][^>]*>(.*?)<\/a>/is",
+            '[URLTEXT]\\2[ENDURLTEXT][LINK]\\1[ENDLINK]',
+            $text,
+            500
+        );
+
+        $text = html_entity_decode($text, ENT_QUOTES | ENT_HTML5, 'UTF-8');
+
+        $text = preg_replace('/###NL###/', "\n", $text);
+        $text = preg_replace("/\n /", "\n", $text);
+        $text = preg_replace("/\t/", ' ', $text);
+
+        // reduce whitespace
+        while (preg_match('/  /', $text)) {
+            $text = preg_replace('/  /', ' ', $text);
+        }
+        while (preg_match("/\n\s*\n\s*\n/", $text)) {
+            $text = preg_replace("/\n\s*\n\s*\n/", "\n\n", $text);
+        }
+        $ww = $this->configProvider->getValue(ConfigOption::WordWrap) ?? self::WORD_WRAP;
+
+        return wordwrap($text, $ww);
+    }
+}

src/Domain/Common/HtmlUrlRewriter.php

@@ -0,0 +1,190 @@
+<?php
+
+declare(strict_types=1);
+
+namespace PhpList\Core\Domain\Common;
+
+use DOMDocument;
+use DOMElement;
+use DOMXPath;
+
+class HtmlUrlRewriter
+{
+    public function addAbsoluteResources(string $html, string $baseUrl): string
+    {
+        $baseUrl = rtrim($baseUrl, "/");
+
+        // 1) Rewrite HTML attributes via DOM (handles quotes, whitespace, etc.)
+        $dom = new DOMDocument();
+        libxml_use_internal_errors(true);
+
+        // Prevent DOMDocument from adding html/body tags if you pass fragments
+        $wrapped = '<!doctype html><meta charset="utf-8"><div id="__wrap__">' . $html . '</div>';
+        $dom->loadHTML($wrapped, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);
+
+        $xpath = new DOMXPath($dom);
+
+        // Attributes to rewrite
+        $attrMap = [
+            '//*[@src]' => 'src',
+            '//*[@href]' => 'href',
+            '//*[@action]' => 'action',
+            '//*[@background]' => 'background',
+        ];
+
+        foreach ($attrMap as $query => $attr) {
+            foreach ($xpath->query($query) as $node) {
+                /** @var DOMElement $node */
+                $val = $node->getAttribute($attr);
+                $node->setAttribute($attr, $this->absolutizeUrl($val, $baseUrl));
+            }
+        }
+
+        // srcset needs special handling (multiple candidates)
+        foreach ($xpath->query('//*[@srcset]') as $node) {
+            /** @var DOMElement $node */
+            $node->setAttribute('srcset', $this->rewriteSrcset($node->getAttribute('srcset'), $baseUrl));
+        }
+
+        // 2) Rewrite inline <style> blocks (CSS)
+        foreach ($xpath->query('//style') as $styleNode) {
+            /** @var DOMElement $styleNode */
+            $css = $styleNode->nodeValue;
+            $styleNode->nodeValue = $this->rewriteCssUrls($css, $baseUrl);
+        }
+
+        // 3) Rewrite style="" attributes (CSS)
+        foreach ($xpath->query('//*[@style]') as $node) {
+            /** @var DOMElement $node */
+            $css = $node->getAttribute('style');
+            $node->setAttribute('style', $this->rewriteCssUrls($css, $baseUrl));
+        }
+
+        // Extract the original fragment back out
+        $wrap = $dom->getElementById('__wrap__');
+        $out = '';
+        foreach ($wrap->childNodes as $child) {
+            $out .= $dom->saveHTML($child);
+        }
+
+        libxml_clear_errors();
+
+        return $out;
+    }
+
+    /**
+     * Convert $url to an absolute URL based on $baseUrl.
+     * Leaves alone:
+     * - already absolute (scheme:)
+     * - protocol-relative (//example.com) => keeps host but adds scheme
+     * - anchors (#...)
+     * - placeholders like [SOMETHING]
+     * - mailto:, tel:, data:, javascript: (etc)
+     */
+    public function absolutizeUrl(string $url, string $baseUrl): string
+    {
+        $url = trim($url);
+        if ($url === '' || $url[0] === '#') return $url;
+        if (preg_match('/\[[^\]]+\]/', $url)) return $url;
+
+        // already has a scheme (http:, https:, mailto:, data:, etc.)
+        if (preg_match('#^[a-z][a-z0-9+.-]*:#i', $url)) return $url;
+
+        $base = parse_url($baseUrl);
+        if (!$base || empty($base['scheme']) || empty($base['host'])) {
+            // If base is invalid, bail out rather than corrupt URLs
+            return $url;
+        }
+
+        // protocol-relative
+        if (str_starts_with($url, '//')) {
+            return $base['scheme'] . ':' . $url;
+        }
+
+        $basePath = $base['path'] ?? '/';
+        // If baseUrl points to a file, use its directory
+        if (!str_ends_with($basePath, '/')) {
+            $basePath = preg_replace('#/[^/]*$#', '/', $basePath);
+        }
+
+        if (str_starts_with($url, '/')) {
+            $path = $url;
+        } else {
+            $path = $basePath . $url;
+        }
+
+        $path = $this->normalizePath($path);
+
+        $port = isset($base['port']) ? ':' . $base['port'] : '';
+        return $base['scheme'] . '://' . $base['host'] . $port . $path;
+    }
+
+    function normalizePath(string $path): string
+    {
+        // Keep query/fragment if present
+        $parts = parse_url($path);
+        $p = $parts['path'] ?? $path;
+
+        $segments = explode('/', $p);
+        $out = [];
+        foreach ($segments as $seg) {
+            if ($seg === '' || $seg === '.') continue;
+            if ($seg === '..') {
+                array_pop($out);
+                continue;
+            }
+            $out[] = $seg;
+        }
+        $norm = '/' . implode('/', $out);
+
+        if (isset($parts['query'])) $norm .= '?' . $parts['query'];
+        if (isset($parts['fragment'])) $norm .= '#' . $parts['fragment'];
+        return $norm;
+    }
+
+    public function rewriteSrcset(string $srcset, string $baseUrl): string
+    {
+        // "a.jpg 1x, /b.jpg 2x" => absolutize each URL part
+        $candidates = array_map('trim', explode(',', $srcset));
+        foreach ($candidates as &$cand) {
+            if ($cand === '') continue;
+            // split at first whitespace: "url descriptor..."
+            if (preg_match('/^(\S+)(\s+.*)?$/', $cand, $m)) {
+                $u = $m[1];
+                $d = $m[2] ?? '';
+                $cand = $this->absolutizeUrl($u, $baseUrl) . $d;
+            }
+        }
+        return implode(', ', $candidates);
+    }
+
+    public function rewriteCssUrls(string $css, string $baseUrl): string
+    {
+        // url(...) handling (supports quotes or no quotes)
+        $css = preg_replace_callback(
+            '#url\(\s*(["\']?)(.*?)\1\s*\)#i',
+            function ($m) use ($baseUrl) {
+                $q = $m[1];
+                $u = $m[2];
+                $abs = $this->absolutizeUrl($u, $baseUrl);
+                return 'url(' . ($q !== '' ? $q : '') . $abs . ($q !== '' ? $q : '') . ')';
+            },
+            $css
+        );
+
+        // @import "..."; or @import url("..."); etc.
+        return preg_replace_callback(
+            '#@import\s+(?:url\()?(\s*["\']?)([^"\')\s;]+)\1\)?#i',
+            function ($m) use ($baseUrl) {
+                $q = trim($m[1]);
+                $u = $m[2];
+                $abs = $this->absolutizeUrl($u, $baseUrl);
+                // Preserve original form loosely
+                return str_starts_with($m[0], '@import url')
+                    ? '@import url(' . ($q ?: '') . $abs . ($q ?: '') . ')'
+                    : '@import ' . ($q ?: '') . $abs . ($q ?: '');
+            },
+            $css
+        );
+    }
+}