From: Andrew Dolgov <noreply@fakecake.org>
Date: Tue, 16 Jun 2015 05:17:49 +0000 (+0300)
Subject: add workaround for mysql only accepting subset of unicode characters
X-Git-Tag: 16.3~307
X-Git-Url: https://git.wh0rd.org/?a=commitdiff_plain;h=35c12dc40a4efb2b3f1ca8a8f822819ce1d34b8a;p=tt-rss.git

add workaround for mysql only accepting subset of unicode characters
---

diff --git a/include/rssfuncs.php b/include/rssfuncs.php
index cc8a8e3f..17233914 100644
--- a/include/rssfuncs.php
+++ b/include/rssfuncs.php
@@ -736,6 +736,13 @@
 
 				_debug("plugin data: $entry_plugin_data", $debug_enabled);
 
+				// Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
+				if (DB_TYPE == "mysql") {
+					foreach ($article as $k => $v) {
+						$article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v);
+					}
+				}
+
 				$entry_tags = $article["tags"];
 				$entry_guid = db_escape_string($entry_guid);
 				$entry_title = db_escape_string($article["title"]);