From cca2d49cc3877c279a24769712306414b18daa04 Mon Sep 17 00:00:00 2001
From: Athou <jeremiepanzer@gmail.com>
Date: Fri, 13 Sep 2024 23:40:13 +0200
Subject: [PATCH] Revert "reduce artifact size by using a smaller library for
 charset detection" because juniversalchardet doesn't support as many charsets
 as icu4j

---
 commafeed-server/pom.xml                        |  6 +++---
 .../backend/feed/parser/EncodingDetector.java   | 17 +++++++++++------
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/commafeed-server/pom.xml b/commafeed-server/pom.xml
index 5dae894e..915e4b65 100644
--- a/commafeed-server/pom.xml
+++ b/commafeed-server/pom.xml
@@ -433,9 +433,9 @@
 			<version>1.18.1</version>
 		</dependency>
 		<dependency>
-			<groupId>com.github.albfernandez</groupId>
-			<artifactId>juniversalchardet</artifactId>
-			<version>2.5.0</version>
+			<groupId>com.ibm.icu</groupId>
+			<artifactId>icu4j</artifactId>
+			<version>75.1</version>
 		</dependency>
 		<dependency>
 			<groupId>net.sourceforge.cssparser</groupId>
diff --git a/commafeed-server/src/main/java/com/commafeed/backend/feed/parser/EncodingDetector.java b/commafeed-server/src/main/java/com/commafeed/backend/feed/parser/EncodingDetector.java
index 6932c61b..0dc092d7 100644
--- a/commafeed-server/src/main/java/com/commafeed/backend/feed/parser/EncodingDetector.java
+++ b/commafeed-server/src/main/java/com/commafeed/backend/feed/parser/EncodingDetector.java
@@ -1,11 +1,12 @@
 package com.commafeed.backend.feed.parser;
 
 import java.nio.charset.Charset;
-import java.util.Optional;
 
 import org.apache.commons.lang3.ArrayUtils;
 import org.apache.commons.lang3.StringUtils;
-import org.mozilla.universalchardet.UniversalDetector;
+
+import com.ibm.icu.text.CharsetDetector;
+import com.ibm.icu.text.CharsetMatch;
 
 import jakarta.inject.Singleton;
 
@@ -52,10 +53,14 @@ class EncodingDetector {
 	 * Detect encoding by analyzing characters in the array
 	 */
 	private Charset detectEncoding(byte[] bytes) {
-		UniversalDetector detector = new UniversalDetector();
-		detector.handleData(bytes);
-		detector.dataEnd();
-		String encoding = Optional.ofNullable(detector.getDetectedCharset()).orElse("UTF-8");
+		String encoding = "UTF-8";
+
+		CharsetDetector detector = new CharsetDetector();
+		detector.setText(bytes);
+		CharsetMatch match = detector.detect();
+		if (match != null) {
+			encoding = match.getName();
+		}
 		if (encoding.equalsIgnoreCase("ISO-8859-1")) {
 			encoding = "windows-1252";
 		}