From 91f454fd6a3279da1d0881464fbb0567e229e495 Mon Sep 17 00:00:00 2001
From: Jesus Maria Mendez Perez <jesus.mendez@ticarum.es>
Date: Fri, 24 Nov 2023 10:09:55 +0100
Subject: [PATCH 1/2] SAK-49497 Update Tika Parsers and dependencies

---
 kernel/kernel-impl/pom.xml                    | 10 +++-
 .../content/impl/BaseContentService.java      |  3 +-
 master/pom.xml                                | 46 +++++++++----------
 search/search-impl/impl/pom.xml               | 12 ++++-
 4 files changed, 42 insertions(+), 29 deletions(-)

diff --git a/kernel/kernel-impl/pom.xml b/kernel/kernel-impl/pom.xml
index c870f1e71c06..cce8c1cc0d15 100644
--- a/kernel/kernel-impl/pom.xml
+++ b/kernel/kernel-impl/pom.xml
@@ -88,7 +88,15 @@
         </dependency>
         <dependency>
             <groupId>org.apache.tika</groupId>
-            <artifactId>tika-parsers</artifactId>
+            <artifactId>tika-parsers-standard-package</artifactId>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.tika</groupId>
+            <artifactId>tika-parser-scientific-package</artifactId>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.tika</groupId>
+            <artifactId>tika-parser-sqlite3-package</artifactId>
         </dependency>
         <dependency>
             <groupId>org.zwobble.mammoth</groupId>
diff --git a/kernel/kernel-impl/src/main/java/org/sakaiproject/content/impl/BaseContentService.java b/kernel/kernel-impl/src/main/java/org/sakaiproject/content/impl/BaseContentService.java
index 00687c0949ee..373b153a94cf 100644
--- a/kernel/kernel-impl/src/main/java/org/sakaiproject/content/impl/BaseContentService.java
+++ b/kernel/kernel-impl/src/main/java/org/sakaiproject/content/impl/BaseContentService.java
@@ -78,6 +78,7 @@
 import org.apache.tika.detect.Detector;
 import org.apache.tika.io.TikaInputStream;
 import org.apache.tika.metadata.Metadata;
+import org.apache.tika.metadata.TikaCoreProperties;
 import org.apache.tika.mime.MimeTypes;
 import org.apache.tika.parser.txt.CharsetDetector;
 import org.apache.tika.parser.txt.CharsetMatch;
@@ -5964,7 +5965,7 @@ public void commitResource(ContentResourceEdit edit, int priority) throws OverQu
 
                 final Metadata metadata = new Metadata();
                 //This might not want to be set as it would advise the detector
-                metadata.set(Metadata.RESOURCE_NAME_KEY, edit.getId());
+                metadata.set(TikaCoreProperties.RESOURCE_NAME_KEY, edit.getId());
                 metadata.set(Metadata.CONTENT_TYPE, currentContentType);
                 String newmatch = "";
                 //If we are ignoring the content for this extension, don't give it any data
diff --git a/master/pom.xml b/master/pom.xml
index 1cdb9d4b58e7..4ee370d5ba73 100644
--- a/master/pom.xml
+++ b/master/pom.xml
@@ -72,12 +72,12 @@
     <sakai.spring.security.version>5.5.7</sakai.spring.security.version>
     <sakai.spring.hateoas.version>1.5.5</sakai.spring.hateoas.version>
     <sakai.spring.plugin.version>2.0.0.RELEASE</sakai.spring.plugin.version>
-    <sakai.tika.version>1.28.5</sakai.tika.version>
+    <sakai.tika.version>2.9.1</sakai.tika.version>
     <sakai.tomcat.version>9.0.83</sakai.tomcat.version>
     <sakai.thymeleaf.version>3.0.15.RELEASE</sakai.thymeleaf.version>
     <sakai.velocity.version>1.6.4</sakai.velocity.version>
     <sakai.log4j.version>2.20.0</sakai.log4j.version>
-    <sakai.slf4j.version>1.7.36</sakai.slf4j.version>
+    <sakai.slf4j.version>2.0.9</sakai.slf4j.version>
     <sakai.poi.version>5.2.4</sakai.poi.version>
     <sakai.mockito.version>3.11.2</sakai.mockito.version>
     <sakai.powermock.version>2.0.9</sakai.powermock.version>
@@ -1165,30 +1165,26 @@
         <scope>test</scope>
       </dependency>
 
-      <!-- Tika includes asm-debug-all which it really shouldn't
-           since it is used in multiple places it is declared here so its done right -->
       <dependency>
-        <groupId>org.apache.tika</groupId>
-        <artifactId>tika-parsers</artifactId>
-        <version>${sakai.tika.version}</version>
-        <exclusions>
-          <exclusion>
-            <groupId>de.l3s.boilerpipe</groupId>
-            <artifactId>boilerpipe</artifactId>
-          </exclusion>
-          <exclusion>
-            <groupId>org.ow2.asm</groupId>
-            <artifactId>asm-debug-all</artifactId>
-          </exclusion>
-          <exclusion>
-            <groupId>commons-logging</groupId>
-            <artifactId>commons-logging</artifactId>
-          </exclusion>
-          <exclusion>
-            <groupId>commons-logging</groupId>
-            <artifactId>commons-logging-api</artifactId>
-          </exclusion>
-        </exclusions>
+         <groupId>org.apache.tika</groupId>
+         <artifactId>tika-parsers-standard-package</artifactId>
+         <version>${sakai.tika.version}</version>
+         <exclusions>
+            <exclusion>
+               <groupId>commons-logging</groupId>
+               <artifactId>commons-logging</artifactId>
+            </exclusion>
+         </exclusions>
+      </dependency>
+      <dependency>
+         <groupId>org.apache.tika</groupId>
+         <artifactId>tika-parser-scientific-package</artifactId>
+         <version>${sakai.tika.version}</version>
+      </dependency>
+      <dependency>
+         <groupId>org.apache.tika</groupId>
+         <artifactId>tika-parser-sqlite3-package</artifactId>
+         <version>${sakai.tika.version}</version>
       </dependency>
       <dependency>
         <groupId>org.apache.tika</groupId>
diff --git a/search/search-impl/impl/pom.xml b/search/search-impl/impl/pom.xml
index 6ce87ab1c27a..2f0e6fd49bab 100644
--- a/search/search-impl/impl/pom.xml
+++ b/search/search-impl/impl/pom.xml
@@ -144,8 +144,16 @@
        <artifactId>tika-core</artifactId>
     </dependency>
     <dependency>
-      <groupId>org.apache.tika</groupId>
-      <artifactId>tika-parsers</artifactId>
+        <groupId>org.apache.tika</groupId>
+        <artifactId>tika-parsers-standard-package</artifactId>
+    </dependency>
+    <dependency>
+        <groupId>org.apache.tika</groupId>
+        <artifactId>tika-parser-scientific-package</artifactId>
+    </dependency>
+    <dependency>
+        <groupId>org.apache.tika</groupId>
+        <artifactId>tika-parser-sqlite3-package</artifactId>
     </dependency>
     <dependency>
       <groupId>commons-codec</groupId>

From 88cc90ecdb7003c0c41bb61824dba5198de04bdc Mon Sep 17 00:00:00 2001
From: David Horwitz <dhorwitz@gmail.com>
Date: Fri, 24 Nov 2023 15:24:08 +0200
Subject: [PATCH 2/2] SAK-49497 Update  pdfbox 2.0.29

---
 search/search-impl/impl/pom.xml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/search/search-impl/impl/pom.xml b/search/search-impl/impl/pom.xml
index 2f0e6fd49bab..e620036edadd 100644
--- a/search/search-impl/impl/pom.xml
+++ b/search/search-impl/impl/pom.xml
@@ -120,7 +120,7 @@
     <dependency>
       <groupId>org.apache.pdfbox</groupId>
       <artifactId>pdfbox</artifactId>
-      <version>2.0.26</version>
+      <version>2.0.29</version>
       <exclusions>
         <exclusion>
           <groupId>commons-logging</groupId>