From ef33864adfd5971073e562c92d0771dc9b855fd6 Mon Sep 17 00:00:00 2001
From: Andy Grunwald <andygrunwald@gmail.com>
Date: Thu, 6 Feb 2025 18:40:21 +0100
Subject: [PATCH] New Script: Apache Tika (#2079)

* New Script: Apache Tika

* Temp: Replace github URLs to my own fork

* Add additional dependencies according to the Docker image installation

See https://github.com/apache/tika-docker/blob/master/full/Dockerfile

* Apache Tika: Set correct tags

* Apache Tika: Set TODO to make it updateable

* Apache Tika: Fix "software-properties-common: command not found"

* Apache Tika: Automate version detection

* Apache Tika: Add `update_script`

* Apache Tika: Added clean up of `/opt/apache-tika/tika-server-standard-prev-version.jar` after upgrade

* Apache Tika: Bump up ram to 2048

* Apache Tika: Set updateable to true

* Apache Tika: Switch from `default-jdk` to `openjdk-17-jre-headless`

* Apache Tika: Removed comment about Docker file

* Apache Tika: Removed empty line

* Revert "Temp: Replace github URLs to my own fork"

This reverts commit f1c5d8720696cdbfde9471abfff07e4b7b71bc6d.
---
 ct/apache-tika.sh              | 69 ++++++++++++++++++++++++++++++
 install/apache-tika-install.sh | 78 ++++++++++++++++++++++++++++++++++
 json/apache-tika.json          | 34 +++++++++++++++
 3 files changed, 181 insertions(+)
 create mode 100755 ct/apache-tika.sh
 create mode 100644 install/apache-tika-install.sh
 create mode 100644 json/apache-tika.json

diff --git a/ct/apache-tika.sh b/ct/apache-tika.sh
new file mode 100755
index 00000000..6227de95
--- /dev/null
+++ b/ct/apache-tika.sh
@@ -0,0 +1,69 @@
+#!/usr/bin/env bash
+source <(curl -s https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/misc/build.func)
+# Copyright (c) 2021-2025 community-scripts ORG
+# Author: Andy Grunwald (andygrunwald)
+# License: MIT | https://github.com/community-scripts/ProxmoxVE/raw/main/LICENSE
+# Source: https://github.com/apache/tika/
+
+# App Default Values
+APP="Apache-Tika"
+var_tags="document"
+var_cpu="1"
+var_ram="2048"
+var_disk="10"
+var_os="debian"
+var_version="12"
+var_unprivileged="1"
+
+# App Output & Base Settings
+header_info "$APP"
+base_settings
+
+# Core
+variables
+color
+catch_errors
+
+function update_script() {
+  header_info
+  check_container_storage
+  check_container_resources
+  if [[ ! -f /etc/systemd/system/apache-tika.service ]]; then
+    msg_error "No ${APP} Installation Found!"
+    exit
+  fi
+  RELEASE="$(wget -qO- https://dlcdn.apache.org/tika/ | grep -oP '(?<=href=")[0-9]+\.[0-9]+\.[0-9]+(?=/")' | sort -V | tail -n1)"
+  if [[ ! -f /opt/${APP}_version.txt ]] || [[ "${RELEASE}" != "$(cat /opt/${APP}_version.txt)" ]]; then
+    msg_info "Stopping ${APP}"
+    systemctl stop apache-tika
+    msg_ok "Stopped ${APP}"
+
+    msg_info "Updating ${APP} to v${RELEASE}"
+    cd /opt/apache-tika
+    wget -q "https://dlcdn.apache.org/tika/${RELEASE}/tika-server-standard-${RELEASE}.jar"
+    mv --force tika-server-standard.jar tika-server-standard-prev-version.jar
+    mv tika-server-standard-${RELEASE}.jar tika-server-standard.jar
+    echo "${RELEASE}" >/opt/${APP}_version.txt
+    msg_ok "Updated ${APP} to v${RELEASE}"
+
+    msg_info "Starting ${APP}"
+    systemctl start apache-tika
+    msg_ok "Started ${APP}"
+    msg_info "Cleaning Up"
+    rm -rf /opt/apache-tika/tika-server-standard-prev-version.jar
+    msg_ok "Cleanup Completed"
+    msg_ok "Updated Successfully"
+  else
+    msg_ok "No update required. ${APP} is already at v${RELEASE}"
+  fi
+  exit
+}
+
+start
+build_container
+description
+
+msg_ok "Completed Successfully!\n"
+echo -e "${CREATING}${GN}${APP} setup has been successfully initialized!${CL}"
+echo -e "${INFO}${YW} Access it using the following URL:${CL}"
+echo -e "${TAB}${GATEWAY}${BGN}http://${IP}:9998${CL}"
\ No newline at end of file
diff --git a/install/apache-tika-install.sh b/install/apache-tika-install.sh
new file mode 100644
index 00000000..1976e60b
--- /dev/null
+++ b/install/apache-tika-install.sh
@@ -0,0 +1,78 @@
+#!/usr/bin/env bash
+
+# Copyright (c) 2021-2025 community-scripts ORG
+# Author: Andy Grunwald (andygrunwald)
+# License: MIT | https://github.com/community-scripts/ProxmoxVE/raw/main/LICENSE
+# Source: https://github.com/apache/tika/
+
+source /dev/stdin <<< "$FUNCTIONS_FILE_PATH"
+color
+verb_ip6
+catch_errors
+setting_up_container
+network_check
+update_os
+
+msg_info "Installing Dependencies"
+$STD apt-get install -y \
+  curl \
+  sudo \
+  mc \
+  software-properties-common \
+  gdal-bin \
+  tesseract-ocr \
+  tesseract-ocr-eng \
+  tesseract-ocr-ita \
+  tesseract-ocr-fra \
+  tesseract-ocr-spa \
+  tesseract-ocr-deu
+$STD echo ttf-mscorefonts-installer msttcorefonts/accepted-mscorefonts-eula select true | debconf-set-selections
+$STD apt-get install -y \
+  xfonts-utils \
+  fonts-freefont-ttf \
+  fonts-liberation \
+  ttf-mscorefonts-installer \
+  cabextract
+msg_ok "Installed Dependencies"
+
+msg_info "Setup OpenJDK"
+$STD apt-get install -y \
+  openjdk-17-jre-headless
+msg_ok "Setup OpenJDK"
+
+msg_info "Installing Apache Tika"
+mkdir -p /opt/apache-tika
+cd /opt/apache-tika
+RELEASE="$(wget -qO- https://dlcdn.apache.org/tika/ | grep -oP '(?<=href=")[0-9]+\.[0-9]+\.[0-9]+(?=/")' | sort -V | tail -n1)"
+wget -q "https://dlcdn.apache.org/tika/${RELEASE}/tika-server-standard-${RELEASE}.jar"
+mv tika-server-standard-${RELEASE}.jar tika-server-standard.jar
+echo "${RELEASE}" >/opt/${APPLICATION}_version.txt
+msg_ok "Installed Apache Tika"
+
+msg_info "Creating Service"
+cat <<EOF >/etc/systemd/system/apache-tika.service
+[Unit]
+Description=Apache Tika
+Documentation=https://tika.apache.org/
+After=syslog.target network.target
+
+[Service]
+User=root
+Restart=always
+Type=simple
+ExecStart=java -jar /opt/apache-tika/tika-server-standard.jar --host 0.0.0.0 --port 9998
+ExecReload=/bin/kill -HUP \$MAINPID
+
+[Install]
+WantedBy=multi-user.target
+EOF
+systemctl enable -q --now apache-tika
+msg_ok "Created Service"
+
+motd_ssh
+customize
+
+msg_info "Cleaning up"
+$STD apt-get -y autoremove
+$STD apt-get -y autoclean
+msg_ok "Cleaned"
diff --git a/json/apache-tika.json b/json/apache-tika.json
new file mode 100644
index 00000000..1e891727
--- /dev/null
+++ b/json/apache-tika.json
@@ -0,0 +1,34 @@
+{
+    "name": "Apache Tika",
+    "slug": "apache-tika",
+    "categories": [
+        12
+    ],
+    "date_created": "2025-02-05",
+    "type": "ct",
+    "updateable": true,
+    "privileged": false,
+    "interface_port": 9998,
+    "documentation": null,
+    "website": "https://tika.apache.org/",
+    "logo": "https://tika.apache.org/tika.png",
+    "description": "The Apache Tika™ toolkit detects and extracts metadata and text from over a thousand different file types (such as PPT, XLS, and PDF). All of these file types can be parsed through a single interface, making Tika useful for search engine indexing, content analysis, translation, and much more.",
+    "install_methods": [
+        {
+            "type": "default",
+            "script": "ct/apache-tika.sh",
+            "resources": {
+                "cpu": 1,
+                "ram": 2024,
+                "hdd": 10,
+                "os": "debian",
+                "version": "12"
+            }
+        }
+    ],
+    "default_credentials": {
+        "username": null,
+        "password": null
+    },
+    "notes": []
+}
\ No newline at end of file