From 2d4970ce5556056207b74e02eb4df44ac61cafc1 Mon Sep 17 00:00:00 2001
From: Anthony Stirling <77850077+Frooodle@users.noreply.github.com>
Date: Mon, 10 Feb 2025 10:07:16 +0000
Subject: [PATCH] Update Dockerfile2

---
 Dockerfile2 | 109 +++++++++++++++++++++++++++++++++-------------------
 1 file changed, 69 insertions(+), 40 deletions(-)

diff --git a/Dockerfile2 b/Dockerfile2
index 21e781a42..8d8b13628 100644
--- a/Dockerfile2
+++ b/Dockerfile2
@@ -25,55 +25,84 @@ WORKDIR /downloader
 RUN curl -s https://api.github.com/repos/Stirling-Tools/Stirling-PDF/releases/latest \
     | jq -r ".assets[] | select(.name == \"Stirling-PDF.jar\") | .browser_download_url" \
     | xargs curl -L -o Stirling-PDF.jar
-    
-    
-# Use the base image
-FROM frooodle/stirling-pdf-base:version8
 
-ARG VERSION_TAG
 
-    
-# Set Environment Variables
-ENV DOCKER_ENABLE_SECURITY=false \
-    HOME=/home/stirlingpdfuser \
-    VERSION_TAG=$VERSION_TAG \
-    JAVA_TOOL_OPTIONS="$JAVA_TOOL_OPTIONS -XX:MaxRAMPercentage=75"
-#    PUID=1000 \
-#    PGID=1000 \
-#    UMASK=022 \
-    
+# Main stage
+FROM alpine:3.21.2@sha256:56fa17d2a7e7f168a043a2712e63aed1f8543aeafdcee47c58dcffe38ed51099
 
-# Create user and group
-##RUN groupadd -g $PGID stirlingpdfgroup && \
-##    useradd -u $PUID -g stirlingpdfgroup -s /bin/sh stirlingpdfuser && \
-##    mkdir -p $HOME && chown stirlingpdfuser:stirlingpdfgroup $HOME
-
-# Set up necessary directories and permissions
-RUN mkdir -p /scripts /usr/share/fonts/opentype/noto /usr/share/tesseract-ocr /configs /customFiles 
-##&& \
-##    chown -R stirlingpdfuser:stirlingpdfgroup /scripts /usr/share/fonts/opentype/noto /usr/share/tesseract-ocr /configs /customFiles && \
-##    chown -R stirlingpdfuser:stirlingpdfgroup /usr/share/tesseract-ocr-original
-
-# Copy the already-converted scripts from the dos2unix stage
-COPY --from=dos2unix-converter /dos2unix-scripts/ /scripts/
 # Copy necessary files
-# COPY src/main/resources/static/fonts/*.ttf /usr/share/fonts/opentype/noto/
-# COPY src/main/resources/static/fonts/*.otf /usr/share/fonts/opentype/noto/
+COPY scripts /scripts
+COPY pipeline /pipeline
+COPY src/main/resources/static/fonts/*.ttf /usr/share/fonts/opentype/noto/
 
 # Copy the application JAR file from the download stage
 COPY --from=downloader /downloader/Stirling-PDF.jar /app.jar
 
-# Set font cache and permissions
-RUN fc-cache -f -v && chmod +x /scripts/*
+ARG VERSION_TAG
 
-##&& \
-##    chown stirlingpdfuser:stirlingpdfgroup /app.jar && \
-##    chmod +x /scripts/init.sh
+# Set Environment Variables
+ENV DOCKER_ENABLE_SECURITY=false \
+    VERSION_TAG=$VERSION_TAG \
+    JAVA_TOOL_OPTIONS="-XX:+UnlockExperimentalVMOptions \
+    -XX:MaxRAMPercentage=75 \
+    -XX:InitiatingHeapOccupancyPercent=20 \
+    -XX:+G1PeriodicGCInvokesConcurrent \
+    -XX:G1PeriodicGCInterval=10000 \
+    -XX:+UseStringDeduplication \
+    -XX:G1PeriodicGCSystemLoadThreshold=70" \
+    HOME=/home/stirlingpdfuser \
+    PUID=1000 \
+    PGID=1000 \
+    UMASK=022 \
+    FAT_DOCKER=true \
+    INSTALL_BOOK_AND_ADVANCED_HTML_OPS=false
 
-# Expose necessary ports
-EXPOSE 8080
+
+# JDK for app
+RUN echo "@testing https://dl-cdn.alpinelinux.org/alpine/edge/main" | tee -a /etc/apk/repositories && \
+    echo "@testing https://dl-cdn.alpinelinux.org/alpine/edge/community" | tee -a /etc/apk/repositories && \
+    echo "@testing https://dl-cdn.alpinelinux.org/alpine/edge/testing" | tee -a /etc/apk/repositories && \
+    apk upgrade --no-cache -a && \
+    apk add --no-cache \
+        ca-certificates \
+        tzdata \
+        tini \
+        bash \
+        curl \
+        shadow \
+        su-exec \
+        openssl \
+        openssl-dev \
+        openjdk21-jre \
+# Doc conversion
+		gcompat \
+    	libc6-compat \
+        libreoffice \
+# pdftohtml
+        poppler-utils \
+# OCR MY PDF (unpaper for descew and other advanced featues)
+        qpdf \
+        tesseract-ocr-data-eng \
+        font-terminus font-dejavu font-noto font-noto-cjk font-awesome font-noto-extra \
+# CV
+		py3-opencv \
+# python3/pip
+        python3 \
+    py3-pip && \
+# uno unoconv and HTML
+    pip install --break-system-packages --no-cache-dir --upgrade unoconv WeasyPrint pdf2image pillow && \
+    mv /usr/share/tessdata /usr/share/tessdata-original && \
+    mkdir -p $HOME /configs /logs /customFiles /pipeline/watchedFolders /pipeline/finishedFolders && \
+    fc-cache -f -v && \
+    chmod +x /scripts/* && \
+    chmod +x /scripts/init.sh && \
+# User permissions
+    addgroup -S stirlingpdfgroup && adduser -S stirlingpdfuser -G stirlingpdfgroup && \
+    chown -R stirlingpdfuser:stirlingpdfgroup $HOME /scripts /usr/share/fonts/opentype/noto /configs /customFiles /pipeline && \
+    chown stirlingpdfuser:stirlingpdfgroup /app.jar
+
+EXPOSE 8080/tcp
 
 # Set user and run command
-##USER stirlingpdfuser
-ENTRYPOINT ["/scripts/init.sh"]
-CMD ["java", "-jar", "/app.jar"]
+ENTRYPOINT ["tini", "--", "/scripts/init.sh"]
+CMD ["java", "-Dfile.encoding=UTF-8", "-jar", "/app.jar"]