docker and ocr updates

DockerfileBase

@@ -1,33 +1,35 @@
 # Main stage
-FROM bellsoft/liberica-openjdk-debian:17 AS base
+FROM ubuntu:latest AS base
+
+RUN apt-get update && apt-get install -y --no-install-recommends software-properties-common gnupg2
+
+RUN add-apt-repository ppa:alex-p/tesseract-ocr5 && apt install -y --no-install-recommends tesseract-ocr
+
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
+    	openjdk-17-jre \
         libreoffice-core-nogui \
         libreoffice-common \
         libreoffice-writer-nogui \
         libreoffice-calc-nogui \
         libreoffice-impress-nogui \
         python3-uno \
+        ghostscript \
         python3-pip \
-        unoconv \
-        pngquant \
-        unpaper \
-        ocrmypdf && \
+        ocrmypdf \
+        unoconv && \
     pip install --upgrade pip && \
+    pip install --no-cache-dir --user --upgrade ocrmypdf && \
+    pip install --no-cache-dir --upgrade pillow==10.0.1 reportlab==3.6.13 wheel==0.38.1 setuptools==65.5.1 pyjwt==2.4.0 cryptography==39.0.1 \
+    pip install --no-cache-dir --upgrade ocrmypdf && \
     pip install --no-cache-dir \
-        pillow==10.0.1 \
-        lxml==4.6.5 \
-        reportlab==3.6.13 \
-        setuptools==65.5.1 \
-        pikepdf==4.4.1 \
-        wheel==0.38.1 \
-        cryptography==39.0.1 \
         opencv-python-headless && \
     rm -rf /var/lib/apt/lists/* && \
     mkdir /usr/share/tesseract-ocr-original && \
     cp -r /usr/share/tesseract-ocr/* /usr/share/tesseract-ocr-original && \
     rm -rf /usr/share/tesseract-ocr
 
+
 # Python packages stage
 FROM base AS python-packages
 # Install build tools and Python libraries
@@ -43,32 +45,4 @@ RUN apt-get update && \
 FROM base
 COPY --from=python-packages /usr/local /usr/local
 
-
-# Install wkhtmltopdf
-RUN apt-get update && \
-    apt-get install -y \
-        fontconfig \
-        libfontconfig1 \
-        libfreetype6 \
-        libx11-6 \
-        libxext6 \
-        libxrender1 \
-        xfonts-75dpi \
-        wget \
-        xfonts-base 
-
-# Set a default value for TARGETARCH if it's not provided
-ARG TARGETARCH=arm64
-
-# Conditional statement to choose the correct wkhtmltopdf package based on architecture
-RUN if [ "$TARGETARCH" = "amd64" ]; then \
-        wget https://github.com/wkhtmltopdf/packaging/releases/download/0.12.6.1-3/wkhtmltox_0.12.6.1-3.bullseye_amd64.deb; \
-    elif [ "$TARGETARCH" = "arm64" ]; then \
-        wget https://github.com/wkhtmltopdf/packaging/releases/download/0.12.6.1-3/wkhtmltox_0.12.6.1-3.bullseye_arm64.deb; \
-    else \
-        wget https://github.com/wkhtmltopdf/packaging/releases/download/0.12.6.1-3/wkhtmltox_0.12.6.1-3.bullseye_amd64.deb; \
-    fi && \
-    dpkg -i wkhtmltox_0.12.6.1-3.bullseye_$TARGETARCH.deb && \
-    rm wkhtmltox_0.12.6.1-3.bullseye_$TARGETARCH.deb && \
-    rm -rf /var/lib/apt/lists/*