Issuue when installing tesseract-ocr, poppler-utils, libmagic-dev, libgl1

[shriharshan]

I am trying to Dockerize my application and the library I am using requires system dependencies like tesseract-ocr, poppler-utils, libmagic-dev, and libgl1 I tried installing these using dnf and microdnf but each time I try to Dockerize my app I am getting some of these packages are not found, Could anyone guide me in the right direction to install these and by the way I am using Lambda/python-3.12

FROM public.ecr.aws/lambda/python:3.12

RUN microdnf install -y dnf-plugins-core
RUN dnf config-manager --add-repo https://download.opensuse.org/repositories/home:Alexander_Pozdnyakov/Fedora_32/home:Alexander_Pozdnyakov.repo
RUN dnf update -y && dnf install -y \
    libmagic-dev \
    poppler-utils \
    tesseract-ocr \
    libgl1
WORKDIR /app

COPY requirements.txt .

# Install dependencies
RUN pip install -r requirements.txt --target "${LAMBDA_TASK_ROOT}"
    
COPY app/main.py "${LAMBDA_TASK_ROOT}/main.py"
COPY app/middleware.py "${LAMBDA_TASK_ROOT}/middleware.py"

CMD ["main.handler"]

This is my Docker file.

[leandrodamascena]

Hi @shriharshan! I'm not sure if you can install these packages using the base Lambda image for Python 3.12 because a few points:

1 - The python3.12 Lambda base image uses minimal linux container which implements microdnf instead of dnf and even if you install dnf-plugins-core you probably won't be able to run dnf config-manager --add-repo to add this new repository.

2 - This repository seems wrong, they don't support Feroda32, I could only see support for CentoOS and RHEL and even using them on an amazonlinux2023 image I wasn't able to install these packages due to dependency conflicts.

3 - You can try downloading the rpm for these libraries and installing it from there, but I think that will be a nightmare to deal with conflict resolution.

4 - If I were you, I would build this image from the python:3.12 image and then install those dependencies + aws_lambda_ric. Check the documentation here on how to build a custom image.

5 - The upstream python3.12 image is built on top of Debian, so you can use this Dockerfile to get some ideas and build your own image:

# Define custom function directory
ARG FUNCTION_DIR="/function"

FROM python:3.12 AS build-image

# Include global arg in this stage of the build
ARG FUNCTION_DIR

# Install necessary packages
RUN apt-get update && apt-get install -y \
    libmagic-dev \
    poppler-utils \
    tesseract-ocr \
    libgl1 \
    && rm -rf /var/lib/apt/lists/*

# Copy function code
RUN mkdir -p ${FUNCTION_DIR}
COPY . ${FUNCTION_DIR}

# Install the function's dependencies
RUN pip install \
    --target ${FUNCTION_DIR} \
        awslambdaric


COPY requirements.txt .

# Install dependencies
RUN pip install -r requirements.txt --target "${FUNCTION_DIR}"
    
COPY app/main.py "${FUNCTION_DIR}/main.py"
COPY app/middleware.py "${FUNCTION_DIR}/middleware.py"

# Set runtime interface client as default command for the container runtime
ENTRYPOINT [ "/usr/local/bin/python", "-m", "awslambdaric" ]
# Pass the name of the function handler as an argument to the runtime
CMD [ "lambda_function.handler" ]

I hope this help.