searchanddisplace-docker/Dockerfile.ingest

# From here we will move the composer binary
FROM composer:1.10 AS composer

# Download base image ubuntu 20.04
FROM ubuntu:20.04

# Changing sh shell to bash
SHELL ["/bin/bash", "-c", "-o", "pipefail"]

# Change working directory
WORKDIR /var/www/html

# Disable Prompt During Packages Installation
ARG DEBIAN_FRONTEND=noninteractive

# Install Apache, PHP7.4-FPM, Supervisor
RUN apt-get update -y && \
    apt-get install -y --no-install-recommends \
    apache2 \
    apache2-doc \
    apache2-utils \
    ca-certificates \
    curl \
    git \
    libapache2-mod-fcgid \
	php7.4 \
	php7.4-common \
	php7.4-fpm \
	php7.4-mbstring \
	php7.4-xml \
    php7.4-zip \
    software-properties-common \
    supervisor && \
    rm -rf /var/lib/apt/lists/*

# Configure packages
RUN a2enmod rewrite actions fcgid alias proxy_fcgi setenvif remoteip && \
    a2enconf php7.4-fpm && \
    bash -c 'echo "RemoteIPHeader X-Forwarded-For" >> /etc/apache2/apache2.conf' && \
    sed -i "s/LogFormat \"%v:%p %h/LogFormat \"%v:%p %a/g" /etc/apache2/apache2.conf && \
    sed -i "s/LogFormat \"%h/LogFormat \"%a/g" /etc/apache2/apache2.conf && \
    sed -i "s/AllowOverride None/AllowOverride All/g" /etc/apache2/apache2.conf && \
    mkdir -p /run/php && \
    mkdir -p /var/run/apache2 && \
    mkdir -p /var/lock/apache2 && \
    mkdir -p /var/log/supervisor && \
    chown -R www-data:www-data /var/www/html && \
    chown -R www-data:www-data /run/php && \
    rm /var/www/html/index.html

# This is a laravel application, change DocRoot
RUN sed -i 's/\/var\/www\/html/&\/public/g' /etc/apache2/sites-available/000-default.conf


# Expose environment variables
RUN sed -i 's/;clear_env = .*/clear_env = no/g' /etc/php/*/fpm/pool.d/www.conf

# Install LibreOffice
RUN apt-get update -y && \
    apt-add-repository -y ppa:libreoffice/ppa && \
    apt-get update -y && \
    apt-get install -y libreoffice

# Install Python3
RUN apt-get update -y && \
    apt-get install -y software-properties-common && \
    add-apt-repository -y ppa:deadsnakes/ppa && \
    apt-get install -y \
        build-essential \
	    libpoppler-cpp-dev \
	    pkg-config \
        python3 \
        python3-dev

# Install PDF Convertor
RUN apt-get update -y && \
    apt-get install -y \
    libpoppler-cpp-dev \
    poppler-utils

# Install Tesseract OCR, Pandoc, DOCX to PDF Convertor, Unpaper
RUN apt-get update -y && \
    add-apt-repository -y ppa:alex-p/tesseract-ocr-devel && \
    apt-get update -y && \
    apt-get install -y \
        pandoc \
        tesseract-ocr \
        unoconv \
        unpaper

# Install Pip
RUN apt-get update -y && \
    curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && \
    python3 get-pip.py && \
    rm -rf get-pip.py && \
    pip3 install --upgrade pip

# Install PdfToText
RUN pip3 install pdftotext

# Prepare Supervisor
RUN mkdir /var/log/queue

## Install Dewarp
WORKDIR /var/www/html/resources/python/dewarp
RUN pip3 install opencv-python
WORKDIR /var/www/html

# Move configuraton files
COPY conf/supervisord_ingest.conf /etc/supervisor/conf.d/supervisord.conf

# Move repository
COPY --chown=www-data:www-data searchanddisplace-ingest/ .

# Here we do the magic!
COPY --from=composer /usr/bin/composer /usr/local/bin/composer

# Install PHP Dependencies
RUN composer install

# Generate key
RUN php artisan key:generate

# Expose Port for the Application 
EXPOSE 80 443

# Copy start.sh script and define default command for the container
CMD ["/usr/bin/supervisord"]