Dockerfile

# Use the Microsoft .NET SDK image to build the project
FROM mcr.microsoft.com/dotnet/sdk:8.0 AS build-env
WORKDIR /ocr

# Copy csproj and restore any dependencies (via NuGet)
COPY . .
RUN dotnet restore "ArkivGPT_OCR.csproj"

# Copy the project files and build our release
RUN dotnet publish -c Release -o out

# Generate runtime image
FROM mcr.microsoft.com/dotnet/sdk:8.0 AS dev-env
WORKDIR /ocr
COPY --from=build-env /ocr/out .
COPY *.py ./
COPY requirements.txt ./

RUN apt-get update && apt-get install -y \
    python3 \
    python3-pip \
    python3-venv \
    tesseract-ocr \ 
    tesseract-ocr-nor \
    poppler-utils
#apt-utils libgdiplus libleptonica-dev libtesseract-dev libc6-dev ghostscript
RUN python3 -m venv /ocr/venv
ENV PATH="/ocr/venv/bin:$PATH"
RUN python3 -m pip install --no-cache-dir -r requirements.txt

RUN dotnet dev-certs https --trust

# Run the gRPC client
ENTRYPOINT ["dotnet", "ArkivGPT_OCR.dll"]