Merge branch 'refactor/prod-and-dev-Dockerfiles' into ci/refactor-ghas

Author: a-klos

services/document-extractor/Dockerfile.dev

@@ -9,11 +9,14 @@ WORKDIR /app
 
 RUN DEBIAN_FRONTEND=noninteractive apt-get update \
     && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
-       build-essential make ffmpeg poppler-utils tesseract-ocr tesseract-ocr-deu tesseract-ocr-eng \
+       build-essential make ffmpeg poppler-utils \
+       tesseract-ocr tesseract-ocr-deu tesseract-ocr-eng \
+       libleptonica-dev pkg-config \
     && python3 -m venv "${POETRY_VIRTUALENVS_PATH}" \
     && ${POETRY_VIRTUALENVS_PATH}/bin/pip install "poetry==${POETRY_VERSION}" \
     && rm -rf /var/lib/apt/lists/*
 ENV PATH="${POETRY_VIRTUALENVS_PATH}/bin:$PATH"
+ENV TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata
 
 # Copy lockfiles first
 COPY services/document-extractor/pyproject.toml services/document-extractor/poetry.lock /app/services/document-extractor/

services/document-extractor/README.md

@@ -7,16 +7,21 @@ The following endpoints are provided by the *documents_extractor*:
 # Requirements
 All required python libraries can be found in the [pyproject.toml](pyproject.toml) file.
 In addition to python libraries the following system packages are required:
-```
+
+```shell
 build-essential
 make
 ffmpeg
 poppler-utils
 tesseract-ocr
 tesseract-ocr-deu
 tesseract-ocr-eng
+libleptonica-dev
+pkg-config
 ```
 
+The Tesseract data path is set via `TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata` in both prod and dev images.
+
 # Endpoints
 
 ## `/extract`
@@ -31,4 +36,3 @@ The following types of information will be extracted:
 A detailed explanation of the deployment can be found in the [project README](../../README.md).
 The *helm-chart* used for the deployment can be found in the [infrastructure directory](../../infrastructure/).
 
-

services/mcp-server/Dockerfile.dev

@@ -1,4 +1,4 @@
-FROM --platform=linux/amd64 python:3.11.7-bookworm
+FROM --platform=linux/amd64 python:3.13.7-bookworm
 
 # Dev image for mcp-server (no local libs)
 ENV POETRY_VIRTUALENVS_PATH=/app/services/mcp-server/.venv
@@ -23,8 +23,8 @@ RUN poetry config virtualenvs.create false \
     && cd /app/services/mcp-server \
     && poetry install --no-interaction --no-ansi --no-root --with dev
 
-# Create non-root user
-RUN adduser --disabled-password --gecos "" --uid 65532 nonroot
+# Create non-root user (align with prod UID for consistent file perms)
+RUN adduser --disabled-password --gecos "" --uid 10001 nonroot
 
 WORKDIR /app/services/mcp-server
 RUN mkdir -p log && chmod 700 log \