Merge pull request #121 from 4teamwork/docker-service

Provide a web service as a Docker container for composing documents

Author: buchi

.dockerignore

@@ -0,0 +1,3 @@
+.github
+.venv
+tests

.github/workflows/build_image.yml

@@ -0,0 +1,51 @@
+name: Build Docker image
+
+on:
+  workflow_dispatch:
+  push:
+    tags:
+      - '*.*.*'
+    branches:
+      - master
+
+jobs:
+  build:
+    runs-on: self-hosted
+    permissions:
+      contents: read
+      packages: write
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          fetch-tags: true
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+        with:
+          driver: remote
+          endpoint: tcp://buildkitd.buildx:1234
+
+      - name: Login to DockerHub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Set Git commit env variables
+        run: |
+          echo "GIT_TAG=$(git describe --tags --candidates=0)" >> $GITHUB_ENV
+          echo "GIT_SHA_TAG=$(git describe --tags)" >> $GITHUB_ENV
+          echo "LATEST_TAG=$(git describe --tags --abbrev=0 master)" >> $GITHUB_ENV
+          echo "BRANCH_NAME=$(git rev-parse --abbrev-ref HEAD)" >> $GITHUB_ENV
+
+      - name: Build and push image
+        uses: docker/bake-action@v6
+        with:
+          source: .
+          files: docker-bake.hcl
+          push: true

Dockerfile

@@ -0,0 +1,43 @@
+FROM python:3.14-alpine AS base
+
+RUN addgroup -S -g 8080 docxcompose \
+    && adduser -S -D -G docxcompose -u 8080 docxcompose
+
+ENV PYTHONUNBUFFERED=1
+WORKDIR /app
+
+RUN echo "/app/lib/python3.14/site-packages/" > /usr/local/lib/python3.14/site-packages/app.pth \
+ && apk add --no-cache \
+    libxml2 \
+    libxslt
+
+
+FROM base AS builder
+
+RUN apk add --no-cache \
+    gcc \
+    musl-dev \
+    libxml2-dev \
+    libxslt-dev \
+    pipx
+
+RUN pipx install poetry \
+ && pipx inject poetry poetry-plugin-export
+
+COPY pyproject.toml poetry.lock ./
+
+RUN /root/.local/bin/poetry export -f requirements.txt --extras server --output requirements.txt \
+ && pip install --no-cache-dir --no-warn-script-location --prefix ./ -r requirements.txt --no-binary lxml
+
+COPY docxcompose docxcompose
+COPY README.rst .
+RUN pip install --no-cache-dir --prefix ./ .
+RUN rm -rf docxcompose pyproject.toml poetry.lock poetry.toml README.rst requirements.txt
+
+
+FROM base AS prod
+
+COPY --from=builder /app /app
+USER docxcompose
+EXPOSE 8080
+CMD ["/app/bin/docxcompose-server"]

README.rst

@@ -31,6 +31,26 @@ line, e.g.:
     $ docxcompose files/master.docx files/content.docx -o files/composed.docx
 
 
+Docker container
+----------------
+
+docxcompose is also available as a Docker container allowing to compose docx 
+documents through a web service.
+
+To start the web service, run:
+
+.. code:: sh
+
+    $ docker run -it --rm -p 8080:8080 4teamwork/docxcompose
+
+To compose documents, just upload them in the desired order as a ``multipart/form-data``
+request to the web service and you will get back the composed document. Example with curl:
+
+.. code:: sh
+
+    $ curl -F "first=@first.docx" -F "second=@second.docx" -o composed.docx http://localhost:8080/
+
+
 Installation for development
 ----------------------------
 

changes/web-service.feature

@@ -0,0 +1 @@
+Provide a web service as a Docker container for composing documents. [buchi]

compose.yaml

@@ -0,0 +1,8 @@
+services:
+  docxcompose:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    image: 4teamwork/docxcompose:latest
+    ports:
+      - 8080:8080

docker-bake.hcl

@@ -0,0 +1,40 @@
+/*
+# To build locally use:
+GIT_TAG=$(git describe --tags --candidates=0) \
+GIT_SHA_TAG=$(git describe --tags) \
+LATEST_TAG=$(git describe --tags --abbrev=0 master) \
+BRANCH_NAME=$(git rev-parse --abbrev-ref HEAD) \
+docker buildx bake -f docker-bake.hcl --load
+*/
+
+variable "IMAGE_NAME" {
+  default = "docker.io/4teamwork/docxcompose"
+}
+variable "GIT_TAG" {
+  default = ""
+}
+variable "GIT_SHA_TAG" {
+  default = ""
+}
+variable "LATEST_TAG" {
+  default = ""
+}
+variable "BRANCH_NAME" {
+  default = ""
+}
+
+target "default" {
+  dockerfile = "./Dockerfile"
+  context = "."
+  target = "prod"
+  tags = [
+    strlen(GIT_TAG) > 0 ? "${IMAGE_NAME}:${GIT_TAG}": "",
+    equal(GIT_TAG, LATEST_TAG) ? "${IMAGE_NAME}:latest": "",
+    equal(GIT_TAG, "") && equal(BRANCH_NAME, "master") ? "${IMAGE_NAME}:edge": "",
+    notequal(BRANCH_NAME, "master") && strlen(GIT_TAG) < 1 && strlen(GIT_SHA_TAG) > 0 ? "${IMAGE_NAME}:${GIT_SHA_TAG}": "",
+  ]
+  platforms = [
+    "linux/amd64",
+    strlen(GIT_TAG) > 0 ? "linux/arm64" : "",
+  ]
+}

docxcompose/server.py

@@ -0,0 +1,121 @@
+try:
+    from aiohttp import web
+except ImportError:
+    raise SystemExit("Install with server extra to use this command.")
+import importlib.metadata
+import logging
+import os.path
+import tempfile
+
+from docx import Document
+
+from docxcompose.composer import Composer
+
+
+CHUNK_SIZE = 65536
+logger = logging.getLogger("docxcompose")
+version = importlib.metadata.version("docxcompose")
+
+
+async def compose(request):
+
+    documents = []
+    temp_dir = None
+
+    if not request.content_type == "multipart/form-data":
+        logger.info(
+            "Bad request. Received content type %s instead of multipart/form-data.",
+            request.content_type,
+        )
+        return web.Response(status=400, text="Multipart request required")
+
+    reader = await request.multipart()
+
+    with tempfile.TemporaryDirectory() as temp_dir:
+        while True:
+            part = await reader.next()
+
+            if part is None:
+                break
+
+            if part.filename is None:
+                continue
+
+            documents.append(await save_part_to_file(part, temp_dir))
+
+        if not documents:
+            return web.Response(status=400, text="No documents provided")
+
+        composed_filename = os.path.join(temp_dir, "composed.docx")
+
+        try:
+            composer = Composer(Document(documents.pop(0)))
+            for document in documents:
+                composer.append(Document(document))
+            composer.save(composed_filename)
+        except Exception:
+            logger.exception("Failed composing documents.")
+            return web.Response(status=500, text="Failed composing documents")
+
+        return await stream_file(
+            request,
+            composed_filename,
+            "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+        )
+
+
+async def save_part_to_file(part, directory):
+    filename = os.path.join(directory, f"{part.name}_{part.filename}")
+    with open(filename, "wb") as file_:
+        while True:
+            chunk = await part.read_chunk(CHUNK_SIZE)
+            if not chunk:
+                break
+            file_.write(chunk)
+    return filename
+
+
+async def stream_file(request, filename, content_type):
+    response = web.StreamResponse(
+        status=200,
+        reason="OK",
+        headers={
+            "Content-Type": content_type,
+            "Content-Disposition": f'attachment; filename="{os.path.basename(filename)}"',
+        },
+    )
+    await response.prepare(request)
+
+    with open(filename, "rb") as outfile:
+        while True:
+            data = outfile.read(CHUNK_SIZE)
+            if not data:
+                break
+            await response.write(data)
+
+    await response.write_eof()
+    return response
+
+
+async def healthcheck(request):
+    return web.Response(status=200, text="OK")
+
+
+def create_app():
+    app = web.Application()
+    app.add_routes([web.post("/", compose)])
+    app.add_routes([web.get("/healthcheck", healthcheck)])
+    return app
+
+
+def main():
+    print(f"docxcompose {version}")
+    logging.basicConfig(
+        format="%(asctime)s %(levelname)s %(name)s %(message)s",
+        level=logging.INFO,
+    )
+    web.run_app(create_app())
+
+
+if __name__ == "__main__":
+    main()