Merge pull request #30 from bcye/feature/docker-multi-stage-builds

Docker multi stage builds

.dockerignore

@@ -1,8 +0,0 @@
-.env
-__pycache__
-.venv
-.pytest_cache
-docs
-node_modules
-output
-sketching

.github/workflows/build-image.yaml

@@ -23,11 +23,15 @@ jobs:
           username: ${{ github.actor }}
           password: ${{ secrets.GITHUB_TOKEN }}
 
+      - name: Set up QEMU for multi-platform builds
+        uses: docker/setup-qemu-action@v3
+
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
 
       - name: Build and push
         uses: docker/build-push-action@v6
         with:
+          platforms: linux/amd64,linux/arm64
           push: true
           tags: ghcr.io/bcye/structured-wikivoyage-exports:latest

.github/workflows/test-parser.yaml

@@ -20,4 +20,4 @@ jobs:
         run: uv sync --locked --dev
 
       - name: Run tests
-        run: PYTHONPATH=. uv run pytest
+        run: PYTHONPATH=src uv run pytest

Dockerfile

@@ -1,11 +1,16 @@
-FROM ghcr.io/astral-sh/uv:0.6-python3.12-bookworm
+# use python 3.12 as a base image
+FROM docker.io/python:3.12-alpine
+# use the latest version of uv, independently of the python version
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
 WORKDIR /app
 
-COPY pyproject.toml uv.lock ./
+# copy the requirements and install them
+COPY pyproject.toml uv.lock .
 RUN uv sync --frozen
 
-COPY . .
+# copy the rest of the code
+COPY src ./
 
 RUN chmod +x entrypoint.sh
 

README.md

@@ -18,7 +18,7 @@ See [docs](docs) for more information on how to use this utility.
 
 ## Testing
 
-Run `PYTHONPATH=. pytest` from inside the venv
+Run `PYTHONPATH=src pytest` from inside the venv, or directly call `PYTHONPATH=src uv run -- pytest`.
 
 ## License
 

entrypoint.sh

@@ -1,2 +0,0 @@
-#!/bin/bash
-uv run main.py

src/entrypoint.sh

@@ -0,0 +1,2 @@
+#!/bin/sh
+uv run main.py

src/transformers/wiki_dump_handler.py

@@ -5,6 +5,7 @@ from .parser import WikivoyageParser
 
 logger = getLogger(__name__)
 
+
 class WikiDumpHandler(xml.sax.ContentHandler):
     """
     SAX handler that, for each <page> whose <id> is in mappings,
@@ -92,9 +93,9 @@ class WikiDumpHandler(xml.sax.ContentHandler):
     async def _process(self, text: str, uid: str, title: str):
         parser = WikivoyageParser()
         entry = parser.parse(text)
-        entry['properties']['title'] = title
+        entry["properties"]["title"] = title
 
         # Write to all handlers concurrently
-        await asyncio.gather(*[
-            handler.write_entry(entry, uid) for handler in self.handlers
-        ])
+        await asyncio.gather(
+            *[handler.write_entry(entry, uid) for handler in self.handlers]
+        )