Automate daily tests of ZIM behavior - Youtube only for now

.github/workflows/DailyTests.yaml

@@ -0,0 +1,27 @@
+name: DailyTests
+
+on:
+  schedule:
+    - cron: "0 4 * * *"
+  workflow_dispatch:
+
+
+jobs:
+  run-daily-tests:
+    runs-on: ubuntu-22.04
+
+    steps:
+      - name: checkout
+        uses: actions/checkout@v4
+
+      - name: build zimit image
+        run: docker build -t local-zimit .
+
+      - name: run crawl of test website
+        run: docker run -v $PWD/output3:/output local-zimit zimit --url https://website.test.openzim.org/ --name tests_eng_test-website --zim-file tests_eng_test-website.zim
+
+      - name: build selenium test image
+        run: docker build -t local-selenium tests-daily
+
+      - name: run integration test suite
+        run: docker run -v $PWD/tests-daily/daily.py:/app/daily.py -v $PWD/output3:/output local-selenium bash -c "cd /app && pytest -h && pytest -v --log-level=INFO --log-format='%(levelname)s - %(message)s' daily.py"

CHANGELOG.md

@@ -10,6 +10,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Added
 
 - Add `--custom-behaviors` argument to support path/HTTP(S) URL custom behaviors to pass to the crawler (#313)
+- Add daily automated end-to-end tests of a page with Youtube player (#330)
 
 ### Changed
 

pyproject.toml

@@ -39,6 +39,7 @@ test = [
 dev = [
   "pre-commit==3.7.1",
   "debugpy==1.8.1",
+  "selenium==4.23.0", # used in daily tests, convenient for dev purpose (autocompletion)
   "zimit[scripts]",
   "zimit[lint]",
   "zimit[test]",

tests-daily/Dockerfile

@@ -0,0 +1,75 @@
+# Let's extract kiwix-tools as usual on alpine temporary build container
+FROM alpine:3.18 as kiwix-serve
+LABEL org.opencontainers.image.source https://github.com/openzim/kiwix-tools
+
+# TARGETPLATFORM is injected by docker build
+ARG TARGETPLATFORM
+ARG KIWIX_TOOLS_VERSION
+
+RUN set -e && \
+    # default (no KIWIX_TOOLS_VERSION set) to today's nightly
+    if [ -z "$KIWIX_TOOLS_VERSION" ] ; then KIWIX_TOOLS_VERSION=$(date +"%Y-%m-%d") ; fi && \
+    apk --no-cache add dumb-init curl && \
+    echo "TARGETPLATFORM: $TARGETPLATFORM" && \
+    if [ "$TARGETPLATFORM" = "linux/386" ]; then ARCH="i586"; \
+    # linux/arm64/v8 points to linux/arm64
+    elif [ "$TARGETPLATFORM" = "linux/arm64/v8" \
+        -o "$TARGETPLATFORM" = "linux/arm64" ]; then ARCH="aarch64"; \
+    # linux/arm translates to linux/arm/v7
+    elif [ "$TARGETPLATFORM" = "linux/arm/v7" ]; then ARCH="armv8"; \
+    elif [ "$TARGETPLATFORM" = "linux/arm/v6" ]; then ARCH="armv6"; \
+    elif [ "$TARGETPLATFORM" = "linux/amd64/v3" \
+        -o "$TARGETPLATFORM" = "linux/amd64/v2" \
+        -o "$TARGETPLATFORM" = "linux/amd64" ]; then ARCH="x86_64"; \
+    # we dont suppot any other arch so let it fail
+    else ARCH="unknown"; fi && \
+    # download requested kiwix-tools version
+    url="http://mirror.download.kiwix.org/nightly/$KIWIX_TOOLS_VERSION/kiwix-tools_linux-$ARCH-$KIWIX_TOOLS_VERSION.tar.gz" && \
+    echo "URL: $url" && \
+    mkdir /kiwix-serve && \
+    curl -k -L $url | tar -xz -C /kiwix-serve --strip-components 1
+
+# Build real "workload" container
+FROM python:3.12-slim-bookworm
+
+# Add kiwix-serve
+COPY --from=kiwix-serve /kiwix-serve /usr/local/bin
+
+# Update apt + install dependencies + install Google Chrome dependencies + clean-up apt lists
+RUN apt-get update -y && \
+    apt-get install -qqy wget xvfb unzip jq && \
+    apt-get install -qqy libxss1 libappindicator1 libgconf-2-4 \
+    fonts-liberation libasound2 libnspr4 libnss3 libx11-xcb1 libxtst6 lsb-release xdg-utils \
+    libgbm1 libnss3 libatk-bridge2.0-0 libgtk-3-0 libx11-xcb1 libxcb-dri3-0 && \
+    rm -rf /var/lib/apt/lists/*
+
+# Fetch the latest version numbers and URLs for Chrome and ChromeDriver
+RUN wget -q -O /tmp/versions.json https://googlechromelabs.github.io/chrome-for-testing/last-known-good-versions-with-downloads.json
+
+# Install chrome
+RUN CHROME_URL=$(jq -r '.channels.Stable.downloads.chrome[] | select(.platform=="linux64") | .url' /tmp/versions.json) && \
+    wget -q --continue -O /tmp/chrome-linux64.zip $CHROME_URL && \
+    unzip /tmp/chrome-linux64.zip -d /opt/chrome
+
+RUN chmod +x /opt/chrome/chrome-linux64/chrome
+
+# Install chromedriver
+RUN CHROMEDRIVER_URL=$(jq -r '.channels.Stable.downloads.chromedriver[] | select(.platform=="linux64") | .url' /tmp/versions.json) && \
+    wget -q --continue -O /tmp/chromedriver-linux64.zip $CHROMEDRIVER_URL && \
+    unzip /tmp/chromedriver-linux64.zip -d /opt/chromedriver && \
+    chmod +x /opt/chromedriver/chromedriver-linux64/chromedriver
+
+# Set up Chromedriver Environment variables
+ENV CHROMEDRIVER_DIR /opt/chromedriver
+ENV PATH $CHROMEDRIVER_DIR:$PATH
+
+# Clean up
+RUN rm /tmp/chrome-linux64.zip /tmp/chromedriver-linux64.zip /tmp/versions.json
+
+# Update pip, install selenium, create work directory
+RUN \
+   python -m pip install --no-cache-dir -U \
+     pip \
+     selenium==4.23.0 \
+     pytest==8.2.2 \
+&& mkdir -p /work

tests-daily/daily.py

@@ -0,0 +1,124 @@
+import logging
+import subprocess
+from time import sleep
+
+import pytest
+from selenium import webdriver
+from selenium.webdriver.chrome.options import Options
+from selenium.webdriver.chrome.service import Service as ChromeService
+from selenium.webdriver.common.by import By
+from selenium.webdriver.support import expected_conditions
+from selenium.webdriver.support.ui import WebDriverWait
+
+KIWIX_SERVE_START_SLEEP = 1
+
+ZIM_NAME = "tests_eng_test-website"
+YOUTUBE_VIDEO_PATH = "youtube.fuzzy.replayweb.page/embed/g5skcrNXdDM"
+
+CHECK_VIDEO_IS_PLAYING_AFTER_SECS = 30
+
+logger = logging.getLogger(__name__)
+
+
+@pytest.fixture(scope="module")
+def chrome_driver():
+    """Start chrome and setup chrome driver / selenium"""
+
+    logger.info("Starting Chrome")
+    chrome_options = Options()
+    chrome_options.add_argument("--headless")
+    chrome_options.add_argument("--no-sandbox")
+    # Other options of interest:
+    # --disable-dev-shm-usage (not needed anymore with recent chrome versions)
+    # --disable-gpu (important for some versions of Chrome)
+    # --remote-debugging-port=9222 (should you need to remote debug)
+
+    # Set path to Chrome binary
+    chrome_options.binary_location = "/opt/chrome/chrome-linux64/chrome"
+
+    # Set path to ChromeDriver
+    chrome_service = ChromeService(
+        executable_path="/opt/chromedriver/chromedriver-linux64/chromedriver"
+    )
+
+    # Set up driver
+    driver = webdriver.Chrome(service=chrome_service, options=chrome_options)
+
+    yield driver
+
+    # Cleanup
+    logger.info("Quitting Chrome")
+    driver.quit()
+
+
+@pytest.fixture(scope="module")
+def kiwix_serve():
+    """Start  kiwix-serve with given ZIM"""
+
+    logger.info("Starting kiwix-serve")
+    process = subprocess.Popen(
+        [
+            "/usr/bin/env",
+            "/usr/local/bin/kiwix-serve",
+            f"/output/{ZIM_NAME}.zim",
+        ]
+    )
+
+    logger.info(
+        f"Waiting {KIWIX_SERVE_START_SLEEP} secs to be 'sure' that kiwix-serve is ready"
+    )
+    sleep(KIWIX_SERVE_START_SLEEP)
+
+    if process.poll() is not None:
+        raise Exception("kiwix-serve has terminated too early")
+
+    yield process
+
+    # Cleanup
+    logger.info("Quitting kiwix-serve")
+    process.terminate()
+
+
+def test_youtube_video(chrome_driver, kiwix_serve):  # noqa: ARG001
+    """Test that youtube video loads, and still plays after a while"""
+
+    chrome_driver.get(f"http://localhost:80/content/{ZIM_NAME}/{YOUTUBE_VIDEO_PATH}")
+
+    if chrome_driver.title == "Content not found":
+        raise Exception("Wrong URL, kiwix-serve said that content is not found")
+
+    button = WebDriverWait(chrome_driver, 1).until(
+        expected_conditions.presence_of_element_located(
+            (By.XPATH, "//button[@title='Play']")
+        )
+    )
+
+    logger.info("Play button found in page")
+
+    button.click()
+
+    video = WebDriverWait(chrome_driver, 1).until(
+        expected_conditions.presence_of_element_located((By.TAG_NAME, "video"))
+    )
+
+    logger.info("Video found in page")
+
+    # arguments[0] is the video tag passed to execute_script
+    if not chrome_driver.execute_script("return arguments[0].paused === false", video):
+        raise Exception("Video is not playing, failed to start probably")
+
+    logger.info("Video is playing")
+
+    logger.info(
+        f"Waiting {CHECK_VIDEO_IS_PLAYING_AFTER_SECS} secs to check video is still "
+        "playing"
+    )
+    sleep(CHECK_VIDEO_IS_PLAYING_AFTER_SECS)
+
+    # arguments[0] is the video tag passed to execute_script
+    if not chrome_driver.execute_script("return arguments[0].paused === false", video):
+        raise Exception(
+            "Video is not playing anymore after "
+            f"{CHECK_VIDEO_IS_PLAYING_AFTER_SECS} secs"
+        )
+    logger.info("Video is still playing")