Add Unit Testing (#16)

* add pytest and tox for unit testing

* add tests for regex checking

* add a base test class

* test Twitter downloads

* test Youtube downloads

* run tests in a PR and pushes

* test only py10 and higher due to used features

* half handle YouTube sign-in error by skipping
skipping is the way to fix this for now, without some OAuth login

* add Instagram tests

Had to modify Instagram login, which might cause issues for other branches we'll see...

* give env variables to tox

* make unit tests run only on main pushes and PRs

* use git lfs for should_be mp4 files

* fix path issue when just using pytest

.gitattributes

@@ -0,0 +1 @@
+*.mp4 filter=lfs diff=lfs merge=lfs -text

.github/workflows/run_tests.yaml

@@ -0,0 +1,40 @@
+name: check
+env:
+  INSTAGRAM_USERNAME: ${{ secrets.INSTAGRAM_USERNAME }}
+  INSTAGRAM_SESSION: ${{ secrets.INSTAGRAM_SESSION_JSON }}
+on:
+  push:
+    branches:
+      - main
+      - master
+  pull_request:
+
+jobs:
+  test:
+    name: test with ${{ matrix.py }} on ${{ matrix.os }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        py:
+          - "3.12"
+          - "3.11"
+          - "3.10"
+        os:
+          - ubuntu-latest
+          - macos-latest
+          - windows-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+      - name: Setup python for test ${{ matrix.py }}
+        uses: actions/setup-python@v4
+        with:
+          python-version: ${{ matrix.py }}
+      - name: Install tox
+        run: python -m pip install tox-gh>=1.3
+      - name: Setup test suite
+        run: tox -vv --notest
+      - name: Run test suite
+        run: tox --skip-pkg-install

Tests/video_system/download_tester.py

@@ -0,0 +1,21 @@
+from src.downloader import VIDEO_RETURN_TYPE
+
+
+class DownloadTester:
+    def download_single_video_test(self, videos: VIDEO_RETURN_TYPE, should_be_path: str):
+        assert len(videos) == 1
+        video = videos[0]
+
+        with open(video.path, "rb") as downloaded, open(
+            should_be_path, "rb"
+        ) as should_be:
+            assert downloaded.read() == should_be.read(), "Downloaded file does not match the expected file"
+
+    def download_multiple_video_test(self, videos: VIDEO_RETURN_TYPE, should_be_paths: list[str]):
+        assert len(videos) == len(should_be_paths), f"len(videos)={len(videos)} len(should_be_paths)={len(should_be_paths)}"
+
+        for video, should_be_path in zip(videos, should_be_paths):
+            with open(video.path, "rb") as downloaded, open(
+                should_be_path, "rb"
+            ) as should_be:
+                assert downloaded.read() == should_be.read(), "Downloaded file does not match the expected file"

Tests/video_system/instagram/test_instagram.py

@@ -0,0 +1,54 @@
+import logging
+import os
+import shutil
+
+import pytest
+
+from Tests.video_system.download_tester import DownloadTester
+from src.instagram import InstagramDownloader
+
+DOWNLOAD_PATH = os.path.join("Tests", "video_system", "instagram", "downloads")
+
+# no params
+TEST_REEL_1 = "https://www.instagram.com/reel/C2-vIqjsosW/"
+#extra params
+TEST_REEL_2 = "https://www.instagram.com/reel/C2-vIqjsosW/?igsh=eGFma20zbWY0bDJs"
+#multiple videos
+TEST_REEL_3 = "https://www.instagram.com/p/C9tjADzPwDz/?igsh=MnF5dzhjajhxZ3Rs"
+
+class TestInstagramDownloader(DownloadTester):
+    @pytest.fixture(autouse=True)
+    def remove_all_cache(self):
+        logging.basicConfig(level=logging.DEBUG)
+        os.makedirs(DOWNLOAD_PATH, exist_ok=True)
+        shutil.rmtree(DOWNLOAD_PATH, ignore_errors=True)
+        yield
+        # delete the downloads folder after the test.
+        assert os.path.exists(DOWNLOAD_PATH)
+        shutil.rmtree(DOWNLOAD_PATH, ignore_errors=True)
+
+    def test_instagram_single_video_download(self):
+        videos = InstagramDownloader.download_video_from_link(
+            TEST_REEL_1, path=DOWNLOAD_PATH
+        )
+        should_be_path = os.path.join("Tests", "video_system", "instagram", "should_be_0.mp4")
+
+        assert videos, "couldn't download, probably due to graphql error (login wasn't successful)"
+        self.download_single_video_test(videos, should_be_path)
+
+    def test_instagram_download_extra_params(self):
+        videos = InstagramDownloader.download_video_from_link(
+            TEST_REEL_2, path=DOWNLOAD_PATH
+        )
+        should_be_path = os.path.join("Tests", "video_system", "instagram", "should_be_0.mp4")
+
+        self.download_single_video_test(videos, should_be_path)
+
+    def test_instagram_download_multiple_videos(self):
+        videos = InstagramDownloader.download_video_from_link(
+            TEST_REEL_3, path=DOWNLOAD_PATH
+        )
+
+        expected_files = [os.path.join("Tests", "video_system", "instagram", should_be) for should_be in [f"should_be_{i}.mp4" for i in range(1, 7)]]
+
+        self.download_multiple_video_test(videos, expected_files)

Tests/video_system/regex/test_instagram_regex.py

@@ -0,0 +1,55 @@
+import pytest
+
+from src.instagram import InstagramDownloader
+from Tests.video_system.regex.test_regex_base import TestDownloaderRegex
+
+
+class TestInstagramRegex(TestDownloaderRegex):
+    @pytest.fixture(autouse=True)
+    def setup_class(self):
+        super().setup(InstagramDownloader)
+
+    def test_direct_reel_link(self):
+        assert self.check_link("https://www.instagram.com/reel/C2-vIqjsosW")
+
+    def test_direct_post_link(self):
+        assert self.check_link("https://www.instagram.com/p/C2-vIqjsosW")
+
+    def test_direct_link_with_params(self):
+        assert self.check_link(
+            "https://www.instagram.com/p/C2-vIqjsosW/?igsh=eGFma20zbWY0bDJs"
+        )
+        assert self.check_link(
+            "https://www.instagram.com/reel/C2-vIqjsosW/?igsh=eGFma20zbWY0bDJs"
+        )
+
+    def test_shortened_link(self):
+        assert self.check_link("https://instagr.am/p/C2-vIqjsosW")
+        assert self.check_link("https://instagr.am/reel/C2-vIqjsosW")
+
+    def test_extra_text_link(self):
+        assert self.check_link(
+            "Check this out: https://www.instagram.com/p/C2-vIqjsosW"
+        )
+        assert self.check_link(
+            "Check this out: https://www.instagram.com/reel/C2-vIqjsosW"
+        )
+        assert self.check_link(
+            "Dude look at this video fr fr https://www.instagram.com/reel/C2-vIqjsosW"
+        )
+
+    def test_invalid_link(self):
+        assert not self.check_link("https://www.instagram.com")
+
+    def test_invalid_link_with_extra(self):
+        assert not self.check_link("Look at this: https://www.instagram.com")
+
+    def test_invalid_shortened_link(self):
+        assert not self.check_link("https://instagr.am")
+
+    def test_different_type_links(self):
+        assert not self.check_link("https://www.youtube.com/watch?v=dQw4w9WgXcQ")
+        assert not self.check_link("https://x.com/MIT_CSAIL/status/1363172815315214336")
+
+    def test_non_link(self):
+        assert not self.check_link("This is not a link")

Tests/video_system/regex/test_other_regex.py

@@ -0,0 +1,20 @@
+from src.downloading_system import get_downloader
+
+
+class TestOtherDownloaderRegex:
+    def check_link(self, link: str):
+        return get_downloader(link) is not None
+
+    def test_not_supported_links(self):
+        assert not self.check_link("https://facebook.com/1234")
+        assert not self.check_link(
+            "https://www.linkedin.com/posts/microsoft_microsoftlife-microsoftcareersemea-softwareengineerjobs-activity-7231661283509460992-3NLl"
+        )
+        assert not self.check_link("https://twitch.tv/1234")
+
+    def test_link_like_non_links(self):
+        assert not self.check_link("This is not a link")
+        assert not self.check_link("alink://link/link")
+        assert not self.check_link("https://")
+        assert not self.check_link("https://youtube./com/123")
+        assert not self.check_link("https://x./com/123")

Tests/video_system/regex/test_regex_base.py

@@ -0,0 +1,13 @@
+from typing import Type
+
+from src.downloader import VideoDownloader
+from src.downloading_system import get_downloader
+
+
+class TestDownloaderRegex:
+    def setup(self, downloader_type_to_check: Type[VideoDownloader]) -> None:
+        self.downloader_type = downloader_type_to_check
+
+    def check_link(self, link: str):
+        downloader = get_downloader(link)
+        return downloader is not None and isinstance(downloader(), self.downloader_type)

Tests/video_system/regex/test_twitter_regex.py

@@ -0,0 +1,34 @@
+import pytest
+
+from src.twitter import TwitterDownloader
+from Tests.video_system.regex.test_regex_base import TestDownloaderRegex
+
+
+class TestTwitterRegex(TestDownloaderRegex):
+    @pytest.fixture(autouse=True)
+    def setup_class(self):
+        super().setup(TwitterDownloader)
+
+    def test_direct_link(self):
+        assert self.check_link("https://x.com/MIT_CSAIL/status/1363172815315214336")
+
+    def test_shortened_link(self):
+        assert self.check_link("https://t.co/1234")
+
+    def test_invalid_link(self):
+        assert not self.check_link("https://x.com")
+
+    def test_extra_text_link(self):
+        assert self.check_link(
+            "Dude check this out: https://x.com/MIT_CSAIL/status/1363172815315214336"
+        )
+
+    def test_non_link(self):
+        assert not self.check_link("This is not a link")
+
+    def test_not_supported_link(self):
+        assert not self.check_link("Check this out: https://facebook.com/1234")
+
+    def test_different_type_links(self):
+        assert not self.check_link("https://www.youtube.com/watch?v=dQw4w9WgXcQ")
+        assert not self.check_link("https://www.instagram.com/p/CMJ1J1JjJjJ/")

Tests/video_system/regex/test_youtube_regex.py

@@ -0,0 +1,40 @@
+import pytest
+
+from src.Youtube import YoutubeDownloader
+from Tests.video_system.regex.test_regex_base import TestDownloaderRegex
+
+
+class TestYoutubeRegex(TestDownloaderRegex):
+    @pytest.fixture(autouse=True)
+    def setup_class(self):
+        super().setup(YoutubeDownloader)
+
+    def test_direct_link(self):
+        assert self.check_link("https://www.youtube.com/watch?v=dQw4w9WgXcQ")
+
+    def test_direct_link_with_timestamp(self):
+        assert self.check_link("https://www.youtube.com/watch?v=dQw4w9WgXcQ&t=10s")
+
+    def test_shortened_link(self):
+        assert self.check_link("https://youtu.be/dQw4w9WgXcQ")
+
+    def test_shortened_link_with_timestamp(self):
+        assert self.check_link("https://youtu.be/dQw4w9WgXcQ?t=10")
+
+    def test_share_parameter_link(self):
+        assert self.check_link("https://youtu.be/dQw4w9WgXcQ?si=T3vbuTjMmskB3x3d")
+
+    def test_extra_text_link(self):
+        assert self.check_link(
+            "Check this out: https://www.youtube.com/watch?v=dQw4w9WgXcQ"
+        )
+
+    def test_invalid_link(self):
+        assert not self.check_link("https://www.youtube.com")
+
+    def test_invalid_shortened_link(self):
+        assert not self.check_link("https://youtu.be")
+
+    def test_different_type_links(self):
+        assert not self.check_link("https://www.instagram.com/p/CMJ1J1JjJjJ/")
+        assert not self.check_link("https://x.com/MIT_CSAIL/status/1363172815315214336")

Tests/video_system/twitter/test_twitter.py

@@ -0,0 +1,53 @@
+import os
+import shutil
+
+import pytest
+
+from Tests.video_system.download_tester import DownloadTester
+from src.twitter import TwitterDownloader
+
+DOWNLOAD_PATH = os.path.join("Tests", "video_system", "twitter", "downloads")
+SHOULD_BE_FILES_PATH = os.path.join("Tests", "video_system", "twitter")
+
+# no lang=en
+TEST_TWEET_1 = "https://x.com/MIT_CSAIL/status/1363172815315214336"
+# lang=en added
+TEST_TWEET_2 = "https://x.com/MIT_CSAIL/status/1363172815315214336?lang=en"
+# multiple videos
+TEST_TWEET_3 = "https://x.com/kytpbs1/status/1820227127532220806"
+
+
+class TestTwitterDownloader(DownloadTester):
+    @pytest.fixture(autouse=True)
+    def remove_all_cache(self):
+        os.makedirs(DOWNLOAD_PATH, exist_ok=True)
+        shutil.rmtree(DOWNLOAD_PATH, ignore_errors=True)
+        yield
+        # delete the downloads folder after the test.
+        assert os.path.exists(DOWNLOAD_PATH)
+        shutil.rmtree(DOWNLOAD_PATH, ignore_errors=True)
+
+    def test_twitter_single_video_download(self):
+        videos = TwitterDownloader.download_video_from_link(
+            TEST_TWEET_1, path=DOWNLOAD_PATH
+        )
+        should_be_path = os.path.join(SHOULD_BE_FILES_PATH, "should_be_0.mp4")
+
+        self.download_single_video_test(videos, should_be_path)
+
+    def test_twitter_download_extra_params(self):
+        videos = TwitterDownloader.download_video_from_link(
+            TEST_TWEET_2, path=DOWNLOAD_PATH
+        )
+        should_be_path = os.path.join(SHOULD_BE_FILES_PATH, "should_be_0.mp4")
+
+        self.download_single_video_test(videos, should_be_path)
+
+    def test_twitter_download_multiple_videos(self):
+        videos = TwitterDownloader.download_video_from_link(
+            TEST_TWEET_3, path=DOWNLOAD_PATH
+        )
+
+        expected_files = [os.path.join(SHOULD_BE_FILES_PATH, should_be) for should_be in ["should_be_1.mp4", "should_be_2.mp4"]]
+
+        self.download_multiple_video_test(videos, expected_files)