[DEV] Helper test to reproduce from user logs easier (#1102)

docs/source/guides/development/index.rst

@@ -62,3 +62,19 @@ is highly recommended.
 
 TODO: screenshots of configuration
 
+Debugging
+---------
+
+Debug Logs
+^^^^^^^^^^^^^^^
+Run with ``--log-level debug`` to show all debug logs when running ytdl-sub.
+
+
+Reproducing a Failing Subscription
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Subscriptions will dump their entire *compiled* yaml at the beginning of exeuction
+when using ``--log-level debug``. This can be copy-pasted into the file
+``resources/file_fixtures/repro.yaml``.
+
+Running the test ``e2e.test_debug_repro.TestReproduce.test_debug_log_repro``
+will fully reproduce that subscription in order to debug it.

tests/e2e/test_debug_repro.py

@@ -0,0 +1,69 @@
+import json
+from typing import Dict
+
+import pytest
+import yaml
+from resources import file_fixture_path
+
+from ytdl_sub.subscriptions.subscription import Subscription
+
+
+def debug_yaml_to_preset_dict(debug_subscription_dict: Dict, output_directory: str) -> Dict:
+    presets_level = debug_subscription_dict["presets"]
+    assert isinstance(presets_level, dict) and len(presets_level) == 1
+    subscription = list(presets_level.values())[0]
+    del subscription["preset"]
+    subscription["output_options"]["output_directory"] = output_directory
+    return subscription
+
+
+@pytest.fixture
+def debug_log_rerpo(output_directory):
+    with open(file_fixture_path("repro.yaml"), "r", encoding="utf-8") as yaml_file:
+        yaml_dict = yaml.safe_load(yaml_file)
+    return debug_yaml_to_preset_dict(
+        debug_subscription_dict=yaml_dict,
+        output_directory=output_directory,
+    )
+
+
+@pytest.fixture
+def subscription_yaml_preset(default_config, output_directory):
+    subscriptions = Subscription.from_file_path(
+        config=default_config, subscription_path=file_fixture_path("repro.yaml")
+    )
+    subscription_dict = yaml.safe_load(subscriptions[0].as_yaml())
+    return debug_yaml_to_preset_dict(
+        debug_subscription_dict=subscription_dict,
+        output_directory=output_directory,
+    )
+
+
+@pytest.mark.skipif(True, reason="Always skip repro, for local testing only")
+class TestReproduce:
+    def test_debug_log_repro(
+        self,
+        default_config,
+        repro_preset_dict,
+        output_directory,
+    ):
+        sub = Subscription.from_dict(
+            config=default_config,
+            preset_name="repro",
+            preset_dict=repro_preset_dict,
+        )
+
+        transaction_log = sub.download(dry_run=False)
+        assert not transaction_log.is_empty
+
+    def test_subscription_yaml_repro(
+        self, default_config, subscription_yaml_preset, output_directory
+    ):
+        sub = Subscription.from_dict(
+            config=default_config,
+            preset_name="repro",
+            preset_dict=subscription_yaml_preset,
+        )
+
+        transaction_log = sub.download(dry_run=False)
+        assert not transaction_log.is_empty

tests/resources/file_fixtures/repro.yaml

@@ -0,0 +1,42 @@
+__preset__:
+  overrides:
+    music_directory: "hi"
+
+YouTube Full Albums:
+
+  = Eastern Bloc:
+    "Eastern Bloc":
+      output_options:
+        migrated_download_archive_name: ".Eastern Bloc-download-archive.json"
+      ytdl_options:
+        max_downloads: 3
+      overrides:
+        title_capture_regex_array:
+          - "^(.*) - (.*) \\|\\|.*"  # Captures artist, album from 'artist - album ||...'
+          - "^(.*) - (.*) \\[.*"  # Captures artist, ablum from 'artist - album [...'
+          - "^(.*) - (.*)"  # Captures artist, ablum from 'artist - album'
+        title_capture: >-
+          {
+            %regex_capture_many( title, title_capture_regex_array )
+          }
+
+        description_capture_regex_array:
+          - '{%unescape("Genre:\s*(.*)\s*\n(?:Rec.+|Year):\s*.*(\d\{4\})\s*\n")}'  # Captures genre and recorded year
+        description_capture: >-
+          {
+            %regex_capture_many( description, description_capture_regex_array)
+          }
+
+        chapter_title_regex_array:
+          - "^(?:\\d+\\.\\s*|)(.*)"  # Captures title from '1. title'
+        chapter_title_capture: >-
+          {
+            %regex_capture_many( chapter_title, chapter_title_regex_array, [chapter_title] )
+          }
+
+        url: "https://www.youtube.com/@heavymetalofeasternbloc"
+        track_artist: "{%array_at(title_capture, 1)}"
+        track_album: "{%array_at(title_capture, 2)}"
+        track_title: "{%array_at(chapter_title_capture, 1)}"
+        track_year: "{%array_at(description_capture, 1)}"
+