working test

src/ytdl_sub/downloaders/youtube/split_video.py

@@ -25,9 +25,13 @@ from ytdl_sub.validators.validators import StringValidator
 _SPLIT_TIMESTAMP_REGEX = re.compile(r"^((?:\d\d:)?(?:\d:)?(?:\d)?\d:\d\d) (.+)$")
 
 
-def _parse_split_timestamp_file(split_timestamp_path: str) -> List[Tuple[str, str]]:
+def _split_video_uid(source_uid: str, idx: int) -> str:
+    return f"{source_uid}___{idx}"
+
+
+def _parse_split_timestamp_file(split_timestamp_path: str) -> Tuple[List[str], List[str]]:
     """
-    Returns list of (00:00:00 title), each timestamp as HH:MM:SS
+    Returns two lists, one containing timestamps in HH:MM:SS format, and the other titles
     """
     if not os.path.isfile(split_timestamp_path):
         raise ValidationException(
@@ -37,7 +41,8 @@ def _parse_split_timestamp_file(split_timestamp_path: str) -> List[Tuple[str, st
     with open(split_timestamp_path, "r", encoding="utf-8") as file:
         lines = file.readlines()
 
-    timestamp_titles: List[Tuple[str, str]] = []
+    timestamps: List[str] = []
+    titles: List[str] = []
     idx = 0
     for idx, line in enumerate(lines):
         match = _SPLIT_TIMESTAMP_REGEX.match(line)
@@ -57,7 +62,8 @@ def _parse_split_timestamp_file(split_timestamp_path: str) -> List[Tuple[str, st
                 pass
 
         assert len(timestamp) == 8
-        timestamp_titles.append((timestamp, title))
+        timestamps.append(timestamp)
+        titles.append(title)
 
     if idx not in (len(lines) - 1, len(lines) - 2):
         raise ValidationException(
@@ -65,7 +71,20 @@ def _parse_split_timestamp_file(split_timestamp_path: str) -> List[Tuple[str, st
             f"Each line must be formatted as '0:00 title' - a timestamp, space, then title."
         )
 
-    return timestamp_titles
+    return timestamps, titles
+
+
+def _split_video_ffmpeg_cmd(
+    input_file: str, output_file: str, timestamps: List[str], idx: int
+) -> List[str]:
+    timestamp_begin = timestamps[idx]
+    timestamp_end = timestamps[idx + 1] if idx + 1 < len(timestamps) else ""
+
+    cmd = ["-i", input_file, "-ss", timestamp_begin]
+    if timestamp_end:
+        cmd += ["-to", timestamp_end]
+    cmd += ["-vcodec", "copy", "-acodec", "copy", output_file]
+    return cmd
 
 
 ###############################################################################
@@ -151,11 +170,29 @@ class YoutubeSplitVideoDownloader(
             **{"break_on_existing": True},
         )
 
+    def _create_split_video_entry(
+        self, source_entry_dict: Dict, title: str, idx: int, split_video_count: int
+    ) -> YoutubePlaylistVideo:
+        """
+        Runs ffmpeg to create the split video
+        """
+        entry_dict = copy.deepcopy(source_entry_dict)
+        entry_dict["title"] = title
+        entry_dict["playlist_index"] = idx + 1
+        entry_dict["playlist_count"] = split_video_count
+        entry_dict["id"] = _split_video_uid(source_uid=entry_dict["id"], idx=idx)
+
+        # Remove track and artist since its now split
+        del entry_dict["track"]
+        del entry_dict["artist"]
+
+        return YoutubePlaylistVideo(entry_dict=entry_dict, working_directory=self.working_directory)
+
     def download(self) -> List[YoutubePlaylistVideo]:
         """Download a single Youtube video, then split it into multiple videos"""
         split_videos: List[YoutubePlaylistVideo] = []
 
-        timestamp_titles = _parse_split_timestamp_file(
+        timestamps, titles = _parse_split_timestamp_file(
             split_timestamp_path=self.download_options.split_timestamps
         )
         entry_dict = self.extract_info(url=self.download_options.video_url)
@@ -165,36 +202,32 @@ class YoutubeSplitVideoDownloader(
         # when copying it
         convert_download_thumbnail(entry=entry)
 
-        for idx, timestamp_title in enumerate(timestamp_titles):
+        for idx, title in enumerate(titles):
-            timestamp_begin, title = timestamp_title
+            new_uid = _split_video_uid(source_uid=entry.uid, idx=idx)
-            timestamp_end = timestamp_titles[idx + 1][0] if idx + 1 < len(timestamp_titles) else ""
-
-            new_uid = f"{entry.uid}___{idx}"
 
+            # Get the input/output file paths
             input_file = entry.get_download_file_path()
             output_file = str(Path(self.working_directory) / f"{new_uid}.{entry.ext}")
             output_thumbnail_file = str(
                 Path(self.working_directory) / f"{new_uid}.{entry.thumbnail_ext}"
             )
 
-            entry_dict_ = copy.deepcopy(entry_dict)
+            # Run ffmpeg to create the split the video
-            entry_dict_["title"] = title
+            FFMPEG.run(
-            entry_dict_["playlist_index"] = idx + 1
+                _split_video_ffmpeg_cmd(
-            entry_dict_["playlist_count"] = len(timestamp_titles)
+                    input_file=input_file, output_file=output_file, timestamps=timestamps, idx=idx
-            entry_dict_["id"] = new_uid
+                )
-
+            )
-            cmd = ["-i", input_file, "-ss", timestamp_begin]
+            # Copy the thumbnail
-            if timestamp_end:
-                cmd += ["-to", timestamp_end]
-            cmd += ["-vcodec", "copy", "-acodec", "copy", output_file]
-
-            FFMPEG.run(cmd)
-
             copyfile(src=entry.get_download_thumbnail_path(), dst=output_thumbnail_file)
 
+            # Format the split video as a YoutubePlaylistVideo
             split_videos.append(
-                YoutubePlaylistVideo(
+                self._create_split_video_entry(
-                    entry_dict=entry_dict_, working_directory=self.working_directory
+                    source_entry_dict=entry_dict,
+                    title=title,
+                    idx=idx,
+                    split_video_count=len(timestamps),
                 )
             )
 

tests/e2e/youtube/test_split_video.py

@@ -21,11 +21,11 @@ def config_path():
 def split_timestamps_file_path():
     timestamps = [
         "0:00 Intro\n",
-        "00:15 Part 1\n",
+        "00:10 Part 1\n",
-        "1:01 Part 2\n",
+        "0:20 Part 2\n",
-        "01:24 Part 3\n",
+        "00:30 Part 3\n",
-        "0:02:01 Part 4\n",
+        "0:00:40 Part 4\n",
-        "00:02:33 Part 5\n",
+        "00:01:01 Part 5\n",
     ]
 
     with NamedTemporaryFile(mode="w", encoding="utf-8", suffix=".txt") as tmp:
@@ -56,7 +56,8 @@ def subscription_dict(output_directory, subscription_name, split_timestamps_file
         # override the output directory with our fixture-generated dir
         "output_options": {
             "output_directory": output_directory,
-            "file_name": "{playlist_index}.{title_sanitized}.{ext}",
+            "file_name": "{channel_sanitized} - {playlist_index}-{playlist_size}"
+            ".{title_sanitized}.{ext}",
         },
         # download the worst format so it is fast
         "ytdl_options": {
@@ -89,9 +90,24 @@ def expected_single_video_download():
     # fmt: off
     return ExpectedDownload(
         expected_md5_file_hashes={
-            Path("JMC - Whale & Wasp.jpg"): "b58377dfe7c39527e1990a24b36bbd77",
+            Path('Project Zombie - 1-6.Intro.mp4'): "eaec6f50f364b13ef1a201e736ec9c05",
-            Path("JMC - Whale & Wasp.mp4"): "931a705864c57d21d6fedebed4af6bbc",
+            Path('Project Zombie - 2-6.Part 1.mp4'): "5850b19acb250cc13db36f80fa1bba5a",
-            Path("JMC - Whale & Wasp.nfo"): "6c2f085adb847c1dcc47c19514c454d8",
+            Path('Project Zombie - 3-6.Part 2.mp4'): "445d95eba437db6df284df7e1ab633e8",
+            Path('Project Zombie - 4-6.Part 3.mp4'): "2b6e7532d515c9e64ed2a33d850cf199",
+            Path('Project Zombie - 5-6.Part 4.mp4'): "842bf3c4d1fcc4c5ab110635935dac66",
+            Path('Project Zombie - 6-6.Part 5.mp4'): "238de99f00f829ab72f042b79da9a33a",
+            Path('Project Zombie - Intro.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Intro.nfo'): "ded59ac906f579312cc3cf98a57e7ea3",
+            Path('Project Zombie - Part 1.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Part 1.nfo'): "70ff5cd0092b8bc22dc4db93a824789b",
+            Path('Project Zombie - Part 2.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Part 2.nfo'): "54450c18a2cbb9d6d2ee5d0a1fb3f279",
+            Path('Project Zombie - Part 3.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Part 3.nfo'): "0effb13fc4039363a95969d1048dde57",
+            Path('Project Zombie - Part 4.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Part 4.nfo'): "74bd0d7c12105469838768a0cc323a8c",
+            Path('Project Zombie - Part 5.jpg'): "e87282e4115baa8b5c727fb4de15316d",
+            Path('Project Zombie - Part 5.nfo'): "a8cf2e77721335ea7c18e22734e7996c",
         }
     )
     # fmt: on