Use a queue+workers instead of semaphore

Looks nicer

.gitignore

@@ -15,3 +15,5 @@ tmp/
 /*.pyz
 /pyrightconfig.json
 /book
+*.mp4
+*.mkv

CHANGELOG.md

@@ -3,6 +3,22 @@ twitch-dl changelog
 
 <!-- Do not edit. This file is automatically generated from changelog.yaml.-->
 
+### [2.5.0 (2024-08-30)](https://github.com/ihabunek/twitch-dl/releases/tag/2.5.0)
+
+* Add support for HD video qualities (#163)
+
+### [2.4.0 (2024-08-30)](https://github.com/ihabunek/twitch-dl/releases/tag/2.4.0)
+
+* Add `clips --target-dir` option. Use in conjunction with `--download` to
+  specify target directory.
+* Fix a crash when downloading clips (#160)
+* Handle video URLs which contain the channel name (#162)
+* Don't stop downloading clips if one download fails
+
+### [2.3.1 (2024-05-19)](https://github.com/ihabunek/twitch-dl/releases/tag/2.3.1)
+
+* Fix fetching access token (#155, thanks @KryptonicDragon)
+
 ### [2.3.0 (2024-04-27)](https://github.com/ihabunek/twitch-dl/releases/tag/2.3.0)
 
 * Show more playlist data when choosing quality

changelog.yaml

@@ -1,3 +1,21 @@
+2.5.0:
+  date: 2024-08-30
+  changes:
+    - "Add support for HD video qualities (#163)"
+
+2.4.0:
+  date: 2024-08-30
+  changes:
+    - "Add `clips --target-dir` option. Use in conjunction with `--download` to specify target directory."
+    - "Fix a crash when downloading clips (#160)"
+    - "Handle video URLs which contain the channel name (#162)"
+    - "Don't stop downloading clips if one download fails"
+
+2.3.1:
+  date: 2024-05-19
+  changes:
+    - "Fix fetching access token (#155, thanks @KryptonicDragon)"
+
 2.3.0:
   date: 2024-04-27
   changes:

docs/changelog.md

@@ -3,6 +3,22 @@ twitch-dl changelog
 
 <!-- Do not edit. This file is automatically generated from changelog.yaml.-->
 
+### [2.5.0 (2024-08-30)](https://github.com/ihabunek/twitch-dl/releases/tag/2.5.0)
+
+* Add support for HD video qualities (#163)
+
+### [2.4.0 (2024-08-30)](https://github.com/ihabunek/twitch-dl/releases/tag/2.4.0)
+
+* Add `clips --target-dir` option. Use in conjunction with `--download` to
+  specify target directory.
+* Fix a crash when downloading clips (#160)
+* Handle video URLs which contain the channel name (#162)
+* Don't stop downloading clips if one download fails
+
+### [2.3.1 (2024-05-19)](https://github.com/ihabunek/twitch-dl/releases/tag/2.3.1)
+
+* Fix fetching access token (#155, thanks @KryptonicDragon)
+
 ### [2.3.0 (2024-04-27)](https://github.com/ihabunek/twitch-dl/releases/tag/2.3.0)
 
 * Show more playlist data when choosing quality

docs/commands/clips.md

@@ -43,6 +43,11 @@ twitch-dl clips [OPTIONS] CHANNEL_NAME
     <td>Period from which to return clips Possible values: <code>last_day</code>, <code>last_week</code>, <code>last_month</code>, <code>all_time</code>. [default: <code>all_time</code>]</td>
 </tr>
 
+<tr>
+    <td class="code">-t, --target-dir</td>
+    <td>Target directory when downloading clips [default: <code>.</code>]</td>
+</tr>
+
 <tr>
     <td class="code">--json</td>
     <td>Print data as JSON rather than human readable text</td>

pyproject.toml

@@ -22,7 +22,7 @@ classifiers = [
 dependencies = [
     "click>=8.0.0,<9.0.0",
     "httpx>=0.17.0,<1.0.0",
-    "m3u8>=1.0.0,<4.0.0",
+    "m3u8>=3.0.0,<7.0.0",
 ]
 
 [tool.setuptools]

tests/test_patterns.py

@@ -7,6 +7,7 @@ TEST_VIDEO_PATTERNS = [
     ("702689313", "https://twitch.tv/videos/702689313"),
     ("702689313", "https://www.twitch.tv/videos/702689313"),
     ("702689313", "https://m.twitch.tv/videos/702689313"),
+    ("2223719525", "https://www.twitch.tv/r0dn3y/video/2223719525"),
 ]
 
 TEST_CLIP_PATTERNS = {

tests/test_progress.py

@@ -8,8 +8,8 @@ def test_initial_values():
     assert progress.progress_perc == 0
     assert progress.remaining_time is None
     assert progress.speed is None
-    assert progress.vod_count == 10
-    assert progress.vod_downloaded_count == 0
+    assert progress.file_count == 10
+    assert progress.downloaded_count == 0
 
 
 def test_downloaded():
@@ -23,26 +23,31 @@ def test_downloaded():
     assert progress.progress_perc == 0
 
     progress.advance(1, 100)
+    progress._recalculate()
     assert progress.downloaded == 100
     assert progress.progress_bytes == 100
     assert progress.progress_perc == 11
 
     progress.advance(2, 200)
+    progress._recalculate()
     assert progress.downloaded == 300
     assert progress.progress_bytes == 300
     assert progress.progress_perc == 33
 
     progress.advance(3, 150)
+    progress._recalculate()
     assert progress.downloaded == 450
     assert progress.progress_bytes == 450
     assert progress.progress_perc == 50
 
     progress.advance(1, 50)
+    progress._recalculate()
     assert progress.downloaded == 500
     assert progress.progress_bytes == 500
     assert progress.progress_perc == 55
 
     progress.abort(2)
+    progress._recalculate()
     assert progress.downloaded == 500
     assert progress.progress_bytes == 300
     assert progress.progress_perc == 33
@@ -52,6 +57,7 @@ def test_downloaded():
     progress.advance(1, 150)
     progress.advance(2, 300)
     progress.advance(3, 150)
+    progress._recalculate()
 
     assert progress.downloaded == 1100
     assert progress.progress_bytes == 900
@@ -71,12 +77,15 @@ def test_estimated_total():
     assert progress.estimated_total is None
 
     progress.start(1, 12000)
+    progress._recalculate()
     assert progress.estimated_total == 12000 * 3
 
     progress.start(2, 11000)
+    progress._recalculate()
     assert progress.estimated_total == 11500 * 3
 
     progress.start(3, 10000)
+    progress._recalculate()
     assert progress.estimated_total == 11000 * 3
 
 
@@ -87,16 +96,16 @@ def test_vod_downloaded_count():
     progress.start(2, 100)
     progress.start(3, 100)
 
-    assert progress.vod_downloaded_count == 0
+    assert progress.downloaded_count == 0
 
     progress.advance(1, 100)
     progress.end(1)
-    assert progress.vod_downloaded_count == 1
+    assert progress.downloaded_count == 1
 
     progress.advance(2, 100)
     progress.end(2)
-    assert progress.vod_downloaded_count == 2
+    assert progress.downloaded_count == 2
 
     progress.advance(3, 100)
     progress.end(3)
-    assert progress.vod_downloaded_count == 3
+    assert progress.downloaded_count == 3

twitchdl/cli.py

@@ -2,12 +2,14 @@ import logging
 import platform
 import re
 import sys
+from pathlib import Path
 from typing import Optional, Tuple
 
 import click
 
 from twitchdl import __version__
 from twitchdl.entities import DownloadOptions
+from twitchdl.naming import DEFAULT_OUTPUT_TEMPLATE
 from twitchdl.twitch import ClipsPeriod, VideosSort, VideosType
 
 # Tweak the Click context
@@ -79,11 +81,12 @@ def validate_rate(_ctx: click.Context, _param: click.Parameter, value: str) -> O
 
 
 @click.group(context_settings=CONTEXT)
-@click.option("--debug/--no-debug", default=False, help="Log debug info to stderr")
+@click.option("--debug/--no-debug", default=False, help="Enable debug logging to stderr")
+@click.option("--verbose/--no-verbose", default=False, help="More verbose debug logging")
 @click.option("--color/--no-color", default=sys.stdout.isatty(), help="Use ANSI color in output")
 @click.version_option(package_name="twitch-dl")
 @click.pass_context
-def cli(ctx: click.Context, color: bool, debug: bool):
+def cli(ctx: click.Context, color: bool, debug: bool, verbose: bool):
     """twitch-dl - twitch.tv downloader
 
     https://twitch-dl.bezdomni.net/
@@ -91,7 +94,7 @@ def cli(ctx: click.Context, color: bool, debug: bool):
     ctx.color = color
 
     if debug:
-        logging.basicConfig(level=logging.DEBUG)
+        logging.basicConfig(level=logging.DEBUG if verbose else logging.INFO)
         logging.getLogger("httpx").setLevel(logging.WARN)
         logging.getLogger("httpcore").setLevel(logging.WARN)
 
@@ -139,6 +142,18 @@ def cli(ctx: click.Context, color: bool, debug: bool):
     default="all_time",
     type=click.Choice(["last_day", "last_week", "last_month", "all_time"]),
 )
+@click.option(
+    "-t",
+    "--target-dir",
+    help="Target directory when downloading clips",
+    type=click.Path(
+        file_okay=False,
+        readable=False,
+        writable=True,
+        path_type=Path,
+    ),
+    default=Path(),
+)
 @json_option
 def clips(
     channel_name: str,
@@ -149,10 +164,14 @@ def clips(
     limit: Optional[int],
     pager: Optional[int],
     period: ClipsPeriod,
+    target_dir: Path,
 ):
     """List or download clips for given CHANNEL_NAME."""
     from twitchdl.commands.clips import clips
 
+    if not target_dir.exists():
+        target_dir.mkdir(parents=True, exist_ok=True)
+
     clips(
         channel_name,
         all=all,
@@ -162,6 +181,7 @@ def clips(
         limit=limit,
         pager=pager,
         period=period,
+        target_dir=target_dir,
     )
 
 
@@ -229,7 +249,7 @@ def clips(
     "-o",
     "--output",
     help="Output file name template. See docs for details.",
-    default="{date}_{id}_{channel_login}_{title_slug}.{format}",
+    default=DEFAULT_OUTPUT_TEMPLATE,
 )
 @click.option(
     "-q",

twitchdl/commands/clips.py

@@ -1,13 +1,15 @@
 import re
 import sys
 from os import path
-from typing import Callable, Generator, Optional
+from pathlib import Path
+from typing import Callable, Generator, List, Optional
 
 import click
 
 from twitchdl import twitch, utils
 from twitchdl.commands.download import get_clip_authenticated_url
-from twitchdl.download import download_file
+from twitchdl.entities import VideoQuality
+from twitchdl.http import download_file
 from twitchdl.output import green, print_clip, print_clip_compact, print_json, print_paged, yellow
 from twitchdl.twitch import Clip, ClipsPeriod
 
@@ -22,6 +24,7 @@ def clips(
     limit: Optional[int] = None,
     pager: Optional[int] = None,
     period: ClipsPeriod = "all_time",
+    target_dir: Path = Path(),
 ):
     # Set different defaults for limit for compact display
     default_limit = 40 if compact else 10
@@ -35,7 +38,7 @@ def clips(
         return print_json(list(generator))
 
     if download:
-        return _download_clips(generator)
+        return _download_clips(target_dir, generator)
 
     print_fn = print_clip_compact if compact else print_clip
 
@@ -45,8 +48,8 @@ def clips(
     return _print_all(generator, print_fn, all)
 
 
-def _target_filename(clip: Clip):
-    url = clip["videoQualities"][0]["sourceURL"]
+def _target_filename(clip: Clip, video_qualities: List[VideoQuality]):
+    url = video_qualities[0]["sourceURL"]
     _, ext = path.splitext(url)
     ext = ext.lstrip(".")
 
@@ -67,16 +70,27 @@ def _target_filename(clip: Clip):
     return f"{name}.{ext}"
 
 
-def _download_clips(generator: Generator[Clip, None, None]):
-    for clip in generator:
-        target = _target_filename(clip)
+def _download_clips(target_dir: Path, generator: Generator[Clip, None, None]):
+    if not target_dir.exists():
+        target_dir.mkdir(parents=True, exist_ok=True)
 
-        if path.exists(target):
+    for clip in generator:
+        # videoQualities can be null in some circumstances, see:
+        # https://github.com/ihabunek/twitch-dl/issues/160
+        if not clip["videoQualities"]:
+            continue
+
+        target = target_dir / _target_filename(clip, clip["videoQualities"])
+
+        if target.exists():
             click.echo(f"Already downloaded: {green(target)}")
         else:
-            url = get_clip_authenticated_url(clip["slug"], "source")
-            click.echo(f"Downloading: {yellow(target)}")
-            download_file(url, target)
+            try:
+                url = get_clip_authenticated_url(clip["slug"], "source")
+                click.echo(f"Downloading: {yellow(target)}")
+                download_file(url, target)
+            except Exception as ex:
+                click.secho(ex, err=True, fg="red")
 
 
 def _print_all(

twitchdl/commands/download.py

@@ -1,35 +1,39 @@
 import asyncio
-import os
 import platform
 import re
+import shlex
 import shutil
 import subprocess
 import tempfile
-from os import path
 from pathlib import Path
-from typing import Dict, List
+from typing import List, Optional
 from urllib.parse import urlencode, urlparse
 
 import click
 import httpx
 
 from twitchdl import twitch, utils
-from twitchdl.download import download_file
 from twitchdl.entities import DownloadOptions
 from twitchdl.exceptions import ConsoleError
-from twitchdl.http import download_all
+from twitchdl.http import download_all, download_file
+from twitchdl.naming import clip_filename, video_filename
 from twitchdl.output import blue, bold, green, print_log, yellow
 from twitchdl.playlists import (
     enumerate_vods,
+    get_init_sections,
     load_m3u8,
     make_join_playlist,
     parse_playlists,
     select_playlist,
 )
-from twitchdl.twitch import Chapter, Clip, ClipAccessToken, Video
+from twitchdl.twitch import Chapter, ClipAccessToken, Video
 
 
 def download(ids: List[str], args: DownloadOptions):
+    if not ids:
+        print_log("No IDs to downlad given")
+        return
+
     for video_id in ids:
         download_one(video_id, args)
 
@@ -46,14 +50,14 @@ def download_one(video: str, args: DownloadOptions):
     raise ConsoleError(f"Invalid input: {video}")
 
 
-def _join_vods(playlist_path: str, target: str, overwrite: bool, video: Video):
+def _join_vods(playlist_path: Path, target: Path, overwrite: bool, video: Video):
     description = video["description"] or ""
     description = description.strip()
 
-    command = [
+    command: List[str] = [
         "ffmpeg",
         "-i",
-        playlist_path,
+        str(playlist_path),
         "-c",
         "copy",
         "-metadata",
@@ -73,15 +77,15 @@ def _join_vods(playlist_path: str, target: str, overwrite: bool, video: Video):
     if overwrite:
         command.append("-y")
 
-    click.secho(f"{' '.join(command)}", dim=True)
+    click.secho(f"{shlex.join(command)}", dim=True)
     result = subprocess.run(command)
     if result.returncode != 0:
         raise ConsoleError("Joining files failed")
 
 
-def _concat_vods(vod_paths: List[str], target: str):
+def _concat_vods(vod_paths: List[Path], target: Path):
     tool = "type" if platform.system() == "Windows" else "cat"
-    command = [tool] + vod_paths
+    command = [tool] + [str(p) for p in vod_paths]
 
     with open(target, "wb") as target_file:
         result = subprocess.run(command, stdout=target_file)
@@ -89,74 +93,15 @@ def _concat_vods(vod_paths: List[str], target: str):
             raise ConsoleError(f"Joining files failed: {result.stderr}")
 
 
-def get_video_placeholders(video: Video, format: str) -> Dict[str, str]:
-    date, time = video["publishedAt"].split("T")
-    game = video["game"]["name"] if video["game"] else "Unknown"
-
-    return {
-        "channel": video["creator"]["displayName"],
-        "channel_login": video["creator"]["login"],
-        "date": date,
-        "datetime": video["publishedAt"],
-        "format": format,
-        "game": game,
-        "game_slug": utils.slugify(game),
-        "id": video["id"],
-        "time": time,
-        "title": utils.titlify(video["title"]),
-        "title_slug": utils.slugify(video["title"]),
-    }
-
-
-def _video_target_filename(video: Video, args: DownloadOptions):
-    subs = get_video_placeholders(video, args.format)
-
-    try:
-        return args.output.format(**subs)
-    except KeyError as e:
-        supported = ", ".join(subs.keys())
-        raise ConsoleError(f"Invalid key {e} used in --output. Supported keys are: {supported}")
-
-
-def _clip_target_filename(clip: Clip, args: DownloadOptions):
-    date, time = clip["createdAt"].split("T")
-    game = clip["game"]["name"] if clip["game"] else "Unknown"
-
-    url = clip["videoQualities"][0]["sourceURL"]
-    _, ext = path.splitext(url)
-    ext = ext.lstrip(".")
-
-    subs = {
-        "channel": clip["broadcaster"]["displayName"],
-        "channel_login": clip["broadcaster"]["login"],
-        "date": date,
-        "datetime": clip["createdAt"],
-        "format": ext,
-        "game": game,
-        "game_slug": utils.slugify(game),
-        "id": clip["id"],
-        "slug": clip["slug"],
-        "time": time,
-        "title": utils.titlify(clip["title"]),
-        "title_slug": utils.slugify(clip["title"]),
-    }
-
-    try:
-        return args.output.format(**subs)
-    except KeyError as e:
-        supported = ", ".join(subs.keys())
-        raise ConsoleError(f"Invalid key {e} used in --output. Supported keys are: {supported}")
-
-
-def _crete_temp_dir(base_uri: str) -> str:
+def _crete_temp_dir(base_uri: str) -> Path:
     """Create a temp dir to store downloads if it doesn't exist."""
     path = urlparse(base_uri).path.lstrip("/")
     temp_dir = Path(tempfile.gettempdir(), "twitch-dl", path)
     temp_dir.mkdir(parents=True, exist_ok=True)
-    return str(temp_dir)
+    return temp_dir
 
 
-def _get_clip_url(access_token: ClipAccessToken, quality: str) -> str:
+def _get_clip_url(access_token: ClipAccessToken, quality: Optional[str]) -> str:
     qualities = access_token["videoQualities"]
 
     # Quality given as an argument
@@ -184,7 +129,7 @@ def _get_clip_url(access_token: ClipAccessToken, quality: str) -> str:
     return selected_quality["sourceURL"]
 
 
-def get_clip_authenticated_url(slug: str, quality: str):
+def get_clip_authenticated_url(slug: str, quality: Optional[str]):
     print_log("Fetching access token...")
     access_token = twitch.get_clip_access_token(slug)
 
@@ -216,10 +161,10 @@ def _download_clip(slug: str, args: DownloadOptions) -> None:
     duration = utils.format_duration(clip["durationSeconds"])
     click.echo(f"Found: {green(title)} by {yellow(user)}, playing {blue(game)} ({duration})")
 
-    target = _clip_target_filename(clip, args)
+    target = Path(clip_filename(clip, args.output))
     click.echo(f"Target: {blue(target)}")
 
-    if not args.overwrite and path.exists(target):
+    if not args.overwrite and target.exists():
         response = click.prompt("File exists. Overwrite? [Y/n]", default="Y", show_default=False)
         if response.lower().strip() != "y":
             raise click.Abort()
@@ -248,10 +193,10 @@ def _download_video(video_id: str, args: DownloadOptions) -> None:
 
     click.echo(f"Found: {blue(video['title'])} by {yellow(video['creator']['displayName'])}")
 
-    target = _video_target_filename(video, args)
+    target = Path(video_filename(video, args.format, args.output))
     click.echo(f"Output: {blue(target)}")
 
-    if not args.overwrite and path.exists(target):
+    if not args.overwrite and target.exists():
         response = click.prompt("File exists. Overwrite? [Y/n]", default="Y", show_default=False)
         if response.lower().strip() != "y":
             raise click.Abort()
@@ -281,19 +226,32 @@ def _download_video(video_id: str, args: DownloadOptions) -> None:
     target_dir = _crete_temp_dir(base_uri)
 
     # Save playlists for debugging purposes
-    with open(path.join(target_dir, "playlists.m3u8"), "w") as f:
+    with open(target_dir / "playlists.m3u8", "w") as f:
         f.write(playlists_text)
-    with open(path.join(target_dir, "playlist.m3u8"), "w") as f:
+    with open(target_dir / "playlist.m3u8", "w") as f:
         f.write(vods_text)
 
-    click.echo(f"\nDownloading {len(vods)} VODs using {args.max_workers} workers to {target_dir}")
+    init_sections = get_init_sections(vods_m3u8)
+    for uri in init_sections:
+        print_log(f"Downloading init section {uri}...")
+        download_file(f"{base_uri}{uri}", target_dir / uri)
+
+    print_log(f"Downloading {len(vods)} VODs using {args.max_workers} workers to {target_dir}")
 
     sources = [base_uri + vod.path for vod in vods]
-    targets = [os.path.join(target_dir, f"{vod.index:05d}.ts") for vod in vods]
-    asyncio.run(download_all(sources, targets, args.max_workers, rate_limit=args.rate_limit))
+    targets = [target_dir / f"{vod.index:05d}.ts" for vod in vods]
+
+    asyncio.run(
+        download_all(
+            zip(sources, targets),
+            args.max_workers,
+            rate_limit=args.rate_limit,
+            count=len(vods),
+        )
+    )
 
     join_playlist = make_join_playlist(vods_m3u8, vods, targets)
-    join_playlist_path = path.join(target_dir, "playlist_downloaded.m3u8")
+    join_playlist_path = target_dir / "playlist_downloaded.m3u8"
     join_playlist.dump(join_playlist_path)  # type: ignore
     click.echo()
 
@@ -312,12 +270,12 @@ def _download_video(video_id: str, args: DownloadOptions) -> None:
     click.echo()
 
     if args.keep:
-        click.echo(f"Temporary files not deleted: {target_dir}")
+        click.echo(f"Temporary files not deleted: {yellow(target_dir)}")
     else:
         print_log("Deleting temporary files...")
         shutil.rmtree(target_dir)
 
-    click.echo(f"\nDownloaded: {green(target)}")
+    click.echo(f"Downloaded: {green(target)}")
 
 
 def http_get(url: str) -> str:

twitchdl/commands/info.py

@@ -4,9 +4,9 @@ import click
 import m3u8
 
 from twitchdl import twitch, utils
-from twitchdl.commands.download import get_video_placeholders
 from twitchdl.exceptions import ConsoleError
-from twitchdl.output import bold, print_clip, print_json, print_log, print_table, print_video
+from twitchdl.naming import video_placeholders
+from twitchdl.output import bold, dim, print_clip, print_json, print_log, print_table, print_video
 from twitchdl.playlists import parse_playlists
 from twitchdl.twitch import Chapter, Clip, Video
 
@@ -55,9 +55,19 @@ def video_info(video: Video, playlists: str, chapters: List[Chapter]):
     click.echo()
     print_video(video)
 
-    click.echo("Playlists:")
-    for p in parse_playlists(playlists):
-        click.echo(f"{bold(p.name)} {p.url}")
+    click.echo("Playlists:\n")
+
+    playlist_headers = ["Name", "Group", "Resolution", "URL"]
+    playlist_data = [
+        [
+            f"{p.name} {dim('source')}" if p.is_source else p.name,
+            p.group_id,
+            f"{p.resolution}",
+            p.url,
+        ]
+        for p in parse_playlists(playlists)
+    ]
+    print_table(playlist_headers, playlist_data)
 
     if chapters:
         click.echo()
@@ -67,7 +77,7 @@ def video_info(video: Video, playlists: str, chapters: List[Chapter]):
             duration = utils.format_time(chapter["durationMilliseconds"] // 1000)
             click.echo(f'{start} {bold(chapter["description"])} ({duration})')
 
-    placeholders = get_video_placeholders(video, format="mkv")
+    placeholders = video_placeholders(video, format="mkv")
     placeholders = [[f"{{{k}}}", v] for k, v in placeholders.items()]
     click.echo("")
     print_table(["Placeholder", "Value"], placeholders)
@@ -98,5 +108,8 @@ def clip_info(clip: Clip):
     click.echo()
     click.echo("Download links:")
 
-    for q in clip["videoQualities"]:
-        click.echo(f"{bold(q['quality'])} [{q['frameRate']} fps] {q['sourceURL']}")
+    if clip["videoQualities"]:
+        for q in clip["videoQualities"]:
+            click.echo(f"{bold(q['quality'])} [{q['frameRate']} fps] {q['sourceURL']}")
+    else:
+        click.echo("No download URLs found")

twitchdl/download.py

@@ -1,37 +0,0 @@
-import os
-
-import httpx
-
-from twitchdl.exceptions import ConsoleError
-
-CHUNK_SIZE = 1024
-CONNECT_TIMEOUT = 5
-RETRY_COUNT = 5
-
-
-def _download(url: str, path: str):
-    tmp_path = path + ".tmp"
-    size = 0
-    with httpx.stream("GET", url, timeout=CONNECT_TIMEOUT) as response:
-        with open(tmp_path, "wb") as target:
-            for chunk in response.iter_bytes(chunk_size=CHUNK_SIZE):
-                target.write(chunk)
-                size += len(chunk)
-
-    os.rename(tmp_path, path)
-    return size
-
-
-def download_file(url: str, path: str, retries: int = RETRY_COUNT):
-    if os.path.exists(path):
-        from_disk = True
-        return (os.path.getsize(path), from_disk)
-
-    from_disk = False
-    for _ in range(retries):
-        try:
-            return (_download(url, path), from_disk)
-        except httpx.RequestError:
-            pass
-
-    raise ConsoleError(f"Failed downloading after {retries} attempts: {url}")

twitchdl/entities.py

@@ -1,5 +1,5 @@
 from dataclasses import dataclass
-from typing import Any, Mapping, Optional
+from typing import Any, List, Literal, Mapping, Optional, TypedDict
 
 
 @dataclass
@@ -20,6 +20,73 @@ class DownloadOptions:
     max_workers: int
 
 
+ClipsPeriod = Literal["last_day", "last_week", "last_month", "all_time"]
+VideosSort = Literal["views", "time"]
+VideosType = Literal["archive", "highlight", "upload"]
+
+
+class AccessToken(TypedDict):
+    signature: str
+    value: str
+
+
+class User(TypedDict):
+    login: str
+    displayName: str
+
+
+class Game(TypedDict):
+    id: str
+    name: str
+
+
+class VideoQuality(TypedDict):
+    frameRate: str
+    quality: str
+    sourceURL: str
+
+
+class ClipAccessToken(TypedDict):
+    id: str
+    playbackAccessToken: AccessToken
+    videoQualities: List[VideoQuality]
+
+
+class Clip(TypedDict):
+    id: str
+    slug: str
+    title: str
+    createdAt: str
+    viewCount: int
+    durationSeconds: int
+    url: str
+    videoQualities: Optional[List[VideoQuality]]
+    game: Game
+    broadcaster: User
+
+
+class Video(TypedDict):
+    id: str
+    title: str
+    description: str
+    publishedAt: str
+    broadcastType: str
+    lengthSeconds: int
+    game: Game
+    creator: User
+
+
+class Chapter(TypedDict):
+    id: str
+    durationMilliseconds: int
+    positionMilliseconds: int
+    type: str
+    description: str
+    subDescription: str
+    thumbnailURL: str
+    game: Game
+
+
 # Type for annotating decoded JSON
 # TODO: make data classes for common structs
 Data = Mapping[str, Any]

twitchdl/http.py

@@ -3,10 +3,12 @@ import logging
 import os
 import time
 from abc import ABC, abstractmethod
-from typing import List, Optional
+from pathlib import Path
+from typing import Iterable, NamedTuple, Optional, Tuple
 
 import httpx
 
+from twitchdl.exceptions import ConsoleError
 from twitchdl.progress import Progress
 
 logger = logging.getLogger(__name__)
@@ -71,7 +73,7 @@ async def download(
     client: httpx.AsyncClient,
     task_id: int,
     source: str,
-    target: str,
+    target: Path,
     progress: Progress,
     token_bucket: TokenBucket,
 ):
@@ -93,52 +95,108 @@ async def download(
 
 async def download_with_retries(
     client: httpx.AsyncClient,
-    semaphore: asyncio.Semaphore,
     task_id: int,
     source: str,
-    target: str,
+    target: Path,
     progress: Progress,
     token_bucket: TokenBucket,
 ):
-    async with semaphore:
-        if os.path.exists(target):
-            size = os.path.getsize(target)
-            progress.already_downloaded(task_id, size)
-            return
+    if target.exists():
+        size = os.path.getsize(target)
+        progress.already_downloaded(task_id, size)
+        return
 
-        for n in range(RETRY_COUNT):
-            try:
-                return await download(client, task_id, source, target, progress, token_bucket)
-            except httpx.RequestError:
-                logger.exception("Task {task_id} failed. Retrying. Maybe.")
-                progress.abort(task_id)
-                if n + 1 >= RETRY_COUNT:
-                    raise
+    for n in range(RETRY_COUNT):
+        try:
+            return await download(client, task_id, source, target, progress, token_bucket)
+        except httpx.RequestError:
+            logger.exception("Task {task_id} failed. Retrying. Maybe.")
+            progress.abort(task_id)
+            if n + 1 >= RETRY_COUNT:
+                raise
 
-        raise Exception("Should not happen")
+    raise Exception("Should not happen")
+
+
+class QueueItem(NamedTuple):
+    task_id: int
+    url: str
+    target: Path
+
+
+async def download_worker(
+    queue: asyncio.Queue[QueueItem],
+    client: httpx.AsyncClient,
+    progress: Progress,
+    token_bucket: TokenBucket,
+):
+    while True:
+        item = await queue.get()
+        await download_with_retries(
+            client,
+            item.task_id,
+            item.url,
+            item.target,
+            progress,
+            token_bucket,
+        )
+        queue.task_done()
 
 
 async def download_all(
-    sources: List[str],
-    targets: List[str],
-    workers: int,
+    source_targets: Iterable[Tuple[str, Path]],
+    worker_count: int,
     *,
+    count: Optional[int] = None,
     rate_limit: Optional[int] = None,
 ):
-    progress = Progress(len(sources))
+    progress = Progress(count)
     token_bucket = LimitingTokenBucket(rate_limit) if rate_limit else EndlessTokenBucket()
+    queue: asyncio.Queue[QueueItem] = asyncio.Queue()
+
     async with httpx.AsyncClient(timeout=TIMEOUT) as client:
-        semaphore = asyncio.Semaphore(workers)
         tasks = [
-            download_with_retries(
-                client,
-                semaphore,
-                task_id,
-                source,
-                target,
-                progress,
-                token_bucket,
-            )
-            for task_id, (source, target) in enumerate(zip(sources, targets))
+            asyncio.create_task(download_worker(queue, client, progress, token_bucket))
+            for _ in range(worker_count)
         ]
-        await asyncio.gather(*tasks)
+
+        for index, (source, target) in enumerate(source_targets):
+            await queue.put(QueueItem(index, source, target))
+
+        # Wait for queue to deplete
+        await queue.join()
+
+        # Cancel tasks and wait until they are cancelled
+        for task in tasks:
+            task.cancel()
+        await asyncio.gather(*tasks, return_exceptions=True)
+
+
+def download_file(url: str, target: Path, retries: int = RETRY_COUNT) -> None:
+    """Download URL to given target path with retries"""
+    error_message = ""
+    for r in range(retries):
+        try:
+            retry_info = f" (retry {r})" if r > 0 else ""
+            logger.info(f"Downloading {url} to {target}{retry_info}")
+            return _do_download_file(url, target)
+        except httpx.HTTPStatusError as ex:
+            logger.error(ex)
+            error_message = f"Server responded with HTTP {ex.response.status_code}"
+        except httpx.RequestError as ex:
+            logger.error(ex)
+            error_message = str(ex)
+
+    raise ConsoleError(f"Failed downloading after {retries} attempts: {error_message}")
+
+
+def _do_download_file(url: str, target: Path) -> None:
+    tmp_path = Path(str(target) + ".tmp")
+
+    with httpx.stream("GET", url, timeout=TIMEOUT, follow_redirects=True) as response:
+        response.raise_for_status()
+        with open(tmp_path, "wb") as f:
+            for chunk in response.iter_bytes(chunk_size=CHUNK_SIZE):
+                f.write(chunk)
+
+    os.rename(tmp_path, target)

twitchdl/naming.py

@@ -0,0 +1,72 @@
+import os
+from typing import Dict
+
+from twitchdl import utils
+from twitchdl.entities import Clip, Video
+from twitchdl.exceptions import ConsoleError
+
+DEFAULT_OUTPUT_TEMPLATE = "{date}_{id}_{channel_login}_{title_slug}.{format}"
+
+
+def video_filename(video: Video, format: str, output: str) -> str:
+    subs = video_placeholders(video, format)
+    return _format(output, subs)
+
+
+def video_placeholders(video: Video, format: str) -> Dict[str, str]:
+    date, time = video["publishedAt"].split("T")
+    game = video["game"]["name"] if video["game"] else "Unknown"
+
+    return {
+        "channel": video["creator"]["displayName"],
+        "channel_login": video["creator"]["login"],
+        "date": date,
+        "datetime": video["publishedAt"],
+        "format": format,
+        "game": game,
+        "game_slug": utils.slugify(game),
+        "id": video["id"],
+        "time": time,
+        "title": utils.titlify(video["title"]),
+        "title_slug": utils.slugify(video["title"]),
+    }
+
+
+def clip_filename(clip: Clip, output: str):
+    subs = clip_placeholders(clip)
+    return _format(output, subs)
+
+
+def clip_placeholders(clip: Clip) -> Dict[str, str]:
+    date, time = clip["createdAt"].split("T")
+    game = clip["game"]["name"] if clip["game"] else "Unknown"
+
+    if clip["videoQualities"]:
+        url = clip["videoQualities"][0]["sourceURL"]
+        _, ext = os.path.splitext(url)
+        ext = ext.lstrip(".")
+    else:
+        ext = "mp4"
+
+    return {
+        "channel": clip["broadcaster"]["displayName"],
+        "channel_login": clip["broadcaster"]["login"],
+        "date": date,
+        "datetime": clip["createdAt"],
+        "format": ext,
+        "game": game,
+        "game_slug": utils.slugify(game),
+        "id": clip["id"],
+        "slug": clip["slug"],
+        "time": time,
+        "title": utils.titlify(clip["title"]),
+        "title_slug": utils.slugify(clip["title"]),
+    }
+
+
+def _format(output: str, subs: Dict[str, str]) -> str:
+    try:
+        return output.format(**subs)
+    except KeyError as e:
+        supported = ", ".join(subs.keys())
+        raise ConsoleError(f"Invalid key {e} used in --output. Supported keys are: {supported}")

twitchdl/output.py

@@ -1,15 +1,21 @@
 import json
+import sys
 from itertools import islice
 from typing import Any, Callable, Generator, List, Optional, TypeVar
 
 import click
 
 from twitchdl import utils
-from twitchdl.twitch import Clip, Video
+from twitchdl.entities import Clip, Video
 
 T = TypeVar("T")
 
 
+def clear_line():
+    sys.stdout.write("\033[1K")
+    sys.stdout.write("\r")
+
+
 def truncate(string: str, length: int) -> str:
     if len(string) > length:
         return string[: length - 1] + "…"
@@ -40,11 +46,8 @@ def print_table(headers: List[str], data: List[List[str]]):
     underlines = ["-" * width for width in widths]
 
     def print_row(row: List[str]):
-        for idx, cell in enumerate(row):
-            width = widths[idx]
-            click.echo(ljust(cell, width), nl=False)
-            click.echo("  ", nl=False)
-        click.echo()
+        parts = (ljust(cell, widths[idx]) for idx, cell in enumerate(row))
+        click.echo("  ".join(parts).strip())
 
     print_row(headers)
     print_row(underlines)
@@ -102,11 +105,12 @@ def print_video(video: Video):
     if channel or playing:
         click.echo(" ".join([channel, playing]))
 
-    if video["description"]:
-        click.echo(f"Description: {video['description']}")
-
     click.echo(f"Published {blue(published_at)}  Length: {blue(length)} ")
     click.secho(url, italic=True)
+
+    if video["description"]:
+        click.echo(f"\nDescription:\n{video['description']}")
+
     click.echo()
 
 

twitchdl/playlists.py

@@ -3,7 +3,8 @@ Parse and manipulate m3u8 playlists.
 """
 
 from dataclasses import dataclass
-from typing import Generator, List, Optional, OrderedDict
+from pathlib import Path
+from typing import Generator, List, Optional, OrderedDict, Set
 
 import click
 import m3u8
@@ -81,7 +82,7 @@ def enumerate_vods(
 def make_join_playlist(
     playlist: m3u8.M3U8,
     vods: List[Vod],
-    targets: List[str],
+    targets: List[Path],
 ) -> m3u8.Playlist:
     """
     Make a modified playlist which references downloaded VODs
@@ -93,7 +94,7 @@ def make_join_playlist(
     playlist.segments.clear()
     for segment in org_segments:
         if segment.uri in path_map:
-            segment.uri = path_map[segment.uri]
+            segment.uri = str(path_map[segment.uri].name)
             playlist.segments.append(segment)
 
     return playlist
@@ -168,3 +169,12 @@ def _playlist_key(playlist: Playlist) -> int:
         pass
 
     return MAX
+
+
+def get_init_sections(playlist: m3u8.M3U8) -> Set[str]:
+    # TODO: we're ignoring initi_section.base_uri and bytes
+    return set(
+        segment.init_section.uri
+        for segment in playlist.segments
+        if segment.init_section is not None
+    )

twitchdl/progress.py

@@ -1,13 +1,13 @@
 import logging
 import time
 from collections import deque
-from dataclasses import dataclass, field
+from dataclasses import dataclass
 from statistics import mean
 from typing import Deque, Dict, NamedTuple, Optional
 
 import click
 
-from twitchdl.output import blue
+from twitchdl.output import blue, clear_line
 from twitchdl.utils import format_size, format_time
 
 logger = logging.getLogger(__name__)
@@ -31,28 +31,25 @@ class Sample(NamedTuple):
     timestamp: float
 
 
-@dataclass
 class Progress:
-    vod_count: int
-    downloaded: int = 0
-    estimated_total: Optional[int] = None
-    last_printed: float = field(default_factory=time.time)
-    progress_bytes: int = 0
-    progress_perc: int = 0
-    remaining_time: Optional[int] = None
-    speed: Optional[float] = None
-    start_time: float = field(default_factory=time.time)
-    tasks: Dict[TaskId, Task] = field(default_factory=dict)
-    vod_downloaded_count: int = 0
-    samples: Deque[Sample] = field(default_factory=lambda: deque(maxlen=100))
+    def __init__(self, file_count: Optional[int] = None):
+        self.downloaded: int = 0
+        self.estimated_total: Optional[int] = None
+        self.last_printed: Optional[float] = None
+        self.progress_bytes: int = 0
+        self.progress_perc: int = 0
+        self.remaining_time: Optional[int] = None
+        self.samples: Deque[Sample] = deque(maxlen=1000)
+        self.speed: Optional[float] = None
+        self.tasks: Dict[TaskId, Task] = {}
+        self.file_count = file_count
+        self.downloaded_count: int = 0
 
     def start(self, task_id: int, size: int):
         if task_id in self.tasks:
             raise ValueError(f"Task {task_id}: cannot start, already started")
 
         self.tasks[task_id] = Task(task_id, size)
-        self._calculate_total()
-        self._calculate_progress()
         self.print()
 
     def advance(self, task_id: int, size: int):
@@ -63,7 +60,6 @@ class Progress:
         self.progress_bytes += size
         self.tasks[task_id].advance(size)
         self.samples.append(Sample(self.downloaded, time.time()))
-        self._calculate_progress()
         self.print()
 
     def already_downloaded(self, task_id: int, size: int):
@@ -72,9 +68,7 @@ class Progress:
 
         self.tasks[task_id] = Task(task_id, size)
         self.progress_bytes += size
-        self.vod_downloaded_count += 1
-        self._calculate_total()
-        self._calculate_progress()
+        self.downloaded_count += 1
         self.print()
 
     def abort(self, task_id: int):
@@ -83,9 +77,6 @@ class Progress:
 
         del self.tasks[task_id]
         self.progress_bytes = sum(t.downloaded for t in self.tasks.values())
-
-        self._calculate_total()
-        self._calculate_progress()
         self.print()
 
     def end(self, task_id: int):
@@ -98,15 +89,15 @@ class Progress:
                 f"Taks {task_id} ended with {task.downloaded}b downloaded, expected {task.size}b."
             )
 
-        self.vod_downloaded_count += 1
+        self.downloaded_count += 1
         self.print()
 
-    def _calculate_total(self):
-        self.estimated_total = (
-            int(mean(t.size for t in self.tasks.values()) * self.vod_count) if self.tasks else None
-        )
+    def _recalculate(self):
+        if self.tasks and self.file_count:
+            self.estimated_total = int(mean(t.size for t in self.tasks.values()) * self.file_count)
+        else:
+            self.estimated_total = None
 
-    def _calculate_progress(self):
         self.speed = self._calculate_speed()
         self.progress_perc = (
             int(100 * self.progress_bytes / self.estimated_total) if self.estimated_total else 0
@@ -133,10 +124,14 @@ class Progress:
         now = time.time()
 
         # Don't print more often than 10 times per second
-        if now - self.last_printed < 0.1:
+        if self.last_printed and now - self.last_printed < 0.1:
             return
 
-        click.echo(f"\rDownloaded {self.vod_downloaded_count}/{self.vod_count} VODs", nl=False)
+        self._recalculate()
+
+        clear_line()
+        total_label = f"/{self.file_count}" if self.file_count else ""
+        click.echo(f"Downloaded {self.downloaded_count}{total_label} VODs", nl=False)
         click.secho(f" {self.progress_perc}%", fg="blue", nl=False)
 
         if self.estimated_total is not None:
@@ -150,6 +145,4 @@ class Progress:
         if self.remaining_time is not None:
             click.echo(f" ETA {blue(format_time(self.remaining_time))}", nl=False)
 
-        click.echo("    ", nl=False)
-
         self.last_printed = now

twitchdl/twitch.py

@@ -2,83 +2,28 @@
 Twitch API access.
 """
 
-import json
 import logging
+import random
 import time
-from typing import Any, Dict, Generator, List, Literal, Mapping, Optional, Tuple, TypedDict, Union
+from typing import Any, Dict, Generator, List, Mapping, Optional, Tuple, Union
 
 import click
 import httpx
 
 from twitchdl import CLIENT_ID
-from twitchdl.entities import Data
+from twitchdl.entities import (
+    AccessToken,
+    Chapter,
+    Clip,
+    ClipAccessToken,
+    ClipsPeriod,
+    Data,
+    Video,
+    VideosSort,
+    VideosType,
+)
 from twitchdl.exceptions import ConsoleError
-
-ClipsPeriod = Literal["last_day", "last_week", "last_month", "all_time"]
-VideosSort = Literal["views", "time"]
-VideosType = Literal["archive", "highlight", "upload"]
-
-
-class AccessToken(TypedDict):
-    signature: str
-    value: str
-
-
-class User(TypedDict):
-    login: str
-    displayName: str
-
-
-class Game(TypedDict):
-    id: str
-    name: str
-
-
-class VideoQuality(TypedDict):
-    frameRate: str
-    quality: str
-    sourceURL: str
-
-
-class ClipAccessToken(TypedDict):
-    id: str
-    playbackAccessToken: AccessToken
-    videoQualities: List[VideoQuality]
-
-
-class Clip(TypedDict):
-    id: str
-    slug: str
-    title: str
-    createdAt: str
-    viewCount: int
-    durationSeconds: int
-    url: str
-    videoQualities: List[VideoQuality]
-    game: Game
-    broadcaster: User
-
-
-class Video(TypedDict):
-    id: str
-    title: str
-    description: str
-    publishedAt: str
-    broadcastType: str
-    lengthSeconds: int
-    game: Game
-    creator: User
-
-
-class Chapter(TypedDict):
-    id: str
-    durationMilliseconds: int
-    positionMilliseconds: int
-    type: str
-    description: str
-    subDescription: str
-    thumbnailURL: str
-    game: Game
+from twitchdl.utils import format_size
 
 
 class GQLError(click.ClickException):
@@ -135,22 +80,23 @@ logger = logging.getLogger(__name__)
 
 
 def log_request(request: httpx.Request):
-    logger.debug(f"--> {request.method} {request.url}")
+    logger.info(f"--> {request.method} {request.url}")
     if request.content:
         logger.debug(f"--> {request.content}")
 
 
-def log_response(response: httpx.Response, duration: float):
+def log_response(response: httpx.Response, duration_seconds: float):
     request = response.request
-    duration_ms = int(1000 * duration)
-    logger.debug(f"<-- {request.method} {request.url} HTTP {response.status_code} {duration_ms}ms")
+    duration = f"{int(1000 * duration_seconds)}ms"
+    size = format_size(len(response.content))
+    logger.info(f"<-- {request.method} {request.url} HTTP {response.status_code} {duration} {size}")
     if response.content:
         logger.debug(f"<-- {response.content}")
 
 
-def gql_post(query: str):
+def gql_persisted_query(query: Data):
     url = "https://gql.twitch.tv/gql"
-    response = authenticated_post(url, content=query)
+    response = authenticated_post(url, json=query)
     gql_raise_on_error(response)
     return response.json()
 
@@ -238,22 +184,18 @@ def get_clip(slug: str) -> Optional[Clip]:
 
 
 def get_clip_access_token(slug: str) -> ClipAccessToken:
-    query = f"""
-    {{
+    query = {
         "operationName": "VideoAccessToken_Clip",
-        "variables": {{
-            "slug": "{slug}"
-        }},
-        "extensions": {{
-            "persistedQuery": {{
+        "variables": {"slug": slug},
+        "extensions": {
+            "persistedQuery": {
                 "version": 1,
-                "sha256Hash": "36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11"
-            }}
-        }}
-    }}
-    """
+                "sha256Hash": "36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11",
+            }
+        },
+    }
 
-    response = gql_post(query.strip())
+    response = gql_persisted_query(query)
     return response["data"]["clip"]
 
 
@@ -325,23 +267,6 @@ def channel_clips_generator(
     return _generator(clips, limit)
 
 
-def channel_clips_generator_old(channel_id: str, period: ClipsPeriod, limit: int):
-    cursor = ""
-    while True:
-        clips = get_channel_clips(channel_id, period, limit, after=cursor)
-
-        if not clips["edges"]:
-            break
-
-        has_next = clips["pageInfo"]["hasNextPage"]
-        cursor = clips["edges"][-1]["cursor"] if has_next else None
-
-        yield clips, has_next
-
-        if not cursor:
-            break
-
-
 def get_channel_videos(
     channel_id: str,
     limit: int,
@@ -422,7 +347,7 @@ def get_access_token(video_id: str, auth_token: Optional[str] = None) -> AccessT
     query = f"""
     {{
         videoPlaybackAccessToken(
-            id: {video_id},
+            id: "{video_id}",
             params: {{
                 platform: "web",
                 playerBackend: "mediaplayer",
@@ -467,8 +392,12 @@ def get_playlists(video_id: str, access_token: AccessToken) -> str:
             "allow_audio_only": "true",
             "allow_source": "true",
             "player": "twitchweb",
+            "platform": "web",
+            "supported_codecs": "av1,h265,h264",
+            "p": random.randint(1000000, 10000000),
         },
     )
+
     response.raise_for_status()
     return response.content.decode("utf-8")
 
@@ -503,7 +432,7 @@ def get_video_chapters(video_id: str) -> List[Chapter]:
         },
     }
 
-    response = gql_post(json.dumps(query))
+    response = gql_persisted_query(query)
     return list(_chapter_nodes(response["data"]["video"]["moments"]))
 
 

twitchdl/utils.py

@@ -85,6 +85,7 @@ def titlify(value: str) -> str:
 VIDEO_PATTERNS = [
     r"^(?P<id>\d+)?$",
     r"^https://(www\.|m\.)?twitch\.tv/videos/(?P<id>\d+)(\?.+)?$",
+    r"^https://(www\.|m\.)?twitch\.tv/\w+/video/(?P<id>\d+)(\?.+)?$",
 ]
 
 CLIP_PATTERNS = [