Override Subtitle download method to convert fTTML/fVTT to TTML/VTT

We want to force convert these Subtitle formats to their respective normal formats as the way they download are not actually usable as-is. As in, even if the user wanted to keep the original Subtitle format, these formats wouldn't be usable as-is.

.pre-commit-config.yaml

@@ -2,11 +2,6 @@
 # See https://pre-commit.com/hooks.html for more hooks
 
 repos:
-  - repo: https://github.com/compilerla/conventional-pre-commit
-    rev: v3.1.0
-    hooks:
-      - id: conventional-pre-commit
-        stages: [commit-msg]
   - repo: https://github.com/mtkennerly/pre-commit-hooks
     rev: v0.3.0
     hooks:

CHANGELOG.md

@@ -2,265 +2,8 @@
 
 All notable changes to this project will be documented in this file.
 
-This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
-
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-Versions [3.0.0] and older use a format based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
-but versions thereafter use a custom changelog format using [git-cliff](https://git-cliff.org).
-
-## [3.1.0] - 2024-03-05
-
-### Features
-
-- *cli*: Implement MultipleChoice click param based on Choice param
-- *dl*: Skip video lang filter if --v-lang unused & only 1 video lang
-- *dl*: Change --vcodec default to None, use any codec
-- *dl*: Support multiple -r/--range and mux ranges separately
-- *Subtitle*: Convert from fTTML->TTML & fVTT->WebVTT post-download
-- *Track*: Make ID optional, Automatically compute one if not provided
-- *Track*: Add a name property to use for the Track Name
-
-### Bug Fixes
-
-- *dl*: Have --sub-format default to None to keep original sub format
-- *HLS*: Use filtered out segment key info
-- *Track*: Don't modify lang when getting name
-- *Track*: Don't use fallback values "Zzzz"/"ZZ" for track name
-- *version*: The `__version__` variable forgot to be updated
-
-### Changes
-
-- Move dl command's download_track() to Track.download()
-- *dl*: Remove unused `get_profiles()` method
-- *DASH*: Move data values from track url to track data property
-- *DASH*: Change how Video FPS is gotten to remove FutureWarning log
-- *Track*: Add type checks, improve typing
-- *Track*: Remove swap() method and it's uses
-- *Track*: Remove unused DRM enum
-- *Track*: Rename Descriptor's M3U & MPD to HLS & DASH
-- *Track*: Remove unnecessary bool casting
-- *Track*: Move the path class instance variable with the rest
-- *Track*: Return new path on move(), raise exceptions on errors
-- *Track*: Move delete and move methods near start of Class
-- *Track*: Rename extra to data, enforce type as dict
-
-### Builds
-
-- Explicitly use marisa-trie==1.1.0 for Python 3.12 wheels
-
-## [3.0.0] - 2024-03-01
-
-### Added
-
-- Support for Python 3.12.
-- Audio track's Codec Enum now has [FLAC](https://en.wikipedia.org/wiki/FLAC) defined.
-- The Downloader to use can now be set in the config under the [downloader key](CONFIG.md#downloader-str).
-- New Multi-Threaded Downloader, `requests`, that makes HTTP(S) calls using [Python-requests](https://requests.readthedocs.io).
-- New Multi-Threaded Downloader, `curl_impersonate`, that makes HTTP(S) calls using [Curl-Impersonate](https://github.com/yifeikong/curl-impersonate) via [Curl_CFFI](https://github.com/yifeikong/curl_cffi).
-- HLS manifests specifying a Byte range value without starting offsets are now supported.
-- HLS segments that use `EXT-X-DISCONTINUITY` are now supported.
-- DASH manifests with SegmentBase or only BaseURL are now supported.
-- Subtitle tracks from DASH manifests now automatically marked as SDH if `urn:tva:metadata:cs:AudioPurposeCS:2007 = 2`.
-- The `--audio-only/--subs-only/--chapters-only` flags can now be used simultaneously. For example, `--subs-only`
-  with `--chapters-only` will get just Subtitles and Chapters.
-- Added `--video-only` flag, which can also still be simultaneously used with the only "only" flags. Using all four
-  of these flags will have the same effect as not using any of them.
-- Added `--no-proxy` flag, disabling all uses of proxies, even if `--proxy` is set.
-- Added `--sub-format` option, which sets the wanted output subtitle format, defaulting to SubRip (SRT).
-- Added `Subtitle.reverse_rtl()` method to use SubtitleEdit's `/ReverseRtlStartEnd` functionality.
-- Added `Subtitle.convert()` method to convert the loaded Subtitle to another format. Note that you cannot convert to
-  fTTML or fVTT, but you can convert from them. SubtitleEdit will be used in precedence over pycaption if available.
-  Converting to SubStationAlphav4 requires SubtitleEdit, but you may want to manually alter the Canvas resolution after
-  the download.
-- Added support for SubRip (SRT) format subtitles in `Subtitle.parse()` via pycaption.
-- Added `API` Vault Client aiming for a RESTful like API.
-- Added `Chapters` Class to hold the new reworked `Chapter` objects, automatically handling stuff like order of the
-  Chapters, Chapter numbers, loading from a chapter file or string, and saving to a chapter file or string.
-- Added new `chapter_fallback_name` config option allowing you to set a Chapter Name Template used when muxing Chapters
-  into an MKV Container with MKVMerge. Do note, it defaults to no Chapter Fallback Name at all, but MKVMerge will force
-  `Chapter {i:02}` at least for me on Windows with the program language set to English. You may want to instead use
-  `Chapter {j:02}` which will do `Chapter 01, Intro, Chapter 02` instead of `Chapter 01, Intro, Chapter 03` (an Intro
-  is not a Chapter of story, but it is the 2nd Chapter marker, so It's up to you how you want to interpret it).
-- Added new `Track.OnSegmentDownloaded` Event, called any time one of the Track's segments were downloaded.
-- Added new `Subtitle.OnConverted` Event, called any time that Subtitle is converted.
-- Implemented `__add__` method to `Tracks` class, allowing you to add to the first Tracks object. For example, making
-  it handy to merge HLS video tracks with DASH tracks, `tracks = dash_tracks + hls_tracks.videos`, or for iterating:
-  `for track in dash.videos + hls.videos: ...`.
-- Added new utility `get_free_port()` to get a free local port to use, though it may be taken by the time it's used.
-
-### Changed
-
-- Moved from my forked release of pymp4 (`rlaphoenix-pymp4`) back to the original `pymp4` release as it is
-  now up-to-date with some of my needed fixes.
-- The DASH manifest is now stored in the Track `url` property to be reused by `DASH.download_track()`.
-- Encrypted DASH streams are now downloaded in full and then decrypted, instead of downloading and decrypting
-  each individual segment. Unlike HLS, DASH cannot dynamically switch out the DRM/Protection information.
-  This brings both CPU and Disk IOPS improvements, as well as fixing rare weird decryption anomalies like broken
-  or odd timestamps, decryption failures, or broken a/v continuity.
-- When a track is being decrypted, it now displays "Decrypting" and afterward "Decrypted" in place of the download
-  speed.
-- When a track finishes downloaded, it now displays "Downloaded" in place of the download speed.
-- When licensing is needed and fails, the track will display "FAILED" in place of the download speed. The track
-  download will cancel and all other track downloads will be skipped/cancelled; downloading will end.
-- The fancy smart quotes (`“` and `”`) are now stripped from filenames.
-- All available services are now listed if you provide an invalid service tag/alias.
-- If a WVD file fails to load and looks to be in the older unsupported v1 format, then instructions on migrating to
-  v2 will be displayed.
-- If Shaka-Packager prints an error (i.e., `:ERROR:` log message) it will now raise a `subprocess.CalledProcessError`
-  exception, even if the process return code is 0.
-- The Video classes' Primaries, Transfer, and Matrix classes had changes to their enum names to better represent their
-  values and uses. See the changed names in the [commit](https://github.com/devine-dl/devine/commit/c159672181ee3bd07b06612f256fa8590d61795c).
-- SubRip (SRT) Subtitles no longer have the `MULTI-LANGUAGE SRT` header forcefully removed. The root cause of the error
-  was identified and fixed in this release.
-- Since `Range.Transfer.SDR_BT_601_625 = 5` has been removed, `Range.from_cicp()` now internally remaps CICP transfer
-  values of `5` to `6` (which is now `Range.Transfer.BT_601 = 6`).
-- Referer and User-Agent Header values passed to the aria2(c) downloader is now set via the dedicated `--referer` and
-  `--user-agent` options respectively, instead of `--header`.
-- The aria2(c) `-j`, `-x`, and `-s` option values can now be set by the config under the `aria2c` key in the options'
-  full names.
-- The aria2(c) `-x`, and `-s` option values now use aria2(c)'s own default values for them instead of `16`. The `j`
-  option value defaults to ThreadPoolExecutor's algorithm of `min(32,(cpu_count+4))`.
-- The download progress bar now states `LICENSING` on the speed text when licensing DRM, and `LICENSED` once finished.
-- The download progress bar now states `CANCELLING`/`CANCELLED` on the speed text when cancelling downloads. This is to
-  make it more clear that it didn't just stop, but stopped as it was cancelled.
-- The download cancel/skip events were moved to `constants.py` so it can be used across the codebase easier without
-  argument drilling. `DL_POOL_STOP` was renamed to `DOWNLOAD_CANCELLED` and `DL_POOL_SKIP` to `DOWNLOAD_LICENCE_ONLY`.
-- The Cookie header is now calculated for each URL passed to the aria2(c) downloader based on the URL. Instead of
-  passing every single cookie, which could have two cookies with the same name aimed for different host names, we now
-  pass only cookies intended for the URL.
-- The aria2(c) process no longer prints output to the terminal directly. Devine now only prints contents of the
-  captured log messages to the terminal. This allows filtering out of errors and warnings that isn't a problem.
-- DASH and HLS no longer download segments silencing errors on all but the last retry as the downloader rework makes
-  this unnecessary. The errors will only be printed on the final retry regardless.
-- `Track.repackage()` now saves as `{name}_repack.{ext}` instead of `{name}.repack.{ext}`.
-- `Video.change_color_range()` now saves as `{name}_{limited|full}_range.{ext}` instead of `{name}.range{0|1}.{ext}`.
-- `Widevine.decrypt()` now saves as `{name}_decrypted.{ext}` instead of `{name}.decrypted.{ext}`.
-- Files starting with the save path's name and using the save path's extension, but not the save path, are no longer
-  deleted on download finish/stop/failure.
-- The output container format is now explicitly specified as `MP4` when calling `shaka-packager`.
-- The default downloader is now `requests` instead of `aria2c` to reduce required external dependencies.
-- Reworked the `Chapter` class to only hold a timestamp and name value with an ID automatically generated as a CRC32 of
-  the Chapter representation.
-- The `--group` option has been renamed to `--tag`.
-- The config file is now read from three more locations in the following order:
-  1) The Devine Namespace Folder (e.g., `%appdata%/Python/Python311/site-packages/devine/devine.yaml`).
-  2) The Parent Folder to the Devine Namespace Folder (e.g., `%appdata%/Python/Python311/site-packages/devine.yaml`).
-  3) The AppDirs User Config Folder (e.g., `%localappdata%/devine/devine.yaml`).
-  Location 2 allows having a config at the root of a portable folder.
-- An empty config file is no longer created when no config file is found.
-- You can now set a default cookie file for a Service, [see README](README.md#cookies--credentials).
-- You can now set a default credential for a Service, [see config](CONFIG.md#credentials-dictstr-strlistdict).
-- Services are now auth-less by default and the error for not having at least a cookie or credential is removed.
-  Cookies/Credentials will only be loaded if a default one for the service is available, or if you use `-p/--profile`
-  and the profile exists.
-- Subtitles when converting to SubRip (SRT) via SubtitleEdit will now use the `/ConvertColorsToDialog` option.
-- HLS segments are now merged by discontinuity instead of all at once. The merged discontinuities are then finally
-  merged to one file using `ffmpeg`. Doing the final merge by byte concatenation did not work for some playlists.
-- The Track is no longer passed through Event Callables. If you are able to set a function on an Even Callable, then
-  you should have access to the track reference to call it directly if needed.
-- The Track.OnDecrypted event callable is now passed the DRM and Segment objects used to Decrypt. The segment object is
-  only passed from HLS downloads.
-- The Track.OnDownloaded event callable is now called BEFORE decryption, right after downloading, not after decryption.
-- All generated Track ID values across the codebase has moved from md5 to crc32 values as code processors complain
-  about its use surrounding security, and it's length is too large for our use case anyway.
-- HLS segments are now downloaded multi-threaded first and then processed in sequence thereafter.
-- HLS segments are no longer decrypted one-by-one, requiring a lot of shaka-packager processes to run and close.
-  They now merged and decrypt in groups based on their EXT-X-KEY, before being merged per discontinuity.
-- The DASH and HLS downloaders now pass multiple URLs to the downloader instead of one-by-one, heavily increasing speed
-  and reliability as connections are kept alive and re-used.
-- Downloaders now yield back progress information in the same convention used by `rich`'s `Progress.update()` method.
-  DASH and HLS now pass the yielded information to their progress callable instead of passing the progress callable to
-  the downloader.
-- The aria2(c) downloader now uses the aria2(c) JSON-RPC interface to query for download progress updates instead of
-  parsing the stdout data in an extremely hacky way.
-- The aria2(c) downloader now re-routes non-HTTP proxies via `pproxy` by a subprocess instead of the now-removed
-  `start_pproxy` utility. This way has proven to be easier, more reliable, and prevents pproxy from messing with rich's
-  terminal output in strange ways.
-- All downloader function's have an altered signature but ultimately similar. `uri` to `urls`, `out` (path) was removed,
-  we now calculate the save path by passing an `output_dir` and `filename`. The `silent`, `segmented`, and `progress`
-  parameters were completely removed.
-- All downloader `urls` can now be a string or a dictionary containing extra URL-specific options to use like
-  URL-specific headers. It can also be a list of the two types of URLs to downloading multi-threaded.
-- All downloader `filenames` can be a static string, or a filename string template with a few variables to use. The
-  template system used is f-string, e.g., `"file_{i:03}{ext}"` (ext starts with `.` if there's an extension).
-- DASH now updates the progress bar when merging segments.
-- The `Widevine.decrypt()` method now also searches for shaka-packager as just `packager` as it is the default build
-  name. (#74)
-
-### Removed
-
-- The `devine auth` command and sub-commands due to lack of support, risk of data, and general quirks with it.
-- Removed `profiles` config, you must now specify which profile you wish to use each time with `-p/--profile`. If you
-  use a specific profile a lot more than others, you should make it the default.
-- The `saldl` downloader has been removed as their binary distribution is whack and development has seemed to stall.
-  It was only used as an alternative to what was at the time the only downloader, aria2(c), as it did not support any
-  form of Byte Range, but `saldl` did, which was crucial for resuming extremely large downloads or complex playlists.
-  However, now we have the requests downloader which does support the Range header.
-- The `Track.needs_proxy` property was removed for a few design architectural reasons.
-  1) Design-wise it isn't valid to have --proxy (or via config/otherwise) set a proxy, then unpredictably have it
-     bypassed or disabled. If I specify `--proxy 127.0.0.1:8080`, I would expect it to use that proxy for all
-     communication indefinitely, not switch in and out depending on the track or service.
-  2) With reason 1, it's also a security problem. The only reason I implemented it in the first place was so I could
-     download faster on my home connection. This means I would authenticate and call APIs under a proxy, then suddenly
-     download manifests and segments e.t.c under my home connection. A competent service could see that as an indicator
-     of bad play and flag you.
-  3) Maintaining this setup across the codebase is extremely annoying, especially because of how proxies are setup/used
-     by Requests in the Session. There's no way to tell a request session to temporarily disable the proxy and turn it
-     back on later, without having to get the proxy from the session (in an annoying way) store it, then remove it,
-     make the calls, then assuming your still in the same function you can add it back. If you're not in the same
-     function, well, time for some spaghetti code.
-- The `Range.Transfer.SDR_BT_601_625 = 5` key and value has been removed as I cannot find any official source to verify
-  it as the correct use. However, usually a `transfer` value of `5` would be PAL SD material so it better matches `6`,
-  which is (now named) `Range.Transfer.BT_601 = 6`. If you have something specifying transfer=5, just remap it to 6.
-- The warning log `There's no ... Audio Tracks, likely part of an invariant playlist, continuing...` message has been
-  removed. So long as your playlist is expecting no audio tracks, or the audio is part of the video transport, then
-  this wouldn't be a problem whatsoever. Therefore, having it log this annoying warning all the time is pointless.
-- The `--min-split-size` argument to the aria2(c) downloader as it was only used to disable splitting on
-  segmented downloads, but the newer downloader system wouldn't really need or want this to be done. If aria2 has
-  decided based on its other settings to have split a segment file, then it likely would benefit from doing so.
-- The `--remote-time` argument from the aria2(c) downloader as it may need to do a GET and a HEAD request to
-  get the remote time information, slowing the download down. We don't need this information anyway as it will likely
-  be repacked with `ffmpeg` or multiplexed with `mkvmerge`, discarding/losing that information.
-- DASH and HLS's 5-attempt retry loop as the downloaders will retry for us.
-- The `start_pproxy` utility has been removed as all uses of it now call `pproxy` via subprocess instead.
-- The `LANGUAGE_MUX_MAP` constant and it's usage has been removed as it is no longer necessary as of MKVToolNix v54.
-
-### Fixed
-
-- Uses of `__ALL__` with Class objects have been correct to `__all__` with string objects, following PEP8.
-- Fixed value of URL passed to `Track.get_key_id()` as it was a tuple rather than the URL string.
-- The `--skip-dl` flag now works again after breaking in v[1.3.0].
-- Move WVD file to correct location on new installations in the `wvd add` command.
-- Cookie data is now passed to downloaders and use URLs based on the URI it will be used for, just like a browser.
-- Failure to get FPS in DASH when SegmentBase isn't used.
-- An error message is now returned if a WVD file fails to load instead of raising an exception.
-- Track language information within M3U playlists are now validated with langcodes before use. Some manifests use the
-  property for arbitrary data that their apps/players use for their own purposes.
-- Attempt to fix non-UTF-8 and mixed-encoding Subtitle downloads by automatically converting to UTF-8. (#43)
-  Decoding is attempted in the following order: UTF-8, CP-1252, then finally chardet detection. If it's neither UTF-8
-  nor CP-1252 and chardet could not detect the encoding, then it is left as-is. Conversion is done per-segment if the
-  Subtitle is segmented, unless it's the fVTT or fTTML formats which are binary.
-- Chapter Character Encoding is now explicitly set to UTF-8 when muxing to an MKV container as Windows seems to default
-  to latin1 or something, breaking Chapter names with any sort of special character within.
-- Subtitle passed through SubtitleEdit now explicitly use UTF-8 character encoding as it usually defaulted to UTF-8
-  with Byte Order Marks (aka UTF-8-SIG/UTF-8-BOM).
-- Subtitles passed through SubtitleEdit now use the same output format as the subtitle being processed instead of SRT.
-- Fixed rare infinite loop when the Server hosting the init/header data/segment file responds with a `Content-Length`
-  header with a value of `0` or smaller.
-- Removed empty caption lists/languages when parsing Subtitles with `Subtitle.parse()`. This stopped conversions to SRT
-  containing the `MULTI-LANGUAGE SRT` header when there was multiple caption lists, even though only one of them
-  actually contained captions.
-- Text-based Subtitle formats now try to automatically convert to UTF-8 when run through `Subtitle.parse()`.
-- Text-based Subtitle formats now have `‎` and `‏` HTML entities unescaped post-download as some rendering
-  libraries seems to not decode them for us. SubtitleEdit also has problems with `/ReverseRtlStartEnd` unless it's
-  already decoded.
-- Fixed two concatenation errors surrounding DASH's BaseURL, sourceURL, and media values that start with or use `../`.
-- Fixed the number values in the `Newly added to x/y Vaults` log, which now states `Cached n Key(s) to x/y Vaults`.
-- File write handler now flushes after appending a new segment to the final save path or checkpoint file, reducing
-  memory usage by quite a bit in some scenarios.
-
-### New Contributors
-
-- [Shivelight](https://github.com/Shivelight)
 
 ## [2.2.0] - 2023-04-23
 
@@ -685,8 +428,6 @@ This release brings a huge change to the fundamentals of Devine's logging, UI, a
 
 Initial public release under the name Devine.
 
-[3.1.0]: https://github.com/devine-dl/devine/releases/tag/v3.1.0
-[3.0.0]: https://github.com/devine-dl/devine/releases/tag/v3.0.0
 [2.2.0]: https://github.com/devine-dl/devine/releases/tag/v2.2.0
 [2.1.0]: https://github.com/devine-dl/devine/releases/tag/v2.1.0
 [2.0.1]: https://github.com/devine-dl/devine/releases/tag/v2.0.1

CONFIG.md

@@ -11,12 +11,13 @@ which does not keep comments.
 ## aria2c (dict)
 
 - `max_concurrent_downloads`
-  Maximum number of parallel downloads. Default: `min(32,(cpu_count+4))`  
+  Maximum number of parallel downloads. Default: `5`  
-  Note: Overrides the `max_workers` parameter of the aria2(c) downloader function.
+  Note: Currently unused as downloads are multi-threaded by Devine rather than Aria2c.
+  Devine internally has a constant set value of 16 for it's parallel downloads.
 - `max_connection_per_server`
   Maximum number of connections to one server for each download. Default: `1`
 - `split`
-  Split a file into N chunks and download each chunk on its own connection. Default: `5`
+  Split a file into N chunks and download each chunk on it's own connection. Default: `5`
 - `file_allocation`
   Specify file allocation method. Default: `"prealloc"`
 
@@ -66,45 +67,25 @@ DSNP:
 default: chromecdm_903_l3
 ```
 
-## chapter_fallback_name (str)
+## credentials (dict)
 
-The Chapter Name to use when exporting a Chapter without a Name.
+Specify login credentials to use for each Service by Profile as Key (case-sensitive).
-The default is no fallback name at all and no Chapter name will be set.
 
-The fallback name can use the following variables in f-string style:
+The value should be `email:password` or `username:password` (with some exceptions).  
-
+The first section does not have to be an email or username. It may also be a Phone number.
-- `{i}`: The Chapter number starting at 1.
-         E.g., `"Chapter {i}"`: "Chapter 1", "Intro", "Chapter 3".
-- `{j}`: A number starting at 1 that increments any time a Chapter has no title.
-         E.g., `"Chapter {j}"`: "Chapter 1", "Intro", "Chapter 2".
-
-These are formatted with f-strings, directives are supported.
-For example, `"Chapter {i:02}"` will result in `"Chapter 01"`.
-
-## credentials (dict[str, str|list|dict])
-
-Specify login credentials to use for each Service, and optionally per-profile.
 
 For example,
 
 ```yaml
-ALL4: jane@gmail.com:LoremIpsum100  # directly
+AMZN:
-AMZN:  # or per-profile, optionally with a default
-  default: jane@example.tld:LoremIpsum99  # <-- used by default if -p/--profile is not used
   james: james@gmail.com:TheFriend97
+  jane: jane@example.tld:LoremIpsum99
   john: john@example.tld:LoremIpsum98
-NF:  # the `default` key is not necessary, but no credential will be used by default
+NF:
   john: john@gmail.com:TheGuyWhoPaysForTheNetflix69420
 ```
 
-The value should be in string form, i.e. `john@gmail.com:password123` or `john:password123`.  
+Credentials must be specified per-profile. You cannot specify a fallback or default credential.
-Any arbitrary values can be used on the left (username/password/phone) and right (password/secret).  
-You can also specify these in list form, i.e., `["john@gmail.com", ":PasswordWithAColon"]`.
-
-If you specify multiple credentials with keys like the `AMZN` and `NF` example above, then you should
-use a `default` key or no credential will be loaded automatically unless you use `-p/--profile`. You
-do not have to use a `default` key at all.
-
 Please be aware that this information is sensitive and to keep it safe. Do not share your config.
 
 ## curl_impersonate (dict)
@@ -160,7 +141,7 @@ AMZN:
   bitrate: CVBR
 ```
 
-or to change the output subtitle format from the default (original format) to WebVTT,
+or to change the output subtitle format from the default (SubRip SRT) to WebVTT,
 
 ```yaml
 sub_format: vtt
@@ -172,8 +153,8 @@ Choose what software to use to download data throughout Devine where needed.
 
 Options:
 
-- `requests` (default) - https://github.com/psf/requests
+- `aria2c` (default) - https://github.com/aria2/aria2
-- `aria2c` - https://github.com/aria2/aria2
+- `requests` - https://github.com/psf/requests
 - `curl_impersonate` - https://github.com/yifeikong/curl-impersonate (via https://github.com/yifeikong/curl_cffi)
 
 Note that aria2c can reach the highest speeds as it utilizes threading and more connections than the other
@@ -207,28 +188,12 @@ provide the same Key ID and CEK for both Video and Audio, as well as for multipl
 You can have as many Key Vaults as you would like. It's nice to share Key Vaults or use a unified Vault on
 Teams as sharing CEKs immediately can help reduce License calls drastically.
 
-Three types of Vaults are in the Core codebase, API, SQLite and MySQL. API makes HTTP requests to a RESTful API,
+Two types of Vaults are in the Core codebase, SQLite and MySQL Vaults. Both directly connect to an SQLite or MySQL
-whereas SQLite and MySQL directly connect to an SQLite or MySQL Database.
+Server. It has to connect directly to the Host/IP. It cannot be in front of a PHP API or such. Beware that some Hosts
+do not let you access the MySQL server outside their intranet (aka Don't port forward or use permissive network
+interfaces).
 
-Note: SQLite and MySQL vaults have to connect directly to the Host/IP. It cannot be in front of a PHP API or such.
+### Connecting to a MySQL Vault
-Beware that some Hosting Providers do not let you access the MySQL server outside their intranet and may not be
-accessible outside their hosting platform.
-
-### Using an API Vault
-
-API vaults use a specific HTTP request format, therefore API or HTTP Key Vault APIs from other projects or services may
-not work in Devine. The API format can be seen in the [API Vault Code](devine/vaults/API.py).
-
-```yaml
-- type: API
-  name: "John#0001's Vault"             # arbitrary vault name
-  uri: "https://key-vault.example.com"  # api base uri (can also be an IP or IP:Port)
-  # uri: "127.0.0.1:80/key-vault"
-  # uri: "https://api.example.com/key-vault"
-  token: "random secret key"            # authorization token
-```
-
-### Using a MySQL Vault
 
 MySQL vaults can be either MySQL or MariaDB servers. I recommend MariaDB.  
 A MySQL Vault can be on a local or remote network, but I recommend SQLite for local Vaults.
@@ -254,7 +219,7 @@ make tables yourself.
 - You may give trusted users CREATE permission so devine can create tables if needed.
 - Other uses should only be given SELECT and INSERT permissions.
 
-### Using an SQLite Vault
+### Connecting to an SQLite Vault
 
 SQLite Vaults are usually only used for locally stored vaults. This vault may be stored on a mounted Cloud storage
 drive, but I recommend using SQLite exclusively as an offline-only vault. Effectively this is your backup vault in
@@ -279,6 +244,34 @@ together.
 - `set_title`
   Set the container title to `Show SXXEXX Episode Name` or `Movie (Year)`. Default: `true`
 
+## profiles (dict)
+
+Pre-define Profiles to use Per-Service.
+
+For example,
+
+```yaml
+AMZN: jane
+DSNP: john
+```
+
+You can also specify a fallback value to pre-define if a match was not made.  
+This can be done using `default` key. This can help reduce redundancy in your specifications.
+
+```yaml
+AMZN: jane
+DSNP: john
+default: james
+```
+
+If a Service doesn't require a profile (as it does not require Credentials or Authorization of any kind), you can
+disable the profile checks by specifying `false` as the profile for the Service.
+
+```yaml
+ALL4: false
+CTV: false
+```
+
 ## proxy_providers (dict)
 
 Enable external proxy provider services.

README.md

@@ -2,7 +2,7 @@
     <img src="https://user-images.githubusercontent.com/17136956/216880837-478f3ec7-6af6-4cca-8eef-5c98ff02104c.png">
     <a href="https://github.com/devine-dl/devine">Devine</a>
     <br/>
-    <sup><em>Modular Movie, TV, and Music Archival Software</em></sup>
+    <sup><em>Open-Source Movie, TV, and Music Downloading Solution</em></sup>
     <br/>
     <a href="https://discord.gg/34K2MGDrBN">
         <img src="https://img.shields.io/discord/841055398240059422?label=&logo=discord&logoColor=ffffff&color=7289DA&labelColor=7289DA" alt="Discord">
@@ -59,23 +59,19 @@ A command-line interface is now available, try `devine --help`.
 
 ### Dependencies
 
-The following is a list of programs that need to be installed by you manually.
+The following is a list of programs that need to be installed manually. I recommend installing these with [winget],
+[chocolatey] or such where possible as it automatically adds them to your `PATH` environment variable and will be
+easier to update in the future.
 
+- [aria2(c)] for downloading streams and large manifests.
 - [CCExtractor] for extracting Closed Caption data like EIA-608 from video streams and converting as SRT.
 - [FFmpeg] (and ffprobe) for repacking/remuxing streams on specific services, and evaluating stream data.
 - [MKVToolNix] v54+ for muxing individual streams to an `.mkv` file.
 - [shaka-packager] for decrypting CENC-CTR and CENC-CBCS video and audio streams.
-- (optional) [aria2(c)] to use as a [downloader](CONFIG.md#downloader-str).
 
-> [!TIP]
+For portable downloads, make sure you put them in your current working directory, in the installation directory,
-> You should install these from a Package Repository if you can; including winget/chocolatey on Windows. They will
+or put the directory path in your `PATH` environment variable. If you do not do this then their binaries will not be
-> automatically add the binary's path to your `PATH` environment variable and will be easier to update in the future.
+able to be found.
-
-> [!IMPORTANT]
-> Most of these dependencies are portable utilities and therefore do not use installers. If you do not install them
-> from a package repository like winget/choco/pacman then make sure you put them in your current working directory, in
-> Devine's installation directory, or the binary's path into your `PATH` environment variable. If you do not do this
-> then Devine will not be able to find the binaries.
 
   [winget]: <https://winget.run>
   [chocolatey]: <https://chocolatey.org>
@@ -252,33 +248,22 @@ sure that the version of devine you have locally is supported by the Service cod
 > automatically download. Python importing the files triggers the download to begin. However, it may cause a delay on
 > startup.
 
-## Cookies & Credentials
+## Profiles (Cookies & Credentials)
 
-Devine can authenticate with Services using Cookies and/or Credentials. Credentials are stored in the config, and
+Just like a streaming service, devine associates both a cookie and/or credential as a Profile. You can associate up to
-Cookies are stored in the data directory which can be found by running `devine env info`.
+one cookie and one credential per-profile, depending on which (or both) are needed by the Service. This system allows
+you to configure multiple accounts per-service and choose which to use at any time.
 
-To add a Credential to a Service, take a look at the [Credentials Config](CONFIG.md#credentials-dictstr-strlistdict)
+Credentials are stored in the config, and Cookies are stored in the data directory. You can find the location of these
-for information on setting up one or more credentials per-service. You can add one or more Credential per-service and
+by running `devine env info`. However, you can manage profiles with `devine auth --help`. E.g. to add a new John
-use `-p/--profile` to choose which Credential to use.
+profile to Netflix with a Cookie and Credential, take a look at the following CLI call,
+`devine auth add John NF --cookie "C:\Users\John\Downloads\netflix.com.txt --credential "john@gmail.com:pass123"`
 
-To add a Cookie to a Service, use a Cookie file extension to make a `cookies.txt` file and move it into the Cookies
+You can also delete a credential with `devine auth delete`. E.g., to delete the cookie for John that we just added, run
-directory. You must rename the `cookies.txt` file to that of the Service tag (case-sensitive), e.g., `NF.txt`. You can
+`devine auth delete John --cookie`. Take a look at `devine auth delete --help` for more information.
-also place it in a Service Cookie folder, e.g., `/Cookies/NF/default.txt` or `/Cookies/NF/.txt`.
 
-You can add multiple Cookies to the `/Cookies/NF/` folder with their own unique name and then use `-p/--profile` to
+> __Note__ Profile names are case-sensitive and unique per-service. They also have no arbitrary character or length
-choose which one to use. E.g., `/Cookies/NF/sam.txt` and then use it with `--profile sam`. If you make a Service Cookie
+> limit, but for convenience I don't recommend using any special characters as your terminal may get confused.
-folder without a `.txt` or `default.txt`, but with another file, then no Cookies will be loaded unless you use
-`-p/--profile` like shown. This allows you to opt in to authentication at whim.
-
-> [!TIP]
-> - If your Service does not require Authentication, then do not define any Credential or Cookie for that Service.
-> - You can use both Cookies and Credentials at the same time, so long as your Service takes and uses both.
-> - If you are using profiles, then make sure you use the same name on the Credential name and Cookie file name when
->   using `-p/--profile`.
-
-> [!WARNING]
-> Profile names are case-sensitive and unique per-service. They have no arbitrary character or length limit, but for
-> convenience sake I don't recommend using any special characters as your terminal may get confused.
 
 ### Cookie file format and Extensions
 
@@ -349,4 +334,4 @@ You can find a copy of the license in the LICENSE file in the root folder.
 
 * * *
 
-© rlaphoenix 2019-2024
+© rlaphoenix 2019-2023

cliff.toml

@@ -1,71 +0,0 @@
-# git-cliff ~ default configuration file
-# https://git-cliff.org/docs/configuration
-
-[changelog]
-header = """
-# Changelog\n
-All notable changes to this project will be documented in this file.
-
-This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-
-Versions [3.0.0] and older use a format based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
-but versions thereafter use a custom changelog format using [git-cliff](https://git-cliff.org).\n
-"""
-body = """
-{% if version -%}
-    ## [{{ version | trim_start_matches(pat="v") }}] - {{ timestamp | date(format="%Y-%m-%d") }}
-{% else -%}
-    ## [Unreleased]
-{% endif -%}
-{% for group, commits in commits | group_by(attribute="group") %}
-    ### {{ group | striptags | trim | upper_first }}
-    {% for commit in commits %}
-        - {% if commit.scope %}*{{ commit.scope }}*: {% endif %}\
-            {% if commit.breaking %}[**breaking**] {% endif %}\
-            {{ commit.message | upper_first }}\
-    {% endfor %}
-{% endfor %}\n
-"""
-footer = """
-{% for release in releases -%}
-    {% if release.version -%}
-        {% if release.previous.version -%}
-            [{{ release.version | trim_start_matches(pat="v") }}]: \
-                https://github.com/{{ remote.github.owner }}/{{ remote.github.repo }}\
-                    /compare/{{ release.previous.version }}..{{ release.version }}
-        {% endif -%}
-    {% else -%}
-        [unreleased]: https://github.com/{{ remote.github.owner }}/{{ remote.github.repo }}\
-            /compare/{{ release.previous.version }}..HEAD
-    {% endif -%}
-{% endfor %}
-"""
-trim = true
-postprocessors = [
-  # { pattern = '<REPO>', replace = "https://github.com/orhun/git-cliff" }, # replace repository URL
-]
-
-[git]
-conventional_commits = true
-filter_unconventional = true
-split_commits = false
-commit_preprocessors = []
-commit_parsers = [
-  { message = "^feat", group = "<!-- 0 -->Features" },
-  { message = "^fix|revert", group = "<!-- 1 -->Bug Fixes" },
-  { message = "^docs", group = "<!-- 2 -->Documentation" },
-  { message = "^style", skip = true },
-  { message = "^refactor", group = "<!-- 3 -->Changes" },
-  { message = "^perf", group = "<!-- 4 -->Performance Improvements" },
-  { message = "^test", skip = true },
-  { message = "^build", group = "<!-- 5 -->Builds" },
-  { message = "^ci", skip = true },
-  { message = "^chore", skip = true },
-]
-protect_breaking_commits = false
-filter_commits = false
-# tag_pattern = "v[0-9].*"
-# skip_tags = ""
-# ignore_tags = ""
-topo_order = false
-sort_commits = "oldest"

devine/commands/auth.py

@@ -0,0 +1,266 @@
+import logging
+import shutil
+import sys
+import tkinter.filedialog
+from collections import defaultdict
+from pathlib import Path
+from typing import Optional
+
+import click
+from ruamel.yaml import YAML
+
+from devine.core.config import Config, config
+from devine.core.constants import context_settings
+from devine.core.credential import Credential
+
+
+@click.group(
+    short_help="Manage cookies and credentials for profiles of services.",
+    context_settings=context_settings)
+@click.pass_context
+def auth(ctx: click.Context) -> None:
+    """Manage cookies and credentials for profiles of services."""
+    ctx.obj = logging.getLogger("auth")
+
+
+@auth.command(
+    name="list",
+    short_help="List profiles and their state for a service or all services.",
+    context_settings=context_settings)
+@click.argument("service", type=str, required=False)
+@click.pass_context
+def list_(ctx: click.Context, service: Optional[str] = None) -> None:
+    """
+    List profiles and their state for a service or all services.
+
+    \b
+    Profile and Service names are case-insensitive.
+    """
+    log = ctx.obj
+    service_f = service
+
+    auth_data: dict[str, dict[str, list]] = defaultdict(lambda: defaultdict(list))
+
+    if config.directories.cookies.exists():
+        for cookie_dir in config.directories.cookies.iterdir():
+            service = cookie_dir.name
+            for cookie in cookie_dir.glob("*.txt"):
+                if cookie.stem not in auth_data[service]:
+                    auth_data[service][cookie.stem].append("Cookie")
+
+    for service, credentials in config.credentials.items():
+        for profile in credentials:
+            auth_data[service][profile].append("Credential")
+
+    for service, profiles in dict(sorted(auth_data.items())).items():  # type:ignore
+        if service_f and service != service_f.upper():
+            continue
+        log.info(service)
+        for profile, authorizations in dict(sorted(profiles.items())).items():
+            log.info(f'  "{profile}": {", ".join(authorizations)}')
+
+
+@auth.command(
+    short_help="View profile cookies and credentials for a service.",
+    context_settings=context_settings)
+@click.argument("profile", type=str)
+@click.argument("service", type=str)
+@click.pass_context
+def view(ctx: click.Context, profile: str, service: str) -> None:
+    """
+    View profile cookies and credentials for a service.
+
+    \b
+    Profile and Service names are case-sensitive.
+    """
+    log = ctx.obj
+    service_f = service
+    profile_f = profile
+    found = False
+
+    for cookie_dir in config.directories.cookies.iterdir():
+        if cookie_dir.name == service_f:
+            for cookie in cookie_dir.glob("*.txt"):
+                if cookie.stem == profile_f:
+                    log.info(f"Cookie: {cookie}")
+                    log.debug(cookie.read_text(encoding="utf8").strip())
+                    found = True
+                    break
+
+    for service, credentials in config.credentials.items():
+        if service == service_f:
+            for profile, credential in credentials.items():
+                if profile == profile_f:
+                    log.info(f"Credential: {':'.join(list(credential))}")
+                    found = True
+                    break
+
+    if not found:
+        raise click.ClickException(
+            f"Could not find Profile '{profile_f}' for Service '{service_f}'."
+            f"\nThe profile and service values are case-sensitive."
+        )
+
+
+@auth.command(
+    short_help="Check what profile is used by services.",
+    context_settings=context_settings)
+@click.argument("service", type=str, required=False)
+@click.pass_context
+def status(ctx: click.Context, service: Optional[str] = None) -> None:
+    """
+    Check what profile is used by services.
+
+    \b
+    Service names are case-sensitive.
+    """
+    log = ctx.obj
+    found_profile = False
+    for service_, profile in config.profiles.items():
+        if not service or service_.upper() == service.upper():
+            log.info(f"{service_}: {profile or '--'}")
+            found_profile = True
+
+    if not found_profile:
+        log.info(f"No profile has been explicitly set for {service}")
+
+    default = config.profiles.get("default", "not set")
+    log.info(f"The default profile is {default}")
+
+
+@auth.command(
+    short_help="Delete a profile and all of its authorization from a service.",
+    context_settings=context_settings)
+@click.argument("profile", type=str)
+@click.argument("service", type=str)
+@click.option("--cookie", is_flag=True, default=False, help="Only delete the cookie.")
+@click.option("--credential", is_flag=True, default=False, help="Only delete the credential.")
+@click.pass_context
+def delete(ctx: click.Context, profile: str, service: str, cookie: bool, credential: bool):
+    """
+    Delete a profile and all of its authorization from a service.
+
+    \b
+    By default this does remove both Cookies and Credentials.
+    You may remove only one of them with --cookie or --credential.
+
+    \b
+    Profile and Service names are case-sensitive.
+    Comments may be removed from config!
+    """
+    log = ctx.obj
+    service_f = service
+    profile_f = profile
+    found = False
+
+    if not credential:
+        for cookie_dir in config.directories.cookies.iterdir():
+            if cookie_dir.name == service_f:
+                for cookie_ in cookie_dir.glob("*.txt"):
+                    if cookie_.stem == profile_f:
+                        cookie_.unlink()
+                        log.info(f"Deleted Cookie: {cookie_}")
+                        found = True
+                        break
+
+    if not cookie:
+        for key, credentials in config.credentials.items():
+            if key == service_f:
+                for profile, credential_ in credentials.items():
+                    if profile == profile_f:
+                        config_path = Config._Directories.user_configs / Config._Filenames.root_config
+                        yaml, data = YAML(), None
+                        yaml.default_flow_style = False
+                        data = yaml.load(config_path)
+                        del data["credentials"][key][profile_f]
+                        yaml.dump(data, config_path)
+                        log.info(f"Deleted Credential: {credential_}")
+                        found = True
+                        break
+
+    if not found:
+        raise click.ClickException(
+            f"Could not find Profile '{profile_f}' for Service '{service_f}'."
+            f"\nThe profile and service values are case-sensitive."
+        )
+
+
+@auth.command(
+    short_help="Add a Credential and/or Cookies to an existing or new profile for a service.",
+    context_settings=context_settings)
+@click.argument("profile", type=str)
+@click.argument("service", type=str)
+@click.option("--cookie", type=str, default=None, help="Direct path to Cookies to add.")
+@click.option("--credential", type=str, default=None, help="Direct Credential string to add.")
+@click.pass_context
+def add(ctx: click.Context, profile: str, service: str, cookie: Optional[str] = None, credential: Optional[str] = None):
+    """
+    Add a Credential and/or Cookies to an existing or new profile for a service.
+
+    \b
+    Cancel the Open File dialogue when presented if you do not wish to provide
+    cookies. The Credential should be in `Username:Password` form. The username
+    may be an email. If you do not wish to add a Credential, just hit enter.
+
+    \b
+    Profile and Service names are case-sensitive!
+    Comments may be removed from config!
+    """
+    log = ctx.obj
+    service = service.upper()
+    profile = profile.lower()
+
+    if cookie:
+        cookie = Path(cookie)
+        if not cookie.is_file():
+            log.error(f"No such file or directory: {cookie}.")
+            sys.exit(1)
+    else:
+        print("Opening File Dialogue, select a Cookie file to import.")
+        cookie = tkinter.filedialog.askopenfilename(
+            title="Select a Cookie file (Cancel to skip)",
+            filetypes=[("Cookies", "*.txt"), ("All files", "*.*")]
+        )
+        if cookie:
+            cookie = Path(cookie)
+        else:
+            log.info("Skipped adding a Cookie...")
+
+    if credential:
+        try:
+            credential = Credential.loads(credential)
+        except ValueError as e:
+            raise click.ClickException(str(e))
+    else:
+        credential = input("Credential: ")
+        if credential:
+            try:
+                credential = Credential.loads(credential)
+            except ValueError as e:
+                raise click.ClickException(str(e))
+        else:
+            log.info("Skipped adding a Credential...")
+
+    if cookie:
+        final_path = (config.directories.cookies / service / profile).with_suffix(".txt")
+        final_path.parent.mkdir(parents=True, exist_ok=True)
+        if final_path.exists():
+            log.error(f"A Cookie file for the Profile {profile} on {service} already exists.")
+            sys.exit(1)
+        shutil.move(cookie, final_path)
+        log.info(f"Moved Cookie file to: {final_path}")
+
+    if credential:
+        config_path = Config._Directories.user_configs / Config._Filenames.root_config
+        yaml, data = YAML(), None
+        yaml.default_flow_style = False
+        data = yaml.load(config_path)
+        if not data:
+            data = {}
+        if "credentials" not in data:
+            data["credentials"] = {}
+        if service not in data["credentials"]:
+            data["credentials"][service] = {}
+        data["credentials"][service][profile] = credential.dumps()
+        yaml.dump(data, config_path)
+        log.info(f"Added Credential: {credential}")

devine/commands/dl.py

@@ -13,8 +13,8 @@ from concurrent import futures
 from concurrent.futures import ThreadPoolExecutor
 from copy import deepcopy
 from functools import partial
-from http.cookiejar import CookieJar, MozillaCookieJar
+from http.cookiejar import MozillaCookieJar
-from itertools import product
+from itertools import zip_longest
 from pathlib import Path
 from threading import Lock
 from typing import Any, Callable, Optional
@@ -32,7 +32,7 @@ from rich.console import Group
 from rich.live import Live
 from rich.padding import Padding
 from rich.panel import Panel
-from rich.progress import BarColumn, Progress, SpinnerColumn, TaskID, TextColumn, TimeRemainingColumn
+from rich.progress import BarColumn, Progress, SpinnerColumn, TextColumn, TimeRemainingColumn
 from rich.rule import Rule
 from rich.table import Table
 from rich.text import Text
@@ -50,14 +50,14 @@ from devine.core.titles import Movie, Song, Title_T
 from devine.core.titles.episode import Episode
 from devine.core.tracks import Audio, Subtitle, Tracks, Video
 from devine.core.utilities import get_binary_path, is_close_match, time_elapsed_since
-from devine.core.utils.click_types import LANGUAGE_RANGE, QUALITY_LIST, SEASON_RANGE, ContextData, MultipleChoice
+from devine.core.utils.click_types import LANGUAGE_RANGE, QUALITY_LIST, SEASON_RANGE, ContextData
 from devine.core.utils.collections import merge_dict
 from devine.core.utils.subprocess import ffprobe
 from devine.core.vaults import Vaults
 
 
 class dl:
-    @click.command(
+    @click.group(
         short_help="Download, Decrypt, and Mux tracks for titles from a Service.",
         cls=Services,
         context_settings=dict(
@@ -66,12 +66,12 @@ class dl:
             token_normalize_func=Services.get_tag
         ))
     @click.option("-p", "--profile", type=str, default=None,
-                  help="Profile to use for Credentials and Cookies (if available).")
+                  help="Profile to use for Credentials and Cookies (if available). Overrides profile set by config.")
     @click.option("-q", "--quality", type=QUALITY_LIST, default=[],
                   help="Download Resolution(s), defaults to the best available resolution.")
     @click.option("-v", "--vcodec", type=click.Choice(Video.Codec, case_sensitive=False),
-                  default=None,
+                  default=Video.Codec.AVC,
-                  help="Video Codec to download, defaults to any codec.")
+                  help="Video Codec to download, defaults to H.264.")
     @click.option("-a", "--acodec", type=click.Choice(Audio.Codec, case_sensitive=False),
                   default=None,
                   help="Audio Codec to download, defaults to any codec.")
@@ -81,9 +81,9 @@ class dl:
     @click.option("-ab", "--abitrate", type=int,
                   default=None,
                   help="Audio Bitrate to download (in kbps), defaults to highest available.")
-    @click.option("-r", "--range", "range_", type=MultipleChoice(Video.Range, case_sensitive=False),
+    @click.option("-r", "--range", "range_", type=click.Choice(Video.Range, case_sensitive=False),
-                  default=[Video.Range.SDR],
+                  default=Video.Range.SDR,
-                  help="Video Color Range(s) to download, defaults to SDR.")
+                  help="Video Color Range, defaults to SDR.")
     @click.option("-c", "--channels", type=float,
                   default=None,
                   help="Audio Channel(s) to download. Matches sub-channel layouts like 5.1 with 6.0 implicitly.")
@@ -97,10 +97,10 @@ class dl:
                   help="Language wanted for Subtitles.")
     @click.option("--proxy", type=str, default=None,
                   help="Proxy URI to use. If a 2-letter country is provided, it will try get a proxy from the config.")
-    @click.option("--tag", type=str, default=None,
+    @click.option("--group", type=str, default=None,
                   help="Set the Group Tag to be used, overriding the one in config if any.")
     @click.option("--sub-format", type=click.Choice(Subtitle.Codec, case_sensitive=False),
-                  default=None,
+                  default=Subtitle.Codec.SubRip,
                   help="Set Output Subtitle Format, only converting if necessary.")
     @click.option("-V", "--video-only", is_flag=True, default=False,
                   help="Only download video tracks.")
@@ -143,7 +143,7 @@ class dl:
         no_proxy: bool,
         profile: Optional[str] = None,
         proxy: Optional[str] = None,
-        tag: Optional[str] = None,
+        group: Optional[str] = None,
         *_: Any,
         **__: Any
     ):
@@ -153,14 +153,17 @@ class dl:
         self.log = logging.getLogger("download")
 
         self.service = Services.get_tag(ctx.invoked_subcommand)
+
+        with console.status("Preparing Service and Profile Authentication...", spinner="dots"):
+            if profile:
                 self.profile = profile
+                self.log.info(f"Profile: '{self.profile}' from the --profile argument")
+            else:
+                self.profile = self.get_profile(self.service)
+                self.log.info(f"Profile: '{self.profile}' from the config")
 
-        if self.profile:
-            self.log.info(f"Using profile: '{self.profile}'")
-
-        with console.status("Loading Service Config...", spinner="dots"):
             service_config_path = Services.get_path(self.service) / config.filenames.config
-            if service_config_path.exists():
+            if service_config_path.is_file():
                 self.service_config = yaml.safe_load(service_config_path.read_text(encoding="utf8"))
                 self.log.info("Service Config loaded")
             else:
@@ -239,8 +242,8 @@ class dl:
             profile=self.profile
         )
 
-        if tag:
+        if group:
-            config.tag = tag
+            config.tag = group
 
         # needs to be added this way instead of @cli.result_callback to be
         # able to keep `self` as the first positional
@@ -250,17 +253,17 @@ class dl:
         self,
         service: Service,
         quality: list[int],
-        vcodec: Optional[Video.Codec],
+        vcodec: Video.Codec,
         acodec: Optional[Audio.Codec],
         vbitrate: int,
         abitrate: int,
-        range_: list[Video.Range],
+        range_: Video.Range,
         channels: float,
         wanted: list[str],
         lang: list[str],
         v_lang: list[str],
         s_lang: list[str],
-        sub_format: Optional[Subtitle.Codec],
+        sub_format: Subtitle.Codec,
         video_only: bool,
         audio_only: bool,
         subs_only: bool,
@@ -284,11 +287,14 @@ class dl:
         else:
             vaults_only = not cdm_only
 
+        if self.profile:
             with console.status("Authenticating with Service...", spinner="dots"):
                 cookies = self.get_cookie_jar(self.service, self.profile)
                 credential = self.get_credentials(self.service, self.profile)
+                if not cookies and not credential:
+                    self.log.error(f"The Profile '{self.profile}' has no Cookies or Credentials, Check for typos")
+                    sys.exit(1)
                 service.authenticate(cookies, credential)
-            if cookies or credential:
                 self.log.info("Authenticated with Service")
 
         with console.status("Fetching Title Metadata...", spinner="dots"):
@@ -325,7 +331,7 @@ class dl:
 
             with console.status("Getting tracks...", spinner="dots"):
                 title.tracks.add(service.get_tracks(title), warn_only=True)
-                title.tracks.chapters = service.get_chapters(title)
+                title.tracks.add(service.get_chapters(title))
 
             # strip SDH subs to non-SDH if no equivalent same-lang non-SDH is available
             # uses a loose check, e.g, wont strip en-US SDH sub if a non-SDH en-GB is available
@@ -338,13 +344,14 @@ class dl:
                     non_sdh_sub = deepcopy(subtitle)
                     non_sdh_sub.id += "_stripped"
                     non_sdh_sub.sdh = False
-                    non_sdh_sub.OnMultiplex = lambda: non_sdh_sub.strip_hearing_impaired()
+                    non_sdh_sub.OnMultiplex = lambda x: x.strip_hearing_impaired()
                     title.tracks.add(non_sdh_sub)
 
             with console.status("Sorting tracks by language and bitrate...", spinner="dots"):
                 title.tracks.sort_videos(by_language=v_lang or lang)
                 title.tracks.sort_audio(by_language=lang)
                 title.tracks.sort_subtitles(by_language=s_lang)
+                title.tracks.sort_chapters()
 
             if list_:
                 available_tracks, _ = title.tracks.tree()
@@ -357,17 +364,14 @@ class dl:
             with console.status("Selecting tracks...", spinner="dots"):
                 if isinstance(title, (Movie, Episode)):
                     # filter video tracks
-                    if vcodec:
                     title.tracks.select_video(lambda x: x.codec == vcodec)
                     if not title.tracks.videos:
                         self.log.error(f"There's no {vcodec.name} Video Track...")
                         sys.exit(1)
 
-                    if range_:
+                    title.tracks.select_video(lambda x: x.range == range_)
-                        title.tracks.select_video(lambda x: x.range in range_)
+                    if not title.tracks.videos:
-                        for color_range in range_:
+                        self.log.error(f"There's no {range_.name} Video Track...")
-                            if not any(x.range == color_range for x in title.tracks.videos):
-                                self.log.error(f"There's no {color_range.name} Video Tracks...")
                         sys.exit(1)
 
                     if vbitrate:
@@ -384,7 +388,7 @@ class dl:
                             sys.exit(1)
 
                     if quality:
-                        title.tracks.by_resolutions(quality)
+                        title.tracks.by_resolutions(quality, per_resolution=1)
                         missing_resolutions = []
                         for resolution in quality:
                             if any(video.height == resolution for video in title.tracks.videos):
@@ -400,27 +404,8 @@ class dl:
                             plural = "s" if len(missing_resolutions) > 1 else ""
                             self.log.error(f"There's no {res_list} Video Track{plural}...")
                             sys.exit(1)
-
+                    else:
-                    # choose best track by range and quality
+                        title.tracks.videos = [title.tracks.videos[0]]
-                    title.tracks.videos = [
-                        track
-                        for resolution, color_range in product(
-                            quality or [None],
-                            range_ or [None]
-                        )
-                        for track in [next(
-                            t
-                            for t in title.tracks.videos
-                            if (not resolution and not color_range) or
-                            (
-                                (not resolution or (
-                                   (t.height == resolution) or
-                                   (int(t.width * (9 / 16)) == resolution)
-                                ))
-                                and (not color_range or t.range == color_range)
-                            )
-                        )]
-                    ]
 
                     # filter subtitle tracks
                     if s_lang and "all" not in s_lang:
@@ -591,7 +576,6 @@ class dl:
                             break
                     video_track_n += 1
 
-                if sub_format:
                 with console.status(f"Converting Subtitles to {sub_format.name}..."):
                     for subtitle in title.tracks.subtitles:
                         if subtitle.codec != sub_format:
@@ -604,7 +588,7 @@ class dl:
                             track.repackage()
                             has_repacked = True
                             if callable(track.OnRepacked):
-                                track.OnRepacked()
+                                track.OnRepacked(track)
                     if has_repacked:
                         # we don't want to fill up the log with "Repacked x track"
                         self.log.info("Repacked one or more tracks with FFMPEG")
@@ -620,33 +604,26 @@ class dl:
                         TimeRemainingColumn(compact=True, elapsed_when_finished=True),
                         console=console
                     )
-
+                    multi_jobs = len(title.tracks.videos) > 1
-                    multiplex_tasks: list[tuple[TaskID, Tracks]] = []
+                    tasks = [
-                    for video_track in title.tracks.videos or [None]:
+                        progress.add_task(
-                        task_description = "Multiplexing"
+                            f"Multiplexing{f' {x.height}p' if multi_jobs else ''}...",
-                        if video_track:
+                            total=None,
-                            if len(quality) > 1:
+                            start=False
-                                task_description += f" {video_track.height}p"
+                        )
-                            if len(range_) > 1:
+                        for x in title.tracks.videos or [None]
-                                task_description += f" {video_track.range.name}"
+                    ]
-
-                        task_id = progress.add_task(f"{task_description}...", total=None, start=False)
-
-                        task_tracks = Tracks(title.tracks) + title.tracks.chapters
-                        if video_track:
-                            task_tracks.videos = [video_track]
-
-                        multiplex_tasks.append((task_id, task_tracks))
-
                     with Live(
                         Padding(progress, (0, 5, 1, 5)),
                         console=console
                     ):
-                        for task_id, task_tracks in multiplex_tasks:
+                        for task, video_track in zip_longest(tasks, title.tracks.videos, fillvalue=None):
-                            progress.start_task(task_id)  # TODO: Needed?
+                            if video_track:
-                            muxed_path, return_code = task_tracks.mux(
+                                title.tracks.videos = [video_track]
+                            progress.start_task(task)  # TODO: Needed?
+                            muxed_path, return_code = title.tracks.mux(
                                 str(title),
-                                progress=partial(progress.update, task_id=task_id),
+                                progress=partial(progress.update, task_id=task),
                                 delete=False
                             )
                             muxed_paths.append(muxed_path)
@@ -655,7 +632,7 @@ class dl:
                             elif return_code >= 2:
                                 self.log.error(f"Failed to Mux video to Matroska file ({return_code})")
                                 sys.exit(1)
-                            for video_track in task_tracks.videos:
+                            if video_track:
                                 video_track.delete()
                         for track in title.tracks:
                             track.delete()
@@ -683,9 +660,13 @@ class dl:
                 ))
 
             # update cookies
-            cookie_file = self.get_cookie_path(self.service, self.profile)
+            cookie_file = config.directories.cookies / service.__class__.__name__ / f"{self.profile}.txt"
-            if cookie_file:
+            if cookie_file.exists():
-                self.save_cookies(cookie_file, service.session.cookies)
+                cookie_jar = MozillaCookieJar(cookie_file)
+                cookie_jar.load()
+                for cookie in service.session.cookies:
+                    cookie_jar.set_cookie(cookie)
+                cookie_jar.save(ignore_discard=True)
 
         dl_time = time_elapsed_since(start_time)
 
@@ -786,11 +767,8 @@ class dl:
                         # So we re-add the keys from vaults earlier overwriting blanks or removed KIDs data.
                         drm.content_keys.update(from_vaults)
 
-                        successful_caches = self.vaults.add_keys(drm.content_keys)
+                        cached_keys = self.vaults.add_keys(drm.content_keys)
-                        self.log.info(
+                        self.log.info(f" + Newly added to {cached_keys}/{len(drm.content_keys)} Vaults")
-                            f"Cached {len(drm.content_keys)} Key{'' if len(drm.content_keys) == 1 else 's'} to "
-                            f"{successful_caches}/{len(self.vaults)} Vaults"
-                        )
                         break  # licensing twice will be unnecessary
 
                 if track_kid and track_kid not in drm.content_keys:
@@ -816,24 +794,22 @@ class dl:
                     export.write_text(jsonpickle.dumps(keys, indent=4), encoding="utf8")
 
     @staticmethod
-    def get_cookie_path(service: str, profile: Optional[str]) -> Optional[Path]:
+    def get_profile(service: str) -> Optional[str]:
-        """Get Service Cookie File Path for Profile."""
+        """Get profile for Service from config."""
-        direct_cookie_file = config.directories.cookies / f"{service}.txt"
+        profile = config.profiles.get(service)
-        profile_cookie_file = config.directories.cookies / service / f"{profile}.txt"
+        if profile is False:
-        default_cookie_file = config.directories.cookies / service / "default.txt"
+            return None  # auth-less service if `false` in config
-
+        if not profile:
-        if direct_cookie_file.exists():
+            profile = config.profiles.get("default")
-            return direct_cookie_file
+        if not profile:
-        elif profile_cookie_file.exists():
+            raise ValueError(f"No profile has been defined for '{service}' in the config.")
-            return profile_cookie_file
+        return profile
-        elif default_cookie_file.exists():
-            return default_cookie_file
 
     @staticmethod
-    def get_cookie_jar(service: str, profile: Optional[str]) -> Optional[MozillaCookieJar]:
+    def get_cookie_jar(service: str, profile: str) -> Optional[MozillaCookieJar]:
-        """Get Service Cookies for Profile."""
+        """Get Profile's Cookies as Mozilla Cookie Jar if available."""
-        cookie_file = dl.get_cookie_path(service, profile)
+        cookie_file = config.directories.cookies / service / f"{profile}.txt"
-        if cookie_file:
+        if cookie_file.is_file():
             cookie_jar = MozillaCookieJar(cookie_file)
             cookie_data = html.unescape(cookie_file.read_text("utf8")).splitlines(keepends=False)
             for i, line in enumerate(cookie_data):
@@ -848,29 +824,17 @@ class dl:
             cookie_file.write_text(cookie_data, "utf8")
             cookie_jar.load(ignore_discard=True, ignore_expires=True)
             return cookie_jar
+        return None
 
     @staticmethod
-    def save_cookies(path: Path, cookies: CookieJar):
+    def get_credentials(service: str, profile: str) -> Optional[Credential]:
-        cookie_jar = MozillaCookieJar(path)
+        """Get Profile's Credential if available."""
-        cookie_jar.load()
+        cred = config.credentials.get(service, {}).get(profile)
-        for cookie in cookies:
+        if cred:
-            cookie_jar.set_cookie(cookie)
+            if isinstance(cred, list):
-        cookie_jar.save(ignore_discard=True)
+                return Credential(*cred)
-
+            return Credential.loads(cred)
-    @staticmethod
+        return None
-    def get_credentials(service: str, profile: Optional[str]) -> Optional[Credential]:
-        """Get Service Credentials for Profile."""
-        credentials = config.credentials.get(service)
-        if credentials:
-            if isinstance(credentials, dict):
-                if profile:
-                    credentials = credentials.get(profile) or credentials.get("default")
-                else:
-                    credentials = credentials.get("default")
-            if credentials:
-                if isinstance(credentials, list):
-                    return Credential(*credentials)
-                return Credential.loads(credentials)  # type: ignore
 
     @staticmethod
     def get_cdm(service: str, profile: Optional[str] = None) -> WidevineCdm:

devine/commands/search.py

@@ -1,166 +0,0 @@
-from __future__ import annotations
-
-import logging
-import re
-import sys
-from typing import Any, Optional
-
-import click
-import yaml
-from rich.padding import Padding
-from rich.rule import Rule
-from rich.tree import Tree
-
-from devine.commands.dl import dl
-from devine.core.config import config
-from devine.core.console import console
-from devine.core.constants import context_settings
-from devine.core.proxies import Basic, Hola, NordVPN
-from devine.core.service import Service
-from devine.core.services import Services
-from devine.core.utilities import get_binary_path
-from devine.core.utils.click_types import ContextData
-from devine.core.utils.collections import merge_dict
-
-
-@click.command(
-    short_help="Search for titles from a Service.",
-    cls=Services,
-    context_settings=dict(
-        **context_settings,
-        token_normalize_func=Services.get_tag
-    ))
-@click.option("-p", "--profile", type=str, default=None,
-              help="Profile to use for Credentials and Cookies (if available).")
-@click.option("--proxy", type=str, default=None,
-              help="Proxy URI to use. If a 2-letter country is provided, it will try get a proxy from the config.")
-@click.option("--no-proxy", is_flag=True, default=False,
-              help="Force disable all proxy use.")
-@click.pass_context
-def search(
-    ctx: click.Context,
-    no_proxy: bool,
-    profile: Optional[str] = None,
-    proxy: Optional[str] = None
-):
-    if not ctx.invoked_subcommand:
-        raise ValueError("A subcommand to invoke was not specified, the main code cannot continue.")
-
-    log = logging.getLogger("search")
-
-    service = Services.get_tag(ctx.invoked_subcommand)
-    profile = profile
-
-    if profile:
-        log.info(f"Using profile: '{profile}'")
-
-    with console.status("Loading Service Config...", spinner="dots"):
-        service_config_path = Services.get_path(service) / config.filenames.config
-        if service_config_path.exists():
-            service_config = yaml.safe_load(service_config_path.read_text(encoding="utf8"))
-            log.info("Service Config loaded")
-        else:
-            service_config = {}
-        merge_dict(config.services.get(service), service_config)
-
-    proxy_providers = []
-    if no_proxy:
-        ctx.params["proxy"] = None
-    else:
-        with console.status("Loading Proxy Providers...", spinner="dots"):
-            if config.proxy_providers.get("basic"):
-                proxy_providers.append(Basic(**config.proxy_providers["basic"]))
-            if config.proxy_providers.get("nordvpn"):
-                proxy_providers.append(NordVPN(**config.proxy_providers["nordvpn"]))
-            if get_binary_path("hola-proxy"):
-                proxy_providers.append(Hola())
-            for proxy_provider in proxy_providers:
-                log.info(f"Loaded {proxy_provider.__class__.__name__}: {proxy_provider}")
-
-        if proxy:
-            requested_provider = None
-            if re.match(r"^[a-z]+:.+$", proxy, re.IGNORECASE):
-                # requesting proxy from a specific proxy provider
-                requested_provider, proxy = proxy.split(":", maxsplit=1)
-            if re.match(r"^[a-z]{2}(?:\d+)?$", proxy, re.IGNORECASE):
-                proxy = proxy.lower()
-                with console.status(f"Getting a Proxy to {proxy}...", spinner="dots"):
-                    if requested_provider:
-                        proxy_provider = next((
-                            x
-                            for x in proxy_providers
-                            if x.__class__.__name__.lower() == requested_provider
-                        ), None)
-                        if not proxy_provider:
-                            log.error(f"The proxy provider '{requested_provider}' was not recognised.")
-                            sys.exit(1)
-                        proxy_uri = proxy_provider.get_proxy(proxy)
-                        if not proxy_uri:
-                            log.error(f"The proxy provider {requested_provider} had no proxy for {proxy}")
-                            sys.exit(1)
-                        proxy = ctx.params["proxy"] = proxy_uri
-                        log.info(f"Using {proxy_provider.__class__.__name__} Proxy: {proxy}")
-                    else:
-                        for proxy_provider in proxy_providers:
-                            proxy_uri = proxy_provider.get_proxy(proxy)
-                            if proxy_uri:
-                                proxy = ctx.params["proxy"] = proxy_uri
-                                log.info(f"Using {proxy_provider.__class__.__name__} Proxy: {proxy}")
-                                break
-            else:
-                log.info(f"Using explicit Proxy: {proxy}")
-
-    ctx.obj = ContextData(
-        config=service_config,
-        cdm=None,
-        proxy_providers=proxy_providers,
-        profile=profile
-    )
-
-
-@search.result_callback()
-def result(service: Service, profile: Optional[str] = None, **_: Any) -> None:
-    log = logging.getLogger("search")
-
-    service_tag = service.__class__.__name__
-
-    with console.status("Authenticating with Service...", spinner="dots"):
-        cookies = dl.get_cookie_jar(service_tag, profile)
-        credential = dl.get_credentials(service_tag, profile)
-        service.authenticate(cookies, credential)
-        if cookies or credential:
-            log.info("Authenticated with Service")
-
-    search_results = Tree("Search Results", hide_root=True)
-    with console.status("Searching...", spinner="dots"):
-        for result in service.search():
-            result_text = f"[bold text]{result.title}[/]"
-            if result.url:
-                result_text = f"[link={result.url}]{result_text}[/link]"
-            if result.label:
-                result_text += f"  [pink]{result.label}[/]"
-            if result.description:
-                result_text += f"\n[text2]{result.description}[/]"
-            result_text += f"\n[bright_black]id: {result.id}[/]"
-            search_results.add(result_text + "\n")
-
-    # update cookies
-    cookie_file = dl.get_cookie_path(service_tag, profile)
-    if cookie_file:
-        dl.save_cookies(cookie_file, service.session.cookies)
-
-    console.print(Padding(
-        Rule(f"[rule.text]{len(search_results.children)} Search Results"),
-        (1, 2)
-    ))
-
-    if search_results.children:
-        console.print(Padding(
-            search_results,
-            (0, 5)
-        ))
-    else:
-        console.print(Padding(
-            "[bold text]No matches[/]\n[bright_black]Please check spelling and search again....[/]",
-            (0, 5)
-        ))

devine/core/__init__.py

@@ -1 +1 @@
-__version__ = "3.1.0"
+__version__ = "2.1.0"

devine/core/__main__.py

@@ -27,7 +27,7 @@ LOGGING_PATH = None
 @click.option("--log", "log_path", type=Path, default=config.directories.logs / config.filenames.log,
               help="Log path (or filename). Path can contain the following f-string args: {name} {time}.")
 def main(version: bool, debug: bool, log_path: Path) -> None:
-    """Devine—Modular Movie, TV, and Music Archival Software."""
+    """Devine—Open-Source Movie, TV, and Music Downloading Solution."""
     logging.basicConfig(
         level=logging.DEBUG if debug else logging.INFO,
         format="%(message)s",

devine/core/config.py

@@ -2,7 +2,7 @@ from __future__ import annotations
 
 import tempfile
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any
 
 import yaml
 from appdirs import AppDirs
@@ -39,7 +39,6 @@ class Config:
         self.dl: dict = kwargs.get("dl") or {}
         self.aria2c: dict = kwargs.get("aria2c") or {}
         self.cdm: dict = kwargs.get("cdm") or {}
-        self.chapter_fallback_name: str = kwargs.get("chapter_fallback_name") or ""
         self.curl_impersonate: dict = kwargs.get("curl_impersonate") or {}
         self.remote_cdm: list[dict] = kwargs.get("remote_cdm") or []
         self.credentials: dict = kwargs.get("credentials") or {}
@@ -51,7 +50,7 @@ class Config:
                 continue
             setattr(self.directories, name, Path(path).expanduser())
 
-        self.downloader = kwargs.get("downloader") or "requests"
+        self.downloader = kwargs.get("downloader") or "aria2c"
 
         self.filenames = self._Filenames()
         for name, filename in (kwargs.get("filenames") or {}).items():
@@ -61,6 +60,7 @@ class Config:
         self.key_vaults: list[dict[str, Any]] = kwargs.get("key_vaults", [])
         self.muxing: dict = kwargs.get("muxing") or {}
         self.nordvpn: dict = kwargs.get("nordvpn") or {}
+        self.profiles: dict = kwargs.get("profiles") or {}
         self.proxy_providers: dict = kwargs.get("proxy_providers") or {}
         self.serve: dict = kwargs.get("serve") or {}
         self.services: dict = kwargs.get("services") or {}
@@ -76,35 +76,11 @@ class Config:
         return cls(**yaml.safe_load(path.read_text(encoding="utf8")) or {})
 
 
-def get_config_path() -> Optional[Path]:
-    """
-    Get Path to Config from various locations.
-
-    Looks for a config file in the following folders in order:
-
-    1. The Devine Namespace Folder (e.g., %appdata%/Python/Python311/site-packages/devine)
-    2. The Parent Folder to the Devine Namespace Folder (e.g., %appdata%/Python/Python311/site-packages)
-    3. The AppDirs User Config Folder (e.g., %localappdata%/devine)
-
-    Returns None if no config file could be found.
-    """
 # noinspection PyProtectedMember
-    path = Config._Directories.namespace_dir / Config._Filenames.root_config
+config_path = Config._Directories.user_configs / Config._Filenames.root_config
-    if not path.exists():
+if not config_path.is_file():
-        # noinspection PyProtectedMember
+    Config._Directories.user_configs.mkdir(parents=True, exist_ok=True)
-        path = Config._Directories.namespace_dir.parent / Config._Filenames.root_config
+    config_path.write_text("")
-    if not path.exists():
-        # noinspection PyProtectedMember
-        path = Config._Directories.user_configs / Config._Filenames.root_config
-    if not path.exists():
-        path = None
-    return path
-
-
-config_path = get_config_path()
-if config_path:
 config = Config.from_yaml(config_path)
-else:
-    config = Config()
 
 __all__ = ("config",)

devine/core/constants.py

@@ -5,6 +5,21 @@ DOWNLOAD_CANCELLED = Event()
 DOWNLOAD_LICENCE_ONLY = Event()
 
 DRM_SORT_MAP = ["ClearKey", "Widevine"]
+LANGUAGE_MUX_MAP = {
+    # List of language tags that cannot be used by mkvmerge and need replacements.
+    # Try get the replacement to be as specific locale-wise as possible.
+    # A bcp47 as the replacement is recommended.
+    "cmn": "zh",
+    "cmn-Hant": "zh-Hant",
+    "cmn-Hans": "zh-Hans",
+    "none": "und",
+    "yue": "zh-yue",
+    "yue-Hant": "zh-yue-Hant",
+    "yue-Hans": "zh-yue-Hans"
+}
+TERRITORY_MAP = {
+    "Hong Kong SAR China": "Hong Kong"
+}
 LANGUAGE_MAX_DISTANCE = 5  # this is max to be considered "same", e.g., en, en-US, en-AU
 VIDEO_CODEC_MAP = {
     "AVC": "H.264",

devine/core/downloaders/__init__.py

@@ -1,5 +1,15 @@
+import asyncio
+
+from ..config import config
 from .aria2c import aria2c
 from .curl_impersonate import curl_impersonate
 from .requests import requests
 
-__all__ = ("aria2c", "curl_impersonate", "requests")
+downloader = {
+    "aria2c": lambda *args, **kwargs: asyncio.run(aria2c(*args, **kwargs)),
+    "curl_impersonate": curl_impersonate,
+    "requests": requests
+}[config.downloader]
+
+
+__all__ = ("downloader", "aria2c", "curl_impersonate", "requests")

devine/core/downloaders/aria2c.py

@@ -1,149 +1,84 @@
-import os
+import asyncio
 import subprocess
 import textwrap
-import time
 from functools import partial
 from http.cookiejar import CookieJar
 from pathlib import Path
-from typing import Any, Callable, Generator, MutableMapping, Optional, Union
+from typing import MutableMapping, Optional, Union
-from urllib.parse import urlparse
 
 import requests
-from Crypto.Random import get_random_bytes
+from requests.cookies import RequestsCookieJar, cookiejar_from_dict, get_cookie_header
-from requests import Session
-from requests.cookies import cookiejar_from_dict, get_cookie_header
-from rich import filesize
 from rich.text import Text
 
 from devine.core.config import config
 from devine.core.console import console
-from devine.core.constants import DOWNLOAD_CANCELLED
+from devine.core.utilities import get_binary_path, start_pproxy
-from devine.core.utilities import get_binary_path, get_extension, get_free_port
 
 
-def rpc(caller: Callable, secret: str, method: str, params: Optional[list[Any]] = None) -> Any:
+async def aria2c(
-    """Make a call to Aria2's JSON-RPC API."""
+    uri: Union[str, list[str]],
-    try:
+    out: Path,
-        rpc_res = caller(
+    headers: Optional[dict] = None,
-            json={
+    cookies: Optional[Union[MutableMapping[str, str], RequestsCookieJar]] = None,
-                "jsonrpc": "2.0",
-                "id": get_random_bytes(16).hex(),
-                "method": method,
-                "params": [f"token:{secret}", *(params or [])]
-            }
-        ).json()
-        if rpc_res.get("code"):
-            # wrap to console width - padding - '[Aria2c]: '
-            error_pretty = "\n          ".join(textwrap.wrap(
-                f"RPC Error: {rpc_res['message']} ({rpc_res['code']})".strip(),
-                width=console.width - 20,
-                initial_indent=""
-            ))
-            console.log(Text.from_ansi("\n[Aria2c]: " + error_pretty))
-        return rpc_res["result"]
-    except requests.exceptions.ConnectionError:
-        # absorb, process likely ended as it was calling RPC
-        return
-
-
-def download(
-    urls: Union[str, list[str], dict[str, Any], list[dict[str, Any]]],
-    output_dir: Path,
-    filename: str,
-    headers: Optional[MutableMapping[str, Union[str, bytes]]] = None,
-    cookies: Optional[Union[MutableMapping[str, str], CookieJar]] = None,
     proxy: Optional[str] = None,
-    max_workers: Optional[int] = None
+    silent: bool = False,
-) -> Generator[dict[str, Any], None, None]:
+    segmented: bool = False,
-    if not urls:
+    progress: Optional[partial] = None,
-        raise ValueError("urls must be provided and not empty")
+    *args: str
-    elif not isinstance(urls, (str, dict, list)):
+) -> int:
-        raise TypeError(f"Expected urls to be {str} or {dict} or a list of one of them, not {type(urls)}")
+    """
+    Download files using Aria2(c).
+    https://aria2.github.io
 
-    if not output_dir:
+    If multiple URLs are provided they will be downloaded in the provided order
-        raise ValueError("output_dir must be provided")
+    to the output directory. They will not be merged together.
-    elif not isinstance(output_dir, Path):
+    """
-        raise TypeError(f"Expected output_dir to be {Path}, not {type(output_dir)}")
+    if not isinstance(uri, list):
+        uri = [uri]
 
-    if not filename:
+    if cookies and not isinstance(cookies, CookieJar):
-        raise ValueError("filename must be provided")
+        cookies = cookiejar_from_dict(cookies)
-    elif not isinstance(filename, str):
-        raise TypeError(f"Expected filename to be {str}, not {type(filename)}")
-
-    if not isinstance(headers, (MutableMapping, type(None))):
-        raise TypeError(f"Expected headers to be {MutableMapping}, not {type(headers)}")
-
-    if not isinstance(cookies, (MutableMapping, CookieJar, type(None))):
-        raise TypeError(f"Expected cookies to be {MutableMapping} or {CookieJar}, not {type(cookies)}")
-
-    if not isinstance(proxy, (str, type(None))):
-        raise TypeError(f"Expected proxy to be {str}, not {type(proxy)}")
-
-    if not max_workers:
-        max_workers = min(32, (os.cpu_count() or 1) + 4)
-    elif not isinstance(max_workers, int):
-        raise TypeError(f"Expected max_workers to be {int}, not {type(max_workers)}")
-
-    if not isinstance(urls, list):
-        urls = [urls]
 
     executable = get_binary_path("aria2c", "aria2")
     if not executable:
         raise EnvironmentError("Aria2c executable not found...")
 
-    if proxy and not proxy.lower().startswith("http://"):
+    if proxy and proxy.lower().split(":")[0] != "http":
-        raise ValueError("Only HTTP proxies are supported by aria2(c)")
+        # HTTPS proxies are not supported by aria2(c).
-
+        # Proxy the proxy via pproxy to access it as an HTTP proxy.
-    if cookies and not isinstance(cookies, CookieJar):
+        async with start_pproxy(proxy) as pproxy_:
-        cookies = cookiejar_from_dict(cookies)
+            return await aria2c(uri, out, headers, cookies, pproxy_, silent, segmented, progress, *args)
 
+    multiple_urls = len(uri) > 1
     url_files = []
-    for i, url in enumerate(urls):
+    for i, url in enumerate(uri):
-        if isinstance(url, str):
+        url_text = url
-            url_data = {
+        if multiple_urls:
-                "url": url
+            url_text += f"\n\tdir={out}"
-            }
+            url_text += f"\n\tout={i:08}.mp4"
         else:
-            url_data: dict[str, Any] = url
+            url_text += f"\n\tdir={out.parent}"
-        url_filename = filename.format(
+            url_text += f"\n\tout={out.name}"
-            i=i,
-            ext=get_extension(url_data["url"])
-        )
-        url_text = url_data["url"]
-        url_text += f"\n\tdir={output_dir}"
-        url_text += f"\n\tout={url_filename}"
         if cookies:
-            mock_request = requests.Request(url=url_data["url"])
+            mock_request = requests.Request(url=url)
             cookie_header = get_cookie_header(cookies, mock_request)
             if cookie_header:
                 url_text += f"\n\theader=Cookie: {cookie_header}"
-        for key, value in url_data.items():
-            if key == "url":
-                continue
-            if key == "headers":
-                for header_name, header_value in value.items():
-                    url_text += f"\n\theader={header_name}: {header_value}"
-            else:
-                url_text += f"\n\t{key}={value}"
         url_files.append(url_text)
     url_file = "\n".join(url_files)
 
-    rpc_port = get_free_port()
+    max_concurrent_downloads = int(config.aria2c.get("max_concurrent_downloads", 5))
-    rpc_secret = get_random_bytes(16).hex()
-    rpc_uri = f"http://127.0.0.1:{rpc_port}/jsonrpc"
-    rpc_session = Session()
-
-    max_concurrent_downloads = int(config.aria2c.get("max_concurrent_downloads", max_workers))
     max_connection_per_server = int(config.aria2c.get("max_connection_per_server", 1))
     split = int(config.aria2c.get("split", 5))
     file_allocation = config.aria2c.get("file_allocation", "prealloc")
-    if len(urls) > 1:
+    if segmented:
         split = 1
         file_allocation = "none"
 
     arguments = [
         # [Basic Options]
         "--input-file", "-",
+        "--out", out.name,
         "--all-proxy", proxy or "",
         "--continue=true",
         # [Connection Options]
@@ -157,13 +92,11 @@ def download(
         "--allow-overwrite=true",
         "--auto-file-renaming=false",
         "--console-log-level=warn",
-        "--download-result=default",
+        f"--download-result={'default' if progress else 'hide'}",
         f"--file-allocation={file_allocation}",
         "--summary-interval=0",
-        # [RPC Options]
+        # [Extra Options]
-        "--enable-rpc=true",
+        *args
-        f"--rpc-listen-port={rpc_port}",
-        f"--rpc-secret={rpc_secret}"
     ]
 
     for header, value in (headers or {}).items():
@@ -181,72 +114,67 @@ def download(
             continue
         arguments.extend(["--header", f"{header}: {value}"])
 
-    yield dict(total=len(urls))
-
     try:
-        p = subprocess.Popen(
+        p = await asyncio.create_subprocess_exec(
-            [
             executable,
-                *arguments
+            *arguments,
-            ],
             stdin=subprocess.PIPE,
-            stdout=subprocess.DEVNULL
+            stdout=subprocess.PIPE
         )
 
         p.stdin.write(url_file.encode())
+        await p.stdin.drain()
         p.stdin.close()
 
-        while p.poll() is None:
+        if p.stdout:
-            global_stats: dict[str, Any] = rpc(
+            is_dl_summary = False
-                caller=partial(rpc_session.post, url=rpc_uri),
+            log_buffer = ""
-                secret=rpc_secret,
+            while True:
-                method="aria2.getGlobalStat"
+                try:
-            ) or {}
+                    chunk = await p.stdout.readuntil(b"\r")
-
+                except asyncio.IncompleteReadError as e:
-            number_stopped = int(global_stats.get("numStoppedTotal", 0))
+                    chunk = e.partial
-            download_speed = int(global_stats.get("downloadSpeed", -1))
+                if not chunk:
-
+                    break
-            if number_stopped:
+                for line in chunk.decode().strip().splitlines():
-                yield dict(completed=number_stopped)
+                    if not line:
-            if download_speed != -1:
+                        continue
-                yield dict(downloaded=f"{filesize.decimal(download_speed)}/s")
+                    if line.startswith("Download Results"):
-
+                        # we know it's 100% downloaded, but let's use the avg dl speed value
-            stopped_downloads: list[dict[str, Any]] = rpc(
+                        is_dl_summary = True
-                caller=partial(rpc_session.post, url=rpc_uri),
+                    elif line.startswith("[") and line.endswith("]"):
-                secret=rpc_secret,
+                        if progress and "%" in line:
-                method="aria2.tellStopped",
+                            # id, dledMiB/totalMiB(x%), CN:xx, DL:xxMiB, ETA:Xs
-                params=[0, 999999]
+                            # eta may not always be available
-            ) or []
+                            data_parts = line[1:-1].split()
-
+                            perc_parts = data_parts[1].split("(")
-            for dl in stopped_downloads:
+                            if len(perc_parts) == 2:
-                if dl["status"] == "error":
+                                # might otherwise be e.g., 0B/0B, with no % symbol provided
-                    used_uri = next(
+                                progress(
-                        uri["uri"]
+                                    total=100,
-                        for file in dl["files"]
+                                    completed=int(perc_parts[1][:-2]),
-                        if file["selected"] == "true"
+                                    downloaded=f"{data_parts[3].split(':')[1]}/s"
-                        for uri in file["uris"]
-                        if uri["status"] == "used"
                                 )
-                    error = f"Download Error (#{dl['gid']}): {dl['errorMessage']} ({dl['errorCode']}), {used_uri}"
+                    elif is_dl_summary and "OK" in line and "|" in line:
-                    error_pretty = "\n          ".join(textwrap.wrap(
+                        gid, status, avg_speed, path_or_uri = line.split("|")
-                        error,
+                        progress(total=100, completed=100, downloaded=avg_speed.strip())
+                    elif not is_dl_summary:
+                        if "aria2 will resume download if the transfer is restarted" in line:
+                            continue
+                        if "If there are any errors, then see the log file" in line:
+                            continue
+                        log_buffer += f"{line.strip()}\n"
+
+            if log_buffer and not silent:
+                # wrap to console width - padding - '[Aria2c]: '
+                log_buffer = "\n          ".join(textwrap.wrap(
+                    log_buffer.rstrip(),
                     width=console.width - 20,
                     initial_indent=""
                 ))
-                    console.log(Text.from_ansi("\n[Aria2c]: " + error_pretty))
+                console.log(Text.from_ansi("\n[Aria2c]: " + log_buffer))
-                    raise ValueError(error)
 
-            if number_stopped == len(urls):
+        await p.wait()
-                rpc(
-                    caller=partial(rpc_session.post, url=rpc_uri),
-                    secret=rpc_secret,
-                    method="aria2.shutdown"
-                )
-                break
-
-            time.sleep(1)
-
-        p.wait()
 
         if p.returncode != 0:
             raise subprocess.CalledProcessError(p.returncode, arguments)
@@ -259,96 +187,8 @@ def download(
             # 0xC000013A is when it never got the chance to
             raise KeyboardInterrupt()
         raise
-    except KeyboardInterrupt:
-        DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-        yield dict(downloaded="[yellow]CANCELLED")
-        raise
-    except Exception:
-        DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-        yield dict(downloaded="[red]FAILED")
-        raise
-    finally:
-        rpc(
-            caller=partial(rpc_session.post, url=rpc_uri),
-            secret=rpc_secret,
-            method="aria2.shutdown"
-        )
 
-
+    return p.returncode
-def aria2c(
-    urls: Union[str, list[str], dict[str, Any], list[dict[str, Any]]],
-    output_dir: Path,
-    filename: str,
-    headers: Optional[MutableMapping[str, Union[str, bytes]]] = None,
-    cookies: Optional[Union[MutableMapping[str, str], CookieJar]] = None,
-    proxy: Optional[str] = None,
-    max_workers: Optional[int] = None
-) -> Generator[dict[str, Any], None, None]:
-    """
-    Download files using Aria2(c).
-    https://aria2.github.io
-
-    Yields the following download status updates while chunks are downloading:
-
-    - {total: 100} (100% download total)
-    - {completed: 1} (1% download progress out of 100%)
-    - {downloaded: "10.1 MB/s"} (currently downloading at a rate of 10.1 MB/s)
-
-    The data is in the same format accepted by rich's progress.update() function.
-
-    Parameters:
-        urls: Web URL(s) to file(s) to download. You can use a dictionary with the key
-            "url" for the URI, and other keys for extra arguments to use per-URL.
-        output_dir: The folder to save the file into. If the save path's directory does
-            not exist then it will be made automatically.
-        filename: The filename or filename template to use for each file. The variables
-            you can use are `i` for the URL index and `ext` for the URL extension.
-        headers: A mapping of HTTP Header Key/Values to use for all downloads.
-        cookies: A mapping of Cookie Key/Values or a Cookie Jar to use for all downloads.
-        proxy: An optional proxy URI to route connections through for all downloads.
-        max_workers: The maximum amount of threads to use for downloads. Defaults to
-            min(32,(cpu_count+4)). Use for the --max-concurrent-downloads option.
-    """
-    if proxy and not proxy.lower().startswith("http://"):
-        # Only HTTP proxies are supported by aria2(c)
-        proxy = urlparse(proxy)
-
-        port = get_free_port()
-        username, password = get_random_bytes(8).hex(), get_random_bytes(8).hex()
-        local_proxy = f"http://{username}:{password}@localhost:{port}"
-
-        scheme = {
-            "https": "http+ssl",
-            "socks5h": "socks"
-        }.get(proxy.scheme, proxy.scheme)
-
-        remote_server = f"{scheme}://{proxy.hostname}"
-        if proxy.port:
-            remote_server += f":{proxy.port}"
-        if proxy.username or proxy.password:
-            remote_server += "#"
-        if proxy.username:
-            remote_server += proxy.username
-        if proxy.password:
-            remote_server += f":{proxy.password}"
-
-        p = subprocess.Popen(
-            [
-                "pproxy",
-                "-l", f"http://:{port}#{username}:{password}",
-                "-r", remote_server
-            ],
-            stdout=subprocess.DEVNULL,
-            stderr=subprocess.DEVNULL
-        )
-
-        try:
-            yield from download(urls, output_dir, filename, headers, cookies, local_proxy, max_workers)
-        finally:
-            p.kill()
-            p.wait()
-        return
-    yield from download(urls, output_dir, filename, headers, cookies, proxy, max_workers)
 
 
 __all__ = ("aria2c",)

devine/core/downloaders/curl_impersonate.py

@@ -1,217 +1,49 @@
-import math
 import time
-from concurrent import futures
+from functools import partial
-from concurrent.futures.thread import ThreadPoolExecutor
-from http.cookiejar import CookieJar
 from pathlib import Path
-from typing import Any, Generator, MutableMapping, Optional, Union
+from typing import Any, MutableMapping, Optional, Union
 
 from curl_cffi.requests import Session
+from requests.cookies import RequestsCookieJar
 from rich import filesize
 
 from devine.core.config import config
 from devine.core.constants import DOWNLOAD_CANCELLED
-from devine.core.utilities import get_extension
 
 MAX_ATTEMPTS = 5
 RETRY_WAIT = 2
-CHUNK_SIZE = 1024
+BROWSER = config.curl_impersonate.get("browser", "chrome110")
-PROGRESS_WINDOW = 5
-BROWSER = config.curl_impersonate.get("browser", "chrome120")
-
-
-def download(
-    url: str,
-    save_path: Path,
-    session: Optional[Session] = None,
-    **kwargs: Any
-) -> Generator[dict[str, Any], None, None]:
-    """
-    Download files using Curl Impersonate.
-    https://github.com/lwthiker/curl-impersonate
-
-    Yields the following download status updates while chunks are downloading:
-
-    - {total: 123} (there are 123 chunks to download)
-    - {total: None} (there are an unknown number of chunks to download)
-    - {advance: 1} (one chunk was downloaded)
-    - {downloaded: "10.1 MB/s"} (currently downloading at a rate of 10.1 MB/s)
-    - {file_downloaded: Path(...), written: 1024} (download finished, has the save path and size)
-
-    The data is in the same format accepted by rich's progress.update() function. The
-    `downloaded` key is custom and is not natively accepted by all rich progress bars.
-
-    Parameters:
-        url: Web URL of a file to download.
-        save_path: The path to save the file to. If the save path's directory does not
-            exist then it will be made automatically.
-        session: The Requests or Curl-Impersonate Session to make HTTP requests with.
-            Useful to set Header, Cookie, and Proxy data. Connections are saved and
-            re-used with the session so long as the server keeps the connection alive.
-        kwargs: Any extra keyword arguments to pass to the session.get() call. Use this
-            for one-time request changes like a header, cookie, or proxy. For example,
-            to request Byte-ranges use e.g., `headers={"Range": "bytes=0-128"}`.
-    """
-    if not session:
-        session = Session(impersonate=BROWSER)
-
-    save_dir = save_path.parent
-    control_file = save_path.with_name(f"{save_path.name}.!dev")
-
-    save_dir.mkdir(parents=True, exist_ok=True)
-
-    if control_file.exists():
-        # consider the file corrupt if the control file exists
-        save_path.unlink(missing_ok=True)
-        control_file.unlink()
-    elif save_path.exists():
-        # if it exists, and no control file, then it should be safe
-        yield dict(
-            file_downloaded=save_path,
-            written=save_path.stat().st_size
-        )
-
-    # TODO: Design a control file format so we know how much of the file is missing
-    control_file.write_bytes(b"")
-
-    attempts = 1
-    try:
-        while True:
-            written = 0
-            download_sizes = []
-            last_speed_refresh = time.time()
-
-            try:
-                stream = session.get(url, stream=True, **kwargs)
-                stream.raise_for_status()
-
-                try:
-                    content_length = int(stream.headers.get("Content-Length", "0"))
-                except ValueError:
-                    content_length = 0
-
-                if content_length > 0:
-                    yield dict(total=math.ceil(content_length / CHUNK_SIZE))
-                else:
-                    # we have no data to calculate total chunks
-                    yield dict(total=None)  # indeterminate mode
-
-                with open(save_path, "wb") as f:
-                    for chunk in stream.iter_content(chunk_size=CHUNK_SIZE):
-                        download_size = len(chunk)
-                        f.write(chunk)
-                        written += download_size
-
-                        yield dict(advance=1)
-
-                        now = time.time()
-                        time_since = now - last_speed_refresh
-
-                        download_sizes.append(download_size)
-                        if time_since > PROGRESS_WINDOW or download_size < CHUNK_SIZE:
-                            data_size = sum(download_sizes)
-                            download_speed = math.ceil(data_size / (time_since or 1))
-                            yield dict(downloaded=f"{filesize.decimal(download_speed)}/s")
-                            last_speed_refresh = now
-                            download_sizes.clear()
-
-                yield dict(
-                    file_downloaded=save_path,
-                    written=written
-                )
-                break
-            except Exception as e:
-                save_path.unlink(missing_ok=True)
-                if DOWNLOAD_CANCELLED.is_set() or attempts == MAX_ATTEMPTS:
-                    raise e
-                time.sleep(RETRY_WAIT)
-                attempts += 1
-    finally:
-        control_file.unlink()
 
 
 def curl_impersonate(
-    urls: Union[str, list[str], dict[str, Any], list[dict[str, Any]]],
+    uri: Union[str, list[str]],
-    output_dir: Path,
+    out: Path,
-    filename: str,
+    headers: Optional[dict] = None,
-    headers: Optional[MutableMapping[str, Union[str, bytes]]] = None,
+    cookies: Optional[Union[MutableMapping[str, str], RequestsCookieJar]] = None,
-    cookies: Optional[Union[MutableMapping[str, str], CookieJar]] = None,
     proxy: Optional[str] = None,
-    max_workers: Optional[int] = None
+    progress: Optional[partial] = None,
-) -> Generator[dict[str, Any], None, None]:
+    *_: Any,
+    **__: Any
+) -> int:
     """
     Download files using Curl Impersonate.
     https://github.com/lwthiker/curl-impersonate
 
-    Yields the following download status updates while chunks are downloading:
+    If multiple URLs are provided they will be downloaded in the provided order
-
+    to the output directory. They will not be merged together.
-    - {total: 123} (there are 123 chunks to download)
-    - {total: None} (there are an unknown number of chunks to download)
-    - {advance: 1} (one chunk was downloaded)
-    - {downloaded: "10.1 MB/s"} (currently downloading at a rate of 10.1 MB/s)
-    - {file_downloaded: Path(...), written: 1024} (download finished, has the save path and size)
-
-    The data is in the same format accepted by rich's progress.update() function.
-    However, The `downloaded`, `file_downloaded` and `written` keys are custom and not
-    natively accepted by rich progress bars.
-
-    Parameters:
-        urls: Web URL(s) to file(s) to download. You can use a dictionary with the key
-            "url" for the URI, and other keys for extra arguments to use per-URL.
-        output_dir: The folder to save the file into. If the save path's directory does
-            not exist then it will be made automatically.
-        filename: The filename or filename template to use for each file. The variables
-            you can use are `i` for the URL index and `ext` for the URL extension.
-        headers: A mapping of HTTP Header Key/Values to use for all downloads.
-        cookies: A mapping of Cookie Key/Values or a Cookie Jar to use for all downloads.
-        proxy: An optional proxy URI to route connections through for all downloads.
-        max_workers: The maximum amount of threads to use for downloads. Defaults to
-            min(32,(cpu_count+4)).
     """
-    if not urls:
+    if isinstance(uri, list) and len(uri) == 1:
-        raise ValueError("urls must be provided and not empty")
+        uri = uri[0]
-    elif not isinstance(urls, (str, dict, list)):
-        raise TypeError(f"Expected urls to be {str} or {dict} or a list of one of them, not {type(urls)}")
 
-    if not output_dir:
+    if isinstance(uri, list):
-        raise ValueError("output_dir must be provided")
+        if out.is_file():
-    elif not isinstance(output_dir, Path):
+            raise ValueError("Expecting out to be a Directory path not a File as multiple URLs were provided")
-        raise TypeError(f"Expected output_dir to be {Path}, not {type(output_dir)}")
+        uri = [
-
+            (url, out / f"{i:08}.mp4")
-    if not filename:
+            for i, url in enumerate(uri)
-        raise ValueError("filename must be provided")
-    elif not isinstance(filename, str):
-        raise TypeError(f"Expected filename to be {str}, not {type(filename)}")
-
-    if not isinstance(headers, (MutableMapping, type(None))):
-        raise TypeError(f"Expected headers to be {MutableMapping}, not {type(headers)}")
-
-    if not isinstance(cookies, (MutableMapping, CookieJar, type(None))):
-        raise TypeError(f"Expected cookies to be {MutableMapping} or {CookieJar}, not {type(cookies)}")
-
-    if not isinstance(proxy, (str, type(None))):
-        raise TypeError(f"Expected proxy to be {str}, not {type(proxy)}")
-
-    if not isinstance(max_workers, (int, type(None))):
-        raise TypeError(f"Expected max_workers to be {int}, not {type(max_workers)}")
-
-    if not isinstance(urls, list):
-        urls = [urls]
-
-    urls = [
-        dict(
-            save_path=save_path,
-            **url
-        ) if isinstance(url, dict) else dict(
-            url=url,
-            save_path=save_path
-        )
-        for i, url in enumerate(urls)
-        for save_path in [output_dir / filename.format(
-            i=i,
-            ext=get_extension(url["url"] if isinstance(url, dict) else url)
-        )]
         ]
+    else:
+        uri = [(uri, out.parent / out.name)]
 
     session = Session(impersonate=BROWSER)
     if headers:
@@ -225,65 +57,49 @@ def curl_impersonate(
         session.cookies.update(cookies)
     if proxy:
         session.proxies.update({
-            "http": proxy.replace("https://", "http://"),
+            "http": proxy,
-            "https": proxy.replace("https://", "http://")
+            "https": proxy
         })
 
-    yield dict(total=len(urls))
+    if progress:
+        progress(total=len(uri))
 
     download_sizes = []
     last_speed_refresh = time.time()
 
-    with ThreadPoolExecutor(max_workers=max_workers) as pool:
+    for url, out_path in uri:
-        for i, future in enumerate(futures.as_completed((
+        out_path.parent.mkdir(parents=True, exist_ok=True)
-            pool.submit(
+        attempts = 1
-                download,
-                session=session,
-                **url
-            )
-            for url in urls
-        ))):
-            file_path, download_size = None, None
         try:
-                for status_update in future.result():
+            stream = session.get(url, stream=True)
-                    if status_update.get("file_downloaded") and status_update.get("written"):
+            stream.raise_for_status()
-                        file_path = status_update["file_downloaded"]
+            with open(out_path, "wb") as f:
-                        download_size = status_update["written"]
+                written = 0
-                    elif len(urls) == 1:
+                for chunk in stream.iter_content(chunk_size=1024):
-                        # these are per-chunk updates, only useful if it's one big file
+                    download_size = len(chunk)
-                        yield status_update
+                    f.write(chunk)
-            except KeyboardInterrupt:
+                    written += download_size
-                DOWNLOAD_CANCELLED.set()  # skip pending track downloads
+                    if progress:
-                yield dict(downloaded="[yellow]CANCELLING")
+                        progress(advance=1)
-                pool.shutdown(wait=True, cancel_futures=True)
-                yield dict(downloaded="[yellow]CANCELLED")
-                # tell dl that it was cancelled
-                # the pool is already shut down, so exiting loop is fine
-                raise
-            except Exception:
-                DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-                yield dict(downloaded="[red]FAILING")
-                pool.shutdown(wait=True, cancel_futures=True)
-                yield dict(downloaded="[red]FAILED")
-                # tell dl that it failed
-                # the pool is already shut down, so exiting loop is fine
-                raise
-            else:
-                yield dict(file_downloaded=file_path)
-                yield dict(advance=1)
 
                         now = time.time()
                         time_since = now - last_speed_refresh
 
-                if download_size:  # no size == skipped dl
                         download_sizes.append(download_size)
-
+                        if time_since > 5 or download_size < 1024:
-                if download_sizes and (time_since > PROGRESS_WINDOW or i == len(urls)):
                             data_size = sum(download_sizes)
-                    download_speed = math.ceil(data_size / (time_since or 1))
+                            download_speed = data_size / (time_since or 1)
-                    yield dict(downloaded=f"{filesize.decimal(download_speed)}/s")
+                            progress(downloaded=f"{filesize.decimal(download_speed)}/s")
                             last_speed_refresh = now
                             download_sizes.clear()
+            break
+        except Exception as e:
+            if DOWNLOAD_CANCELLED.is_set() or attempts == MAX_ATTEMPTS:
+                raise e
+            time.sleep(RETRY_WAIT)
+            attempts += 1
+
+    return 0
 
 
 __all__ = ("curl_impersonate",)

devine/core/downloaders/requests.py

@@ -1,228 +1,50 @@
 import math
-import os
 import time
-from concurrent import futures
+from functools import partial
-from concurrent.futures.thread import ThreadPoolExecutor
-from http.cookiejar import CookieJar
 from pathlib import Path
-from typing import Any, Generator, MutableMapping, Optional, Union
+from typing import Any, MutableMapping, Optional, Union
 
 from requests import Session
-from requests.adapters import HTTPAdapter
+from requests.cookies import RequestsCookieJar
 from rich import filesize
 
 from devine.core.constants import DOWNLOAD_CANCELLED
-from devine.core.utilities import get_extension
 
 MAX_ATTEMPTS = 5
 RETRY_WAIT = 2
-CHUNK_SIZE = 1024
-PROGRESS_WINDOW = 5
-
-
-def download(
-    url: str,
-    save_path: Path,
-    session: Optional[Session] = None,
-    **kwargs: Any
-) -> Generator[dict[str, Any], None, None]:
-    """
-    Download a file using Python Requests.
-    https://requests.readthedocs.io
-
-    Yields the following download status updates while chunks are downloading:
-
-    - {total: 123} (there are 123 chunks to download)
-    - {total: None} (there are an unknown number of chunks to download)
-    - {advance: 1} (one chunk was downloaded)
-    - {downloaded: "10.1 MB/s"} (currently downloading at a rate of 10.1 MB/s)
-    - {file_downloaded: Path(...), written: 1024} (download finished, has the save path and size)
-
-    The data is in the same format accepted by rich's progress.update() function. The
-    `downloaded` key is custom and is not natively accepted by all rich progress bars.
-
-    Parameters:
-        url: Web URL of a file to download.
-        save_path: The path to save the file to. If the save path's directory does not
-            exist then it will be made automatically.
-        session: The Requests Session to make HTTP requests with. Useful to set Header,
-            Cookie, and Proxy data. Connections are saved and re-used with the session
-            so long as the server keeps the connection alive.
-        kwargs: Any extra keyword arguments to pass to the session.get() call. Use this
-            for one-time request changes like a header, cookie, or proxy. For example,
-            to request Byte-ranges use e.g., `headers={"Range": "bytes=0-128"}`.
-    """
-    session = session or Session()
-
-    save_dir = save_path.parent
-    control_file = save_path.with_name(f"{save_path.name}.!dev")
-
-    save_dir.mkdir(parents=True, exist_ok=True)
-
-    if control_file.exists():
-        # consider the file corrupt if the control file exists
-        save_path.unlink(missing_ok=True)
-        control_file.unlink()
-    elif save_path.exists():
-        # if it exists, and no control file, then it should be safe
-        yield dict(
-            file_downloaded=save_path,
-            written=save_path.stat().st_size
-        )
-
-    # TODO: Design a control file format so we know how much of the file is missing
-    control_file.write_bytes(b"")
-
-    attempts = 1
-    try:
-        while True:
-            written = 0
-            download_sizes = []
-            last_speed_refresh = time.time()
-
-            try:
-                stream = session.get(url, stream=True, **kwargs)
-                stream.raise_for_status()
-
-                try:
-                    content_length = int(stream.headers.get("Content-Length", "0"))
-                except ValueError:
-                    content_length = 0
-
-                if content_length > 0:
-                    yield dict(total=math.ceil(content_length / CHUNK_SIZE))
-                else:
-                    # we have no data to calculate total chunks
-                    yield dict(total=None)  # indeterminate mode
-
-                with open(save_path, "wb") as f:
-                    for chunk in stream.iter_content(chunk_size=CHUNK_SIZE):
-                        download_size = len(chunk)
-                        f.write(chunk)
-                        written += download_size
-
-                        yield dict(advance=1)
-
-                        now = time.time()
-                        time_since = now - last_speed_refresh
-
-                        download_sizes.append(download_size)
-                        if time_since > PROGRESS_WINDOW or download_size < CHUNK_SIZE:
-                            data_size = sum(download_sizes)
-                            download_speed = math.ceil(data_size / (time_since or 1))
-                            yield dict(downloaded=f"{filesize.decimal(download_speed)}/s")
-                            last_speed_refresh = now
-                            download_sizes.clear()
-
-                yield dict(
-                    file_downloaded=save_path,
-                    written=written
-                )
-                break
-            except Exception as e:
-                save_path.unlink(missing_ok=True)
-                if DOWNLOAD_CANCELLED.is_set() or attempts == MAX_ATTEMPTS:
-                    raise e
-                time.sleep(RETRY_WAIT)
-                attempts += 1
-    finally:
-        control_file.unlink()
 
 
 def requests(
-    urls: Union[str, list[str], dict[str, Any], list[dict[str, Any]]],
+    uri: Union[str, list[str]],
-    output_dir: Path,
+    out: Path,
-    filename: str,
+    headers: Optional[dict] = None,
-    headers: Optional[MutableMapping[str, Union[str, bytes]]] = None,
+    cookies: Optional[Union[MutableMapping[str, str], RequestsCookieJar]] = None,
-    cookies: Optional[Union[MutableMapping[str, str], CookieJar]] = None,
     proxy: Optional[str] = None,
-    max_workers: Optional[int] = None
+    progress: Optional[partial] = None,
-) -> Generator[dict[str, Any], None, None]:
+    *_: Any,
+    **__: Any
+) -> int:
     """
-    Download a file using Python Requests.
+    Download files using Python Requests.
     https://requests.readthedocs.io
 
-    Yields the following download status updates while chunks are downloading:
+    If multiple URLs are provided they will be downloaded in the provided order
-
+    to the output directory. They will not be merged together.
-    - {total: 123} (there are 123 chunks to download)
-    - {total: None} (there are an unknown number of chunks to download)
-    - {advance: 1} (one chunk was downloaded)
-    - {downloaded: "10.1 MB/s"} (currently downloading at a rate of 10.1 MB/s)
-    - {file_downloaded: Path(...), written: 1024} (download finished, has the save path and size)
-
-    The data is in the same format accepted by rich's progress.update() function.
-    However, The `downloaded`, `file_downloaded` and `written` keys are custom and not
-    natively accepted by rich progress bars.
-
-    Parameters:
-        urls: Web URL(s) to file(s) to download. You can use a dictionary with the key
-            "url" for the URI, and other keys for extra arguments to use per-URL.
-        output_dir: The folder to save the file into. If the save path's directory does
-            not exist then it will be made automatically.
-        filename: The filename or filename template to use for each file. The variables
-            you can use are `i` for the URL index and `ext` for the URL extension.
-        headers: A mapping of HTTP Header Key/Values to use for all downloads.
-        cookies: A mapping of Cookie Key/Values or a Cookie Jar to use for all downloads.
-        proxy: An optional proxy URI to route connections through for all downloads.
-        max_workers: The maximum amount of threads to use for downloads. Defaults to
-            min(32,(cpu_count+4)).
     """
-    if not urls:
+    if isinstance(uri, list) and len(uri) == 1:
-        raise ValueError("urls must be provided and not empty")
+        uri = uri[0]
-    elif not isinstance(urls, (str, dict, list)):
-        raise TypeError(f"Expected urls to be {str} or {dict} or a list of one of them, not {type(urls)}")
 
-    if not output_dir:
+    if isinstance(uri, list):
-        raise ValueError("output_dir must be provided")
+        if out.is_file():
-    elif not isinstance(output_dir, Path):
+            raise ValueError("Expecting out to be a Directory path not a File as multiple URLs were provided")
-        raise TypeError(f"Expected output_dir to be {Path}, not {type(output_dir)}")
+        uri = [
-
+            (url, out / f"{i:08}.mp4")
-    if not filename:
+            for i, url in enumerate(uri)
-        raise ValueError("filename must be provided")
-    elif not isinstance(filename, str):
-        raise TypeError(f"Expected filename to be {str}, not {type(filename)}")
-
-    if not isinstance(headers, (MutableMapping, type(None))):
-        raise TypeError(f"Expected headers to be {MutableMapping}, not {type(headers)}")
-
-    if not isinstance(cookies, (MutableMapping, CookieJar, type(None))):
-        raise TypeError(f"Expected cookies to be {MutableMapping} or {CookieJar}, not {type(cookies)}")
-
-    if not isinstance(proxy, (str, type(None))):
-        raise TypeError(f"Expected proxy to be {str}, not {type(proxy)}")
-
-    if not isinstance(max_workers, (int, type(None))):
-        raise TypeError(f"Expected max_workers to be {int}, not {type(max_workers)}")
-
-    if not isinstance(urls, list):
-        urls = [urls]
-
-    if not max_workers:
-        max_workers = min(32, (os.cpu_count() or 1) + 4)
-
-    urls = [
-        dict(
-            save_path=save_path,
-            **url
-        ) if isinstance(url, dict) else dict(
-            url=url,
-            save_path=save_path
-        )
-        for i, url in enumerate(urls)
-        for save_path in [output_dir / filename.format(
-            i=i,
-            ext=get_extension(url["url"] if isinstance(url, dict) else url)
-        )]
         ]
+    else:
+        uri = [(uri, out.parent / out.name)]
 
     session = Session()
-    session.mount("https://", HTTPAdapter(
-        pool_connections=max_workers,
-        pool_maxsize=max_workers,
-        pool_block=True
-    ))
-    session.mount("http://", session.adapters["https://"])
-
     if headers:
         headers = {
             k: v
@@ -235,61 +57,53 @@ def requests(
     if proxy:
         session.proxies.update({"all": proxy})
 
-    yield dict(total=len(urls))
+    if progress:
+        progress(total=len(uri))
 
     download_sizes = []
     last_speed_refresh = time.time()
 
-    with ThreadPoolExecutor(max_workers=max_workers) as pool:
+    for url, out_path in uri:
-        for i, future in enumerate(futures.as_completed((
+        out_path.parent.mkdir(parents=True, exist_ok=True)
-            pool.submit(
+        attempts = 1
-                download,
+
-                session=session,
+        while True:
-                **url
-            )
-            for url in urls
-        ))):
-            file_path, download_size = None, None
             try:
-                for status_update in future.result():
+                stream = session.get(url, stream=True)
-                    if status_update.get("file_downloaded") and status_update.get("written"):
+                stream.raise_for_status()
-                        file_path = status_update["file_downloaded"]
+
-                        download_size = status_update["written"]
+                if len(uri) == 1 and progress:
-                    elif len(urls) == 1:
+                    content_length = int(stream.headers.get("Content-Length", "0"))
-                        # these are per-chunk updates, only useful if it's one big file
+                    if content_length > 0:
-                        yield status_update
+                        progress(total=math.ceil(content_length / 1024))
-            except KeyboardInterrupt:
+
-                DOWNLOAD_CANCELLED.set()  # skip pending track downloads
+                with open(out_path, "wb") as f:
-                yield dict(downloaded="[yellow]CANCELLING")
+                    written = 0
-                pool.shutdown(wait=True, cancel_futures=True)
+                    for chunk in stream.iter_content(chunk_size=1024):
-                yield dict(downloaded="[yellow]CANCELLED")
+                        download_size = len(chunk)
-                # tell dl that it was cancelled
+                        f.write(chunk)
-                # the pool is already shut down, so exiting loop is fine
+                        written += download_size
-                raise
+                        if progress:
-            except Exception:
+                            progress(advance=1)
-                DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-                yield dict(downloaded="[red]FAILING")
-                pool.shutdown(wait=True, cancel_futures=True)
-                yield dict(downloaded="[red]FAILED")
-                # tell dl that it failed
-                # the pool is already shut down, so exiting loop is fine
-                raise
-            else:
-                yield dict(file_downloaded=file_path, written=download_size)
-                yield dict(advance=1)
 
                             now = time.time()
                             time_since = now - last_speed_refresh
 
-                if download_size:  # no size == skipped dl
                             download_sizes.append(download_size)
-
+                            if time_since > 5 or download_size < 1024:
-                if download_sizes and (time_since > PROGRESS_WINDOW or i == len(urls)):
                                 data_size = sum(download_sizes)
-                    download_speed = math.ceil(data_size / (time_since or 1))
+                                download_speed = data_size / (time_since or 1)
-                    yield dict(downloaded=f"{filesize.decimal(download_speed)}/s")
+                                progress(downloaded=f"{filesize.decimal(download_speed)}/s")
                                 last_speed_refresh = now
                                 download_sizes.clear()
+                break
+            except Exception as e:
+                if DOWNLOAD_CANCELLED.is_set() or attempts == MAX_ATTEMPTS:
+                    raise e
+                time.sleep(RETRY_WAIT)
+                attempts += 1
+
+    return 0
 
 
 __all__ = ("requests",)

devine/core/drm/clearkey.py

@@ -6,10 +6,10 @@ from pathlib import Path
 from typing import Optional, Union
 from urllib.parse import urljoin
 
+import requests
 from Cryptodome.Cipher import AES
 from Cryptodome.Util.Padding import pad, unpad
 from m3u8.model import Key
-from requests import Session
 
 
 class ClearKey:
@@ -58,33 +58,14 @@ class ClearKey:
         shutil.move(decrypted_path, path)
 
     @classmethod
-    def from_m3u_key(cls, m3u_key: Key, session: Optional[Session] = None) -> ClearKey:
+    def from_m3u_key(cls, m3u_key: Key, proxy: Optional[str] = None) -> ClearKey:
-        """
-        Load a ClearKey from an M3U(8) Playlist's EXT-X-KEY.
-
-        Parameters:
-            m3u_key: A Key object parsed from a m3u(8) playlist using
-                the `m3u8` library.
-            session: Optional session used to request external URIs with.
-                Useful to set headers, proxies, cookies, and so forth.
-        """
         if not isinstance(m3u_key, Key):
             raise ValueError(f"Provided M3U Key is in an unexpected type {m3u_key!r}")
-        if not isinstance(session, (Session, type(None))):
-            raise TypeError(f"Expected session to be a {Session}, not a {type(session)}")
-
         if not m3u_key.method.startswith("AES"):
             raise ValueError(f"Provided M3U Key is not an AES Clear Key, {m3u_key.method}")
         if not m3u_key.uri:
             raise ValueError("No URI in M3U Key, unable to get Key.")
 
-        if not session:
-            session = Session()
-
-        if not session.headers.get("User-Agent"):
-            # commonly needed default for HLS playlists
-            session.headers["User-Agent"] = "smartexoplayer/1.1.0 (Linux;Android 8.0.0) ExoPlayerLib/2.13.3"
-
         if m3u_key.uri.startswith("data:"):
             media_types, data = m3u_key.uri[5:].split(",")
             media_types = media_types.split(";")
@@ -93,7 +74,13 @@ class ClearKey:
             key = data
         else:
             url = urljoin(m3u_key.base_uri, m3u_key.uri)
-            res = session.get(url)
+            res = requests.get(
+                url=url,
+                headers={
+                    "User-Agent": "smartexoplayer/1.1.0 (Linux;Android 8.0.0) ExoPlayerLib/2.13.3"
+                },
+                proxies={"all": proxy} if proxy else None
+            )
             res.raise_for_status()
             if not res.content:
                 raise EOFError("Unexpected Empty Response by M3U Key URI.")

devine/core/drm/widevine.py

@@ -78,7 +78,7 @@ class Widevine:
         pssh_boxes: list[Container] = []
         tenc_boxes: list[Container] = []
 
-        if track.descriptor == track.Descriptor.HLS:
+        if track.descriptor == track.Descriptor.M3U:
             m3u_url = track.url
             master = m3u8.loads(session.get(m3u_url).text, uri=m3u_url)
             pssh_boxes.extend(
@@ -224,7 +224,7 @@ class Widevine:
             raise ValueError("Cannot decrypt a Track without any Content Keys...")
 
         platform = {"win32": "win", "darwin": "osx"}.get(sys.platform, sys.platform)
-        executable = get_binary_path("shaka-packager", "packager", f"packager-{platform}", f"packager-{platform}-x64")
+        executable = get_binary_path("shaka-packager", f"packager-{platform}", f"packager-{platform}-x64")
         if not executable:
             raise EnvironmentError("Shaka Packager executable not found but is required.")
         if not path or not path.exists():

devine/core/manifests/dash.py

@@ -6,22 +6,28 @@ import logging
 import math
 import re
 import sys
+import time
+from concurrent import futures
+from concurrent.futures import ThreadPoolExecutor
 from copy import copy
 from functools import partial
+from hashlib import md5
 from pathlib import Path
-from typing import Any, Callable, Optional, Union
+from typing import Any, Callable, MutableMapping, Optional, Union
 from urllib.parse import urljoin, urlparse
 from uuid import UUID
-from zlib import crc32
 
 import requests
 from langcodes import Language, tag_is_valid
-from lxml.etree import Element, ElementTree
+from lxml.etree import Element
 from pywidevine.cdm import Cdm as WidevineCdm
 from pywidevine.pssh import PSSH
 from requests import Session
+from requests.cookies import RequestsCookieJar
+from rich import filesize
 
 from devine.core.constants import DOWNLOAD_CANCELLED, DOWNLOAD_LICENCE_ONLY, AnyTrack
+from devine.core.downloaders import downloader
 from devine.core.downloaders import requests as requests_downloader
 from devine.core.drm import Widevine
 from devine.core.tracks import Audio, Subtitle, Tracks, Video
@@ -113,7 +119,6 @@ class DASH:
                 for rep in adaptation_set.findall("Representation"):
                     get = partial(self._get, adaptation_set=adaptation_set, representation=rep)
                     findall = partial(self._findall, adaptation_set=adaptation_set, representation=rep, both=True)
-                    segment_base = rep.find("SegmentBase")
 
                     codecs = get("codecs")
                     content_type = get("contentType")
@@ -141,10 +146,6 @@ class DASH:
                     if content_type == "video":
                         track_type = Video
                         track_codec = Video.Codec.from_codecs(codecs)
-                        track_fps = get("frameRate")
-                        if not track_fps and segment_base is not None:
-                            track_fps = segment_base.get("timescale")
-
                         track_args = dict(
                             range_=self.get_video_range(
                                 codecs,
@@ -154,7 +155,7 @@ class DASH:
                             bitrate=get("bandwidth") or None,
                             width=get("width") or 0,
                             height=get("height") or 0,
-                            fps=track_fps or None
+                            fps=get("frameRate") or (rep.find("SegmentBase") or {}).get("timescale") or None
                         )
                     elif content_type == "audio":
                         track_type = Audio
@@ -172,9 +173,8 @@ class DASH:
                         track_type = Subtitle
                         track_codec = Subtitle.Codec.from_codecs(codecs or "vtt")
                         track_args = dict(
-                            cc=self.is_closed_caption(adaptation_set),
+                            forced=self.is_forced(adaptation_set),
-                            sdh=self.is_sdh(adaptation_set),
+                            cc=self.is_closed_caption(adaptation_set)
-                            forced=self.is_forced(adaptation_set)
                         )
                     elif content_type == "image":
                         # we don't want what's likely thumbnails for the seekbar
@@ -195,30 +195,23 @@ class DASH:
                     # a good and actually unique track ID, sometimes because of the lang
                     # dialect not being represented in the id, or the bitrate, or such.
                     # this combines all of them as one and hashes it to keep it small(ish).
-                    track_id = hex(crc32("{codec}-{lang}-{bitrate}-{base_url}-{ids}-{track_args}".format(
+                    track_id = md5("{codec}-{lang}-{bitrate}-{base_url}-{ids}-{track_args}".format(
                         codec=codecs,
                         lang=track_lang,
                         bitrate=get("bitrate"),
                         base_url=(rep.findtext("BaseURL") or "").split("?")[0],
                         ids=[get("audioTrackId"), get("id"), period.get("id")],
                         track_args=track_args
-                    ).encode()))[2:]
+                    ).encode()).hexdigest()
 
                     tracks.add(track_type(
                         id_=track_id,
-                        url=self.url,
+                        url=(self.url, self.manifest, rep, adaptation_set, period),
                         codec=track_codec,
                         language=track_lang,
                         is_original_lang=language and is_close_match(track_lang, [language]),
-                        descriptor=Video.Descriptor.DASH,
+                        descriptor=Video.Descriptor.MPD,
-                        data={
+                        extra=(rep, adaptation_set),
-                            "dash": {
-                                "manifest": self.manifest,
-                                "period": period,
-                                "adaptation_set": adaptation_set,
-                                "representation": rep
-                            }
-                        },
                         **track_args
                     ))
 
@@ -249,21 +242,18 @@ class DASH:
 
         log = logging.getLogger("DASH")
 
-        manifest: ElementTree = track.data["dash"]["manifest"]
+        manifest_url, manifest, representation, adaptation_set, period = track.url
-        period: Element = track.data["dash"]["period"]
-        adaptation_set: Element = track.data["dash"]["adaptation_set"]
-        representation: Element = track.data["dash"]["representation"]
 
         track.drm = DASH.get_drm(
             representation.findall("ContentProtection") +
             adaptation_set.findall("ContentProtection")
         )
 
+        manifest_url_query = urlparse(manifest_url).query
+
         manifest_base_url = manifest.findtext("BaseURL")
-        if not manifest_base_url:
+        if not manifest_base_url or not re.match("^https?://", manifest_base_url, re.IGNORECASE):
-            manifest_base_url = track.url
+            manifest_base_url = urljoin(manifest_url, "./", manifest_base_url)
-        elif not re.match("^https?://", manifest_base_url, re.IGNORECASE):
-            manifest_base_url = urljoin(track.url, f"./{manifest_base_url}")
         period_base_url = urljoin(manifest_base_url, period.findtext("BaseURL"))
         rep_base_url = urljoin(period_base_url, representation.findtext("BaseURL"))
 
@@ -278,10 +268,13 @@ class DASH:
         if segment_list is None:
             segment_list = adaptation_set.find("SegmentList")
 
-        segment_base = representation.find("SegmentBase")
+        if segment_template is None and segment_list is None and rep_base_url:
-        if segment_base is None:
+            # If there's no SegmentTemplate and no SegmentList, then SegmentBase is used or just BaseURL
-            segment_base = adaptation_set.find("SegmentBase")
+            # Regardless which of the two is used, we can just directly grab the BaseURL
-
+            # Players would normally calculate segments via Byte-Ranges, but we don't care
+            track.url = rep_base_url
+            track.descriptor = track.Descriptor.URL
+        else:
             segments: list[tuple[str, Optional[str]]] = []
             track_kid: Optional[UUID] = None
 
@@ -298,9 +291,7 @@ class DASH:
                         if not rep_base_url:
                             raise ValueError("Resolved Segment URL is not absolute, and no Base URL is available.")
                         value = urljoin(rep_base_url, value)
-                if not urlparse(value).query:
+                    if not urlparse(value).query and manifest_url_query:
-                    manifest_url_query = urlparse(track.url).query
-                    if manifest_url_query:
                         value += f"?{manifest_url_query}"
                     segment_template.set(item, value)
 
@@ -359,10 +350,8 @@ class DASH:
                 initialization = segment_list.find("Initialization")
                 if initialization is not None:
                     source_url = initialization.get("sourceURL")
-                if not source_url:
+                    if source_url is None:
                         source_url = rep_base_url
-                elif not re.match("^https?://", source_url, re.IGNORECASE):
-                    source_url = urljoin(rep_base_url, f"./{source_url}")
 
                     if initialization.get("range"):
                         init_range_header = {"Range": f"bytes={initialization.get('range')}"}
@@ -377,45 +366,16 @@ class DASH:
                 segment_urls = segment_list.findall("SegmentURL")
                 for segment_url in segment_urls:
                     media_url = segment_url.get("media")
-                if not media_url:
+                    if media_url is None:
                         media_url = rep_base_url
-                elif not re.match("^https?://", media_url, re.IGNORECASE):
-                    media_url = urljoin(rep_base_url, f"./{media_url}")
 
                     segments.append((
                         media_url,
                         segment_url.get("mediaRange")
                     ))
-        elif segment_base is not None:
-            media_range = None
-            init_data = None
-            initialization = segment_base.find("Initialization")
-            if initialization is not None:
-                if initialization.get("range"):
-                    init_range_header = {"Range": f"bytes={initialization.get('range')}"}
-                else:
-                    init_range_header = None
-
-                res = session.get(url=rep_base_url, headers=init_range_header)
-                res.raise_for_status()
-                init_data = res.content
-                track_kid = track.get_key_id(init_data)
-                total_size = res.headers.get("Content-Range", "").split("/")[-1]
-                if total_size:
-                    media_range = f"{len(init_data)}-{total_size}"
-
-            segments.append((
-                rep_base_url,
-                media_range
-            ))
-        elif rep_base_url:
-            segments.append((
-                rep_base_url,
-                None
-            ))
             else:
                 log.error("Could not find a way to get segments from this MPD manifest.")
-            log.debug(track.url)
+                log.debug(manifest_url)
                 sys.exit(1)
 
             if not track.drm and isinstance(track, (Video, Audio)):
@@ -451,52 +411,62 @@ class DASH:
 
             progress(total=len(segments))
 
-        downloader = track.downloader
+            download_sizes = []
-        if downloader.__name__ == "aria2c" and any(bytes_range is not None for url, bytes_range in segments):
+            download_speed_window = 5
-            # aria2(c) is shit and doesn't support the Range header, fallback to the requests downloader
+            last_speed_refresh = time.time()
-            downloader = requests_downloader
 
-        for status_update in downloader(
+            with ThreadPoolExecutor(max_workers=16) as pool:
-            urls=[
+                for i, download in enumerate(futures.as_completed((
-                {
+                    pool.submit(
-                    "url": url,
+                        DASH.download_segment,
-                    "headers": {
+                        url=url,
-                        "Range": f"bytes={bytes_range}"
+                        out_path=(save_dir / str(n).zfill(len(str(len(segments))))).with_suffix(".mp4"),
-                    } if bytes_range else {}
+                        track=track,
-                }
+                        proxy=proxy,
-                for url, bytes_range in segments
-            ],
-            output_dir=save_dir,
-            filename="{i:0%d}.mp4" % (len(str(len(segments)))),
                         headers=session.headers,
                         cookies=session.cookies,
-            proxy=proxy,
+                        bytes_range=bytes_range
-            max_workers=16
+                    )
-        ):
+                    for n, (url, bytes_range) in enumerate(segments)
-            file_downloaded = status_update.get("file_downloaded")
+                ))):
-            if file_downloaded and callable(track.OnSegmentDownloaded):
+                    try:
-                track.OnSegmentDownloaded(file_downloaded)
+                        download_size = download.result()
+                    except KeyboardInterrupt:
+                        DOWNLOAD_CANCELLED.set()  # skip pending track downloads
+                        progress(downloaded="[yellow]CANCELLING")
+                        pool.shutdown(wait=True, cancel_futures=True)
+                        progress(downloaded="[yellow]CANCELLED")
+                        # tell dl that it was cancelled
+                        # the pool is already shut down, so exiting loop is fine
+                        raise
+                    except Exception:
+                        DOWNLOAD_CANCELLED.set()  # skip pending track downloads
+                        progress(downloaded="[red]FAILING")
+                        pool.shutdown(wait=True, cancel_futures=True)
+                        progress(downloaded="[red]FAILED")
+                        # tell dl that it failed
+                        # the pool is already shut down, so exiting loop is fine
+                        raise
                     else:
-                downloaded = status_update.get("downloaded")
+                        progress(advance=1)
-                if downloaded and downloaded.endswith("/s"):
-                    status_update["downloaded"] = f"DASH {downloaded}"
-                progress(**status_update)
 
-        # see https://github.com/devine-dl/devine/issues/71
+                        now = time.time()
-        for control_file in save_dir.glob("*.aria2__temp"):
+                        time_since = now - last_speed_refresh
-            control_file.unlink()
+
+                        if download_size:  # no size == skipped dl
+                            download_sizes.append(download_size)
+
+                        if download_sizes and (time_since > download_speed_window or i == len(segments)):
+                            data_size = sum(download_sizes)
+                            download_speed = data_size / (time_since or 1)
+                            progress(downloaded=f"DASH {filesize.decimal(download_speed)}/s")
+                            last_speed_refresh = now
+                            download_sizes.clear()
 
-        segments_to_merge = [
-            x
-            for x in sorted(save_dir.iterdir())
-            if x.is_file()
-        ]
             with open(save_path, "wb") as f:
                 if init_data:
                     f.write(init_data)
-            if len(segments_to_merge) > 1:
+                for segment_file in sorted(save_dir.iterdir()):
-                progress(downloaded="Merging", completed=0, total=len(segments_to_merge))
-            for segment_file in segments_to_merge:
                     segment_data = segment_file.read_bytes()
                     # TODO: fix encoding after decryption?
                     if (
@@ -504,31 +474,86 @@ class DASH:
                         track.codec not in (Subtitle.Codec.fVTT, Subtitle.Codec.fTTML)
                     ):
                         segment_data = try_ensure_utf8(segment_data)
-                    segment_data = segment_data.decode("utf8"). \
+                        segment_data = html.unescape(segment_data.decode("utf8")).encode("utf8")
-                        replace("‎", html.unescape("‎")). \
-                        replace("‏", html.unescape("‏")). \
-                        encode("utf8")
                     f.write(segment_data)
-                f.flush()
                     segment_file.unlink()
-                progress(advance=1)
-
-        track.path = save_path
-        if callable(track.OnDownloaded):
-            track.OnDownloaded()
 
             if drm:
                 progress(downloaded="Decrypting", completed=0, total=100)
                 drm.decrypt(save_path)
                 track.drm = None
                 if callable(track.OnDecrypted):
-                track.OnDecrypted(drm)
+                    track.OnDecrypted(track)
-            progress(downloaded="Decrypting", advance=100)
+                progress(downloaded="Decrypted", completed=100)
 
+            track.path = save_path
             save_dir.rmdir()
 
             progress(downloaded="Downloaded")
 
+    @staticmethod
+    def download_segment(
+        url: str,
+        out_path: Path,
+        track: AnyTrack,
+        proxy: Optional[str] = None,
+        headers: Optional[MutableMapping[str, str | bytes]] = None,
+        cookies: Optional[Union[MutableMapping[str, str], RequestsCookieJar]] = None,
+        bytes_range: Optional[str] = None
+    ) -> int:
+        """
+        Download a DASH Media Segment.
+
+        Parameters:
+            url: Full HTTP(S) URL to the Segment you want to download.
+            out_path: Path to save the downloaded Segment file to.
+            track: The Track object of which this Segment is for. Currently only used to
+                fix an invalid value in the TFHD box of Audio Tracks.
+            proxy: Proxy URI to use when downloading the Segment file.
+            headers: HTTP Headers to send when requesting the Segment file.
+            cookies: Cookies to send when requesting the Segment file. The actual cookies sent
+                will be resolved based on the URI among other parameters. Multiple cookies with
+                the same name but a different domain/path are resolved.
+            bytes_range: Download only specific bytes of the Segment file using the Range header.
+
+        Returns the file size of the downloaded Segment in bytes.
+        """
+        if DOWNLOAD_CANCELLED.is_set():
+            raise KeyboardInterrupt()
+
+        if bytes_range:
+            # aria2(c) doesn't support byte ranges, use python-requests
+            downloader_ = requests_downloader
+            headers_ = dict(**headers, Range=f"bytes={bytes_range}")
+        else:
+            downloader_ = downloader
+            headers_ = headers
+
+        downloader_(
+            uri=url,
+            out=out_path,
+            headers=headers_,
+            cookies=cookies,
+            proxy=proxy,
+            segmented=True
+        )
+
+        # fix audio decryption on ATVP by fixing the sample description index
+        # TODO: Should this be done in the video data or the init data?
+        if isinstance(track, Audio):
+            with open(out_path, "rb+") as f:
+                segment_data = f.read()
+                fixed_segment_data = re.sub(
+                    b"(tfhd\x00\x02\x00\x1a\x00\x00\x00\x01\x00\x00\x00)\x02",
+                    b"\\g<1>\x01",
+                    segment_data
+                )
+                if fixed_segment_data != segment_data:
+                    f.seek(0)
+                    f.write(fixed_segment_data)
+
+        return out_path.stat().st_size
+
     @staticmethod
     def _get(
         item: str,
@@ -664,14 +689,6 @@ class DASH:
             for x in adaptation_set.findall("Role")
         )
 
-    @staticmethod
-    def is_sdh(adaptation_set: Element) -> bool:
-        """Check if contents of Adaptation Set is for the Hearing Impaired."""
-        return any(
-            (x.get("schemeIdUri"), x.get("value")) == ("urn:tva:metadata:cs:AudioPurposeCS:2007", "2")
-            for x in adaptation_set.findall("Accessibility")
-        )
-
     @staticmethod
     def is_closed_caption(adaptation_set: Element) -> bool:
         """Check if contents of Adaptation Set is a Closed Caption Subtitle."""

devine/core/manifests/hls.py

@@ -2,14 +2,18 @@ from __future__ import annotations
 
 import html
 import logging
-import shutil
+import re
-import subprocess
 import sys
+import time
+from concurrent import futures
+from concurrent.futures import ThreadPoolExecutor
 from functools import partial
+from hashlib import md5
 from pathlib import Path
+from queue import Queue
+from threading import Lock
 from typing import Any, Callable, Optional, Union
 from urllib.parse import urljoin
-from zlib import crc32
 
 import m3u8
 import requests
@@ -18,12 +22,14 @@ from m3u8 import M3U8
 from pywidevine.cdm import Cdm as WidevineCdm
 from pywidevine.pssh import PSSH
 from requests import Session
+from rich import filesize
 
 from devine.core.constants import DOWNLOAD_CANCELLED, DOWNLOAD_LICENCE_ONLY, AnyTrack
+from devine.core.downloaders import downloader
 from devine.core.downloaders import requests as requests_downloader
 from devine.core.drm import DRM_T, ClearKey, Widevine
 from devine.core.tracks import Audio, Subtitle, Tracks, Video
-from devine.core.utilities import get_binary_path, get_extension, is_close_match, try_ensure_utf8
+from devine.core.utilities import is_close_match, try_ensure_utf8
 
 
 class HLS:
@@ -87,7 +93,7 @@ class HLS:
         All Track objects' URL will be to another M3U(8) document. However, these documents
         will be Invariant Playlists and contain the list of segments URIs among other metadata.
         """
-        session_drm = HLS.get_all_drm(self.manifest.session_keys)
+        session_drm = HLS.get_drm(self.manifest.session_keys)
 
         audio_codecs_by_group_id: dict[str, Audio.Codec] = {}
         tracks = Tracks()
@@ -107,19 +113,15 @@ class HLS:
                 primary_track_type = Video
 
             tracks.add(primary_track_type(
-                id_=hex(crc32(str(playlist).encode()))[2:],
+                id_=md5(str(playlist).encode()).hexdigest()[0:7],  # 7 chars only for filename length
                 url=urljoin(playlist.base_uri, playlist.uri),
                 codec=primary_track_type.Codec.from_codecs(playlist.stream_info.codecs),
                 language=language,  # HLS manifests do not seem to have language info
                 is_original_lang=True,  # TODO: All we can do is assume Yes
                 bitrate=playlist.stream_info.average_bandwidth or playlist.stream_info.bandwidth,
-                descriptor=Video.Descriptor.HLS,
+                descriptor=Video.Descriptor.M3U,
                 drm=session_drm,
-                data={
+                extra=playlist,
-                    "hls": {
-                        "playlist": playlist
-                    }
-                },
                 # video track args
                 **(dict(
                     range_=Video.Range.DV if any(
@@ -162,18 +164,14 @@ class HLS:
                 raise ValueError(msg)
 
             tracks.add(track_type(
-                id_=hex(crc32(str(media).encode()))[2:],
+                id_=md5(str(media).encode()).hexdigest()[0:6],  # 6 chars only for filename length
                 url=urljoin(media.base_uri, media.uri),
                 codec=codec,
                 language=track_lang,  # HLS media may not have language info, fallback if needed
                 is_original_lang=language and is_close_match(track_lang, [language]),
-                descriptor=Audio.Descriptor.HLS,
+                descriptor=Audio.Descriptor.M3U,
                 drm=session_drm if media.type == "AUDIO" else None,
-                data={
+                extra=media,
-                    "hls": {
-                        "media": media
-                    }
-                },
                 # audio track args
                 **(dict(
                     bitrate=0,  # TODO: M3U doesn't seem to state bitrate?
@@ -238,442 +236,295 @@ class HLS:
         else:
             session_drm = None
 
-        unwanted_segments = [
+        progress(total=len(master.segments))
-            segment for segment in master.segments
-            if callable(track.OnSegmentFilter) and track.OnSegmentFilter(segment)
-        ]
 
-        total_segments = len(master.segments) - len(unwanted_segments)
+        download_sizes = []
-        progress(total=total_segments)
+        download_speed_window = 5
+        last_speed_refresh = time.time()
 
-        downloader = track.downloader
+        segment_key = Queue(maxsize=1)
+        segment_key.put((session_drm, None))
+        init_data = Queue(maxsize=1)
+        init_data.put(None)
+        range_offset = Queue(maxsize=1)
+        range_offset.put(0)
+        drm_lock = Lock()
 
-        urls: list[dict[str, Any]] = []
+        with ThreadPoolExecutor(max_workers=16) as pool:
-        range_offset = 0
+            for i, download in enumerate(futures.as_completed((
-        for segment in master.segments:
+                pool.submit(
-            if segment in unwanted_segments:
+                    HLS.download_segment,
-                continue
+                    segment=segment,
-
+                    out_path=(save_dir / str(n).zfill(len(str(len(master.segments))))).with_suffix(".mp4"),
-            if segment.byterange:
+                    track=track,
-                if downloader.__name__ == "aria2c":
+                    init_data=init_data,
-                    # aria2(c) is shit and doesn't support the Range header, fallback to the requests downloader
+                    segment_key=segment_key,
-                    downloader = requests_downloader
+                    range_offset=range_offset,
-                byte_range = HLS.calculate_byte_range(segment.byterange, range_offset)
+                    drm_lock=drm_lock,
-                range_offset = byte_range.split("-")[0]
+                    progress=progress,
-            else:
+                    license_widevine=license_widevine,
-                byte_range = None
+                    session=session,
-
+                    proxy=proxy
-            urls.append({
-                "url": urljoin(segment.base_uri, segment.uri),
-                "headers": {
-                    "Range": f"bytes={byte_range}"
-                } if byte_range else {}
-            })
-
-        segment_save_dir = save_dir / "segments"
-
-        for status_update in downloader(
-            urls=urls,
-            output_dir=segment_save_dir,
-            filename="{i:0%d}{ext}" % len(str(len(urls))),
-            headers=session.headers,
-            cookies=session.cookies,
-            proxy=proxy,
-            max_workers=16
-        ):
-            file_downloaded = status_update.get("file_downloaded")
-            if file_downloaded and callable(track.OnSegmentDownloaded):
-                track.OnSegmentDownloaded(file_downloaded)
-            else:
-                downloaded = status_update.get("downloaded")
-                if downloaded and downloaded.endswith("/s"):
-                    status_update["downloaded"] = f"HLS {downloaded}"
-                progress(**status_update)
-
-        # see https://github.com/devine-dl/devine/issues/71
-        for control_file in segment_save_dir.glob("*.aria2__temp"):
-            control_file.unlink()
-
-        progress(total=total_segments, completed=0, downloaded="Merging")
-
-        name_len = len(str(total_segments))
-        discon_i = 0
-        range_offset = 0
-        map_data: Optional[tuple[m3u8.model.InitializationSection, bytes]] = None
-        if session_drm:
-            encryption_data: Optional[tuple[Optional[m3u8.Key], DRM_T]] = (None, session_drm)
-        else:
-            encryption_data: Optional[tuple[Optional[m3u8.Key], DRM_T]] = None
-
-        i = -1
-        for real_i, segment in enumerate(master.segments):
-            if segment not in unwanted_segments:
-                i += 1
-
-            is_last_segment = (real_i + 1) == len(master.segments)
-
-            def merge(to: Path, via: list[Path], delete: bool = False, include_map_data: bool = False):
-                """
-                Merge all files to a given path, optionally including map data.
-
-                Parameters:
-                    to: The output file with all merged data.
-                    via: List of files to merge, in sequence.
-                    delete: Delete the file once it's been merged.
-                    include_map_data: Whether to include the init map data.
-                """
-                with open(to, "wb") as x:
-                    if include_map_data and map_data and map_data[1]:
-                        x.write(map_data[1])
-                    for file in via:
-                        x.write(file.read_bytes())
-                        x.flush()
-                        if delete:
-                            file.unlink()
-
-            def decrypt(include_this_segment: bool) -> Path:
-                """
-                Decrypt all segments that uses the currently set DRM.
-
-                All segments that will be decrypted with this DRM will be merged together
-                in sequence, prefixed with the init data (if any), and then deleted. Once
-                merged they will be decrypted. The merged and decrypted file names state
-                the range of segments that were used.
-
-                Parameters:
-                    include_this_segment: Whether to include the current segment in the
-                        list of segments to merge and decrypt. This should be False if
-                        decrypting on EXT-X-KEY changes, or True when decrypting on the
-                        last segment.
-
-                Returns the decrypted path.
-                """
-                drm = encryption_data[1]
-                first_segment_i = next(
-                    int(file.stem)
-                    for file in sorted(segment_save_dir.iterdir())
-                    if file.stem.isdigit()
                 )
-                last_segment_i = max(0, i - int(not include_this_segment))
+                for n, segment in enumerate(master.segments)
-                range_len = (last_segment_i - first_segment_i) + 1
+            ))):
-
-                segment_range = f"{str(first_segment_i).zfill(name_len)}-{str(last_segment_i).zfill(name_len)}"
-                merged_path = segment_save_dir / f"{segment_range}{get_extension(master.segments[last_segment_i].uri)}"
-                decrypted_path = segment_save_dir / f"{merged_path.stem}_decrypted{merged_path.suffix}"
-
-                files = [
-                    file
-                    for file in sorted(segment_save_dir.iterdir())
-                    if file.stem.isdigit() and first_segment_i <= int(file.stem) <= last_segment_i
-                ]
-                if not files:
-                    raise ValueError(f"None of the segment files for {segment_range} exist...")
-                elif len(files) != range_len:
-                    raise ValueError(f"Missing {range_len - len(files)} segment files for {segment_range}...")
-
-                merge(
-                    to=merged_path,
-                    via=files,
-                    delete=True,
-                    include_map_data=True
-                )
-
-                drm.decrypt(merged_path)
-                merged_path.rename(decrypted_path)
-
-                if callable(track.OnDecrypted):
-                    track.OnDecrypted(drm, decrypted_path)
-
-                return decrypted_path
-
-            def merge_discontinuity(include_this_segment: bool, include_map_data: bool = True):
-                """
-                Merge all segments of the discontinuity.
-
-                All segment files for this discontinuity must already be downloaded and
-                already decrypted (if it needs to be decrypted).
-
-                Parameters:
-                    include_this_segment: Whether to include the current segment in the
-                        list of segments to merge and decrypt. This should be False if
-                        decrypting on EXT-X-KEY changes, or True when decrypting on the
-                        last segment.
-                    include_map_data: Whether to prepend the init map data before the
-                        segment files when merging.
-                """
-                last_segment_i = max(0, i - int(not include_this_segment))
-
-                files = [
-                    file
-                    for file in sorted(segment_save_dir.iterdir())
-                    if int(file.stem.replace("_decrypted", "").split("-")[-1]) <= last_segment_i
-                ]
-                if files:
-                    to_dir = segment_save_dir.parent
-                    to_path = to_dir / f"{str(discon_i).zfill(name_len)}{files[-1].suffix}"
-                    merge(
-                        to=to_path,
-                        via=files,
-                        delete=True,
-                        include_map_data=include_map_data
-                    )
-
-            if segment not in unwanted_segments:
-                if isinstance(track, Subtitle):
-                    segment_file_ext = get_extension(segment.uri)
-                    segment_file_path = segment_save_dir / f"{str(i).zfill(name_len)}{segment_file_ext}"
-                    segment_data = try_ensure_utf8(segment_file_path.read_bytes())
-                    if track.codec not in (Subtitle.Codec.fVTT, Subtitle.Codec.fTTML):
-                        segment_data = segment_data.decode("utf8"). \
-                            replace("&lrm;", html.unescape("&lrm;")). \
-                            replace("&rlm;", html.unescape("&rlm;")). \
-                            encode("utf8")
-                    segment_file_path.write_bytes(segment_data)
-
-                if segment.discontinuity and i != 0:
-                    if encryption_data:
-                        decrypt(include_this_segment=False)
-                    merge_discontinuity(
-                        include_this_segment=False,
-                        include_map_data=not encryption_data or not encryption_data[1]
-                    )
-
-                    discon_i += 1
-                    range_offset = 0  # TODO: Should this be reset or not?
-                    map_data = None
-                    if encryption_data:
-                        encryption_data = (encryption_data[0], encryption_data[1])
-
-                if segment.init_section and (not map_data or segment.init_section != map_data[0]):
-                    if segment.init_section.byterange:
-                        init_byte_range = HLS.calculate_byte_range(
-                            segment.init_section.byterange,
-                            range_offset
-                        )
-                        range_offset = init_byte_range.split("-")[0]
-                        init_range_header = {
-                            "Range": f"bytes={init_byte_range}"
-                        }
-                    else:
-                        init_range_header = {}
-
-                    res = session.get(
-                        url=urljoin(segment.init_section.base_uri, segment.init_section.uri),
-                        headers=init_range_header
-                    )
-                    res.raise_for_status()
-                    map_data = (segment.init_section, res.content)
-
-            if segment.keys:
-                key = HLS.get_supported_key(segment.keys)
-                if encryption_data and encryption_data[0] != key and i != 0 and segment not in unwanted_segments:
-                    decrypt(include_this_segment=False)
-
-                if key is None:
-                    encryption_data = None
-                elif not encryption_data or encryption_data[0] != key:
-                    drm = HLS.get_drm(key, session)
-                    if isinstance(drm, Widevine):
                 try:
-                            if map_data:
+                    download_size = download.result()
-                                track_kid = track.get_key_id(map_data[1])
+                except KeyboardInterrupt:
-                            else:
-                                track_kid = None
-                            progress(downloaded="LICENSING")
-                            license_widevine(drm, track_kid=track_kid)
-                            progress(downloaded="[yellow]LICENSED")
-                        except Exception:  # noqa
                     DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-                            progress(downloaded="[red]FAILED")
+                    progress(downloaded="[yellow]CANCELLING")
+                    pool.shutdown(wait=True, cancel_futures=True)
+                    progress(downloaded="[yellow]CANCELLED")
+                    # tell dl that it was cancelled
+                    # the pool is already shut down, so exiting loop is fine
                     raise
-                    encryption_data = (key, drm)
+                except Exception as e:
-
+                    DOWNLOAD_CANCELLED.set()  # skip pending track downloads
-            # TODO: This wont work as we already downloaded
+                    progress(downloaded="[red]FAILING")
-            if DOWNLOAD_LICENCE_ONLY.is_set():
+                    pool.shutdown(wait=True, cancel_futures=True)
-                continue
+                    progress(downloaded="[red]FAILED")
-
+                    # tell dl that it failed
-            if is_last_segment:
+                    # the pool is already shut down, so exiting loop is fine
-                # required as it won't end with EXT-X-DISCONTINUITY nor a new key
+                    raise e
-                if encryption_data:
+                else:
-                    decrypt(include_this_segment=True)
+                    # it successfully downloaded, and it was not cancelled
-                merge_discontinuity(
-                    include_this_segment=True,
-                    include_map_data=not encryption_data or not encryption_data[1]
-                )
-
                     progress(advance=1)
 
-        # TODO: Again still wont work, we've already downloaded
+                    if download_size == -1:  # skipped for --skip-dl
+                        progress(downloaded="[yellow]SKIPPING")
+                        continue
+
+                    now = time.time()
+                    time_since = now - last_speed_refresh
+
+                    if download_size:  # no size == skipped dl
+                        download_sizes.append(download_size)
+
+                    if download_sizes and (time_since > download_speed_window or i == len(master.segments)):
+                        data_size = sum(download_sizes)
+                        download_speed = data_size / (time_since or 1)
+                        progress(downloaded=f"HLS {filesize.decimal(download_speed)}/s")
+                        last_speed_refresh = now
+                        download_sizes.clear()
+
         if DOWNLOAD_LICENCE_ONLY.is_set():
             return
 
-        segment_save_dir.rmdir()
-
-        # finally merge all the discontinuity save files together to the final path
-        segments_to_merge = [
-            x
-            for x in sorted(save_dir.iterdir())
-            if x.is_file()
-        ]
-        if len(segments_to_merge) == 1:
-            shutil.move(segments_to_merge[0], save_path)
-        else:
-            progress(downloaded="Merging")
-            if isinstance(track, (Video, Audio)):
-                HLS.merge_segments(
-                    segments=segments_to_merge,
-                    save_path=save_path
-                )
-            else:
         with open(save_path, "wb") as f:
-                    for discontinuity_file in segments_to_merge:
+            for segment_file in sorted(save_dir.iterdir()):
-                        discontinuity_data = discontinuity_file.read_bytes()
+                segment_data = segment_file.read_bytes()
-                        f.write(discontinuity_data)
+                if isinstance(track, Subtitle):
-                        f.flush()
+                    segment_data = try_ensure_utf8(segment_data)
-
+                    if track.codec not in (Subtitle.Codec.fVTT, Subtitle.Codec.fTTML):
-        save_dir.rmdir()
+                        segment_data = html.unescape(segment_data.decode("utf8")).encode("utf8")
+                f.write(segment_data)
+                segment_file.unlink()
 
         progress(downloaded="Downloaded")
 
         track.path = save_path
-        if callable(track.OnDownloaded):
+        save_dir.rmdir()
-            track.OnDownloaded()
 
     @staticmethod
-    def merge_segments(segments: list[Path], save_path: Path) -> int:
+    def download_segment(
+        segment: m3u8.Segment,
+        out_path: Path,
+        track: AnyTrack,
+        init_data: Queue,
+        segment_key: Queue,
+        range_offset: Queue,
+        drm_lock: Lock,
+        progress: partial,
+        license_widevine: Optional[Callable] = None,
+        session: Optional[Session] = None,
+        proxy: Optional[str] = None
+    ) -> int:
         """
-        Concatenate Segments by first demuxing with FFmpeg.
+        Download (and Decrypt) an HLS Media Segment.
 
-        Returns the file size of the merged file.
+        Note: Make sure all Queue objects passed are appropriately initialized with
+              a starting value or this function may get permanently stuck.
+
+        Parameters:
+            segment: The m3u8.Segment Object to Download.
+            out_path: Path to save the downloaded Segment file to.
+            track: The Track object of which this Segment is for. Currently used to fix an
+                invalid value in the TFHD box of Audio Tracks, for the OnSegmentFilter, and
+                for DRM-related operations like getting the Track ID and Decryption.
+            init_data: Queue for saving and loading the most recent init section data.
+            segment_key: Queue for saving and loading the most recent DRM object, and it's
+                adjacent Segment.Key object.
+            range_offset: Queue for saving and loading the most recent Segment Bytes Range.
+            drm_lock: Prevent more than one Download from doing anything DRM-related at the
+                same time. Make sure all calls to download_segment() use the same Lock object.
+            progress: Rich Progress bar to provide progress updates to.
+            license_widevine: Function used to license Widevine DRM objects. It must be passed
+                if the Segment's DRM uses Widevine.
+            proxy: Proxy URI to use when downloading the Segment file.
+            session: Python-Requests Session used when requesting init data.
+
+        Returns the file size of the downloaded Segment in bytes.
         """
-        ffmpeg = get_binary_path("ffmpeg")
+        if DOWNLOAD_CANCELLED.is_set():
-        if not ffmpeg:
+            raise KeyboardInterrupt()
-            raise EnvironmentError("FFmpeg executable was not found but is required to merge HLS segments.")
 
-        demuxer_file = segments[0].parent / "ffmpeg_concat_demuxer.txt"
+        if callable(track.OnSegmentFilter) and track.OnSegmentFilter(segment):
-        demuxer_file.write_text("\n".join([
+            return 0
-            f"file '{segment}'"
-            for segment in segments
-        ]))
 
-        subprocess.check_call([
+        # handle init section changes
-            ffmpeg, "-hide_banner",
+        newest_init_data = init_data.get()
-            "-loglevel", "panic",
+        try:
-            "-f", "concat",
+            if segment.init_section and (not newest_init_data or segment.discontinuity):
-            "-safe", "0",
+                # Only use the init data if there's no init data yet (e.g., start of file)
-            "-i", demuxer_file,
+                # or if EXT-X-DISCONTINUITY is reached at the same time as EXT-X-MAP.
-            "-map", "0",
+                # Even if a new EXT-X-MAP is supplied, it may just be duplicate and would
-            "-c", "copy",
+                # be unnecessary and slow to re-download the init data each time.
-            save_path
+                if segment.init_section.byterange:
-        ])
+                    previous_range_offset = range_offset.get()
-        demuxer_file.unlink()
+                    byte_range = HLS.calculate_byte_range(segment.init_section.byterange, previous_range_offset)
-
+                    range_offset.put(byte_range.split("-")[0])
-        return save_path.stat().st_size
+                    range_header = {
-
+                        "Range": f"bytes={byte_range}"
-    @staticmethod
+                    }
-    def get_supported_key(keys: list[Union[m3u8.model.SessionKey, m3u8.model.Key]]) -> Optional[m3u8.Key]:
-        """
-        Get a support Key System from a list of Key systems.
-
-        Note that the key systems are chosen in an opinionated order.
-
-        Returns None if one of the key systems is method=NONE, which means all segments
-        from hence forth should be treated as plain text until another key system is
-        encountered, unless it's also method=NONE.
-
-        Raises NotImplementedError if none of the key systems are supported.
-        """
-        if any(key.method == "NONE" for key in keys):
-            return None
-
-        unsupported_systems = []
-        for key in keys:
-            if not key:
-                continue
-            # TODO: Add a way to specify which supported key system to use
-            # TODO: Add support for 'SAMPLE-AES', 'AES-CTR', 'AES-CBC', 'ClearKey'
-            elif key.method == "AES-128":
-                return key
-            elif key.method == "ISO-23001-7":
-                return key
-            elif key.keyformat and key.keyformat.lower() == WidevineCdm.urn:
-                return key
                 else:
-                unsupported_systems.append(key.method + (f" ({key.keyformat})" if key.keyformat else ""))
+                    range_header = {}
+                res = session.get(
+                    url=urljoin(segment.init_section.base_uri, segment.init_section.uri),
+                    headers=range_header
+                )
+                res.raise_for_status()
+                newest_init_data = res.content
+        finally:
+            init_data.put(newest_init_data)
+
+        # handle segment key changes
+        with drm_lock:
+            newest_segment_key = segment_key.get()
+            try:
+                if segment.keys and newest_segment_key[1] != segment.keys:
+                    drm = HLS.get_drm(
+                        keys=segment.keys,
+                        proxy=proxy
+                    )
+                    if drm:
+                        track.drm = drm
+                        # license and grab content keys
+                        # TODO: What if we don't want to use the first DRM system?
+                        drm = drm[0]
+                        if isinstance(drm, Widevine):
+                            track_kid = track.get_key_id(newest_init_data)
+                            if not license_widevine:
+                                raise ValueError("license_widevine func must be supplied to use Widevine DRM")
+                            progress(downloaded="LICENSING")
+                            license_widevine(drm, track_kid=track_kid)
+                            progress(downloaded="[yellow]LICENSED")
+                        newest_segment_key = (drm, segment.keys)
+            finally:
+                segment_key.put(newest_segment_key)
+
+            if DOWNLOAD_LICENCE_ONLY.is_set():
+                return -1
+
+        headers_ = session.headers
+        if segment.byterange:
+            # aria2(c) doesn't support byte ranges, use python-requests
+            downloader_ = requests_downloader
+            previous_range_offset = range_offset.get()
+            byte_range = HLS.calculate_byte_range(segment.byterange, previous_range_offset)
+            range_offset.put(byte_range.split("-")[0])
+            headers_["Range"] = f"bytes={byte_range}"
         else:
-            raise NotImplementedError(f"None of the key systems are supported: {', '.join(unsupported_systems)}")
+            downloader_ = downloader
+
+        downloader_(
+            uri=urljoin(segment.base_uri, segment.uri),
+            out=out_path,
+            headers=headers_,
+            cookies=session.cookies,
+            proxy=proxy,
+            segmented=True
+        )
+
+        download_size = out_path.stat().st_size
+
+        # fix audio decryption on ATVP by fixing the sample description index
+        # TODO: Should this be done in the video data or the init data?
+        if isinstance(track, Audio):
+            with open(out_path, "rb+") as f:
+                segment_data = f.read()
+                fixed_segment_data = re.sub(
+                    b"(tfhd\x00\x02\x00\x1a\x00\x00\x00\x01\x00\x00\x00)\x02",
+                    b"\\g<1>\x01",
+                    segment_data
+                )
+                if fixed_segment_data != segment_data:
+                    f.seek(0)
+                    f.write(fixed_segment_data)
+
+        # prepend the init data to be able to decrypt
+        if newest_init_data:
+            with open(out_path, "rb+") as f:
+                segment_data = f.read()
+                f.seek(0)
+                f.write(newest_init_data)
+                f.write(segment_data)
+
+        # decrypt segment if encrypted
+        if newest_segment_key[0]:
+            newest_segment_key[0].decrypt(out_path)
+            track.drm = None
+            if callable(track.OnDecrypted):
+                track.OnDecrypted(track)
+
+        return download_size
 
     @staticmethod
     def get_drm(
-        key: Union[m3u8.model.SessionKey, m3u8.model.Key],
-        session: Optional[requests.Session] = None
-    ) -> DRM_T:
-        """
-        Convert HLS EXT-X-KEY data to an initialized DRM object.
-
-        Parameters:
-            key: m3u8 key system (EXT-X-KEY) object.
-            session: Optional session used to request AES-128 URIs.
-                Useful to set headers, proxies, cookies, and so forth.
-
-        Raises a NotImplementedError if the key system is not supported.
-        """
-        if not isinstance(session, (Session, type(None))):
-            raise TypeError(f"Expected session to be a {Session}, not {type(session)}")
-        if not session:
-            session = Session()
-
-        # TODO: Add support for 'SAMPLE-AES', 'AES-CTR', 'AES-CBC', 'ClearKey'
-        if key.method == "AES-128":
-            drm = ClearKey.from_m3u_key(key, session)
-        elif key.method == "ISO-23001-7":
-            drm = Widevine(
-                pssh=PSSH.new(
-                    key_ids=[key.uri.split(",")[-1]],
-                    system_id=PSSH.SystemId.Widevine
-                )
-            )
-        elif key.keyformat and key.keyformat.lower() == WidevineCdm.urn:
-            drm = Widevine(
-                pssh=PSSH(key.uri.split(",")[-1]),
-                **key._extra_params  # noqa
-            )
-        else:
-            raise NotImplementedError(f"The key system is not supported: {key}")
-
-        return drm
-
-    @staticmethod
-    def get_all_drm(
         keys: list[Union[m3u8.model.SessionKey, m3u8.model.Key]],
         proxy: Optional[str] = None
     ) -> list[DRM_T]:
         """
         Convert HLS EXT-X-KEY data to initialized DRM objects.
 
-        Parameters:
+        You can supply key data for a single segment or for the entire manifest.
-            keys: m3u8 key system (EXT-X-KEY) objects.
+        This lets you narrow the results down to each specific segment's DRM status.
-            proxy: Optional proxy string used for requesting AES-128 URIs.
 
-        Raises a NotImplementedError if none of the key systems are supported.
+        Returns an empty list if there were no supplied EXT-X-KEY data, or if all the
+        EXT-X-KEY's were of blank data. An empty list signals a DRM-free stream or segment.
+
+        Will raise a NotImplementedError if EXT-X-KEY data was supplied and none of them
+        were supported. A DRM-free track will never raise NotImplementedError.
         """
-        unsupported_keys: list[m3u8.Key] = []
+        drm = []
-        drm_objects: list[DRM_T] = []
+        unsupported_systems = []
-
-        if any(key.method == "NONE" for key in keys):
-            return []
 
         for key in keys:
-            try:
+            if not key:
-                drm = HLS.get_drm(key, proxy)
+                continue
-                drm_objects.append(drm)
+            # TODO: Add support for 'SAMPLE-AES', 'AES-CTR', 'AES-CBC', 'ClearKey'
-            except NotImplementedError:
+            if key.method == "NONE":
-                unsupported_keys.append(key)
+                return []
+            elif key.method == "AES-128":
+                drm.append(ClearKey.from_m3u_key(key, proxy))
+            elif key.method == "ISO-23001-7":
+                drm.append(Widevine(
+                    pssh=PSSH.new(
+                        key_ids=[key.uri.split(",")[-1]],
+                        system_id=PSSH.SystemId.Widevine
+                    )
+                ))
+            elif key.keyformat and key.keyformat.lower() == WidevineCdm.urn:
+                drm.append(Widevine(
+                    pssh=PSSH(key.uri.split(",")[-1]),
+                    **key._extra_params  # noqa
+                ))
+            else:
+                unsupported_systems.append(key.method + (f" ({key.keyformat})" if key.keyformat else ""))
 
-        if not drm_objects and unsupported_keys:
+        if not drm and unsupported_systems:
-            raise NotImplementedError(f"None of the key systems are supported: {unsupported_keys}")
+            raise NotImplementedError(f"No support for any of the key systems: {', '.join(unsupported_systems)}")
 
-        return drm_objects
+        return drm
 
     @staticmethod
     def calculate_byte_range(m3u_range: str, fallback_offset: int = 0) -> str:

devine/core/search_result.py

@@ -1,44 +0,0 @@
-from typing import Optional, Union
-
-
-class SearchResult:
-    def __init__(
-        self,
-        id_: Union[str, int],
-        title: str,
-        description: Optional[str] = None,
-        label: Optional[str] = None,
-        url: Optional[str] = None
-    ):
-        """
-        A Search Result for any support Title Type.
-
-        Parameters:
-            id_: The search result's Title ID.
-            title: The primary display text, e.g., the Title's Name.
-            description: The secondary display text, e.g., the Title's Description or
-                further title information.
-            label: The tertiary display text. This will typically be used to display
-                an informative label or tag to the result. E.g., "unavailable", the
-                title's price tag, region, etc.
-            url: A hyperlink to the search result or title's page.
-        """
-        if not isinstance(id_, (str, int)):
-            raise TypeError(f"Expected id_ to be a {str} or {int}, not {type(id_)}")
-        if not isinstance(title, str):
-            raise TypeError(f"Expected title to be a {str}, not {type(title)}")
-        if not isinstance(description, (str, type(None))):
-            raise TypeError(f"Expected description to be a {str}, not {type(description)}")
-        if not isinstance(label, (str, type(None))):
-            raise TypeError(f"Expected label to be a {str}, not {type(label)}")
-        if not isinstance(url, (str, type(None))):
-            raise TypeError(f"Expected url to be a {str}, not {type(url)}")
-
-        self.id = id_
-        self.title = title
-        self.description = description
-        self.label = label
-        self.url = url
-
-
-__all__ = ("SearchResult",)

devine/core/service.py

@@ -1,8 +1,7 @@
 import base64
 import logging
 from abc import ABCMeta, abstractmethod
-from collections.abc import Generator
+from http.cookiejar import CookieJar, MozillaCookieJar
-from http.cookiejar import CookieJar
 from typing import Optional, Union
 from urllib.parse import urlparse
 
@@ -17,9 +16,8 @@ from devine.core.config import config
 from devine.core.console import console
 from devine.core.constants import AnyTrack
 from devine.core.credential import Credential
-from devine.core.search_result import SearchResult
 from devine.core.titles import Title_T, Titles_T
-from devine.core.tracks import Chapters, Tracks
+from devine.core.tracks import Chapter, Tracks
 from devine.core.utilities import get_ip_info
 
 
@@ -98,12 +96,15 @@ class Service(metaclass=ABCMeta):
                 backoff_factor=0.2,
                 status_forcelist=[429, 500, 502, 503, 504]
             ),
+            # 16 connections is used for byte-ranged downloads
+            # double it to allow for 16 non-related connections
+            pool_maxsize=16 * 2,
             pool_block=True
         ))
         session.mount("http://", session.adapters["https://"])
         return session
 
-    def authenticate(self, cookies: Optional[CookieJar] = None, credential: Optional[Credential] = None) -> None:
+    def authenticate(self, cookies: Optional[MozillaCookieJar] = None, credential: Optional[Credential] = None) -> None:
         """
         Authenticate the Service with Cookies and/or Credentials (Email/Username and Password).
 
@@ -119,20 +120,9 @@ class Service(metaclass=ABCMeta):
         """
         if cookies is not None:
             if not isinstance(cookies, CookieJar):
-                raise TypeError(f"Expected cookies to be a {CookieJar}, not {cookies!r}.")
+                raise TypeError(f"Expected cookies to be a {MozillaCookieJar}, not {cookies!r}.")
             self.session.cookies.update(cookies)
 
-    def search(self) -> Generator[SearchResult, None, None]:
-        """
-        Search by query for titles from the Service.
-
-        The query must be taken as a CLI argument by the Service class.
-        Ideally just re-use the title ID argument (i.e. self.title).
-
-        Search results will be displayed in the order yielded.
-        """
-        raise NotImplementedError(f"Search functionality has not been implemented by {self.__class__.__name__}")
-
     def get_widevine_service_certificate(self, *, challenge: bytes, title: Title_T, track: AnyTrack) \
             -> Union[bytes, str]:
         """
@@ -217,22 +207,24 @@ class Service(metaclass=ABCMeta):
         """
 
     @abstractmethod
-    def get_chapters(self, title: Title_T) -> Chapters:
+    def get_chapters(self, title: Title_T) -> list[Chapter]:
         """
-        Get Chapters for the Title.
+        Get Chapter objects of the Title.
 
-        Parameters:
+        Return a list of Chapter objects. This will be run after get_tracks. If there's anything
-            title: The current Title from `get_titles` that is being processed.
+        from the get_tracks that may be needed, e.g. "device_id" or a-like, store it in the class
+        via `self` and re-use the value in get_chapters.
 
-        You must return a Chapters object containing 0 or more Chapter objects.
+        How it's used is generally the same as get_titles. These are only separated as to reduce
+        function complexity and keep them focused on simple tasks.
 
-        You do not need to set a Chapter number or sort/order the chapters in any way as
+        You do not need to sort or order the chapters in any way. However, you do need to filter
-        the Chapters class automatically handles all of that for you. If there's no
+        and alter them as needed by the service. No modification is made after get_chapters is
-        descriptive name for a Chapter then do not set a name at all.
+        ran. So that means ensure that the Chapter objects returned have consistent Chapter Titles
+        and Chapter Numbers.
 
-        You must not set Chapter names to "Chapter {n}" or such. If you (or the user)
+        :param title: The current `Title` from get_titles that is being executed.
-        wants "Chapter {n}" style Chapter names (or similar) then they can use the config
+        :return: List of Chapter objects, if available, empty list otherwise.
-        option `chapter_fallback_name`. For example, `"Chapter {i:02}"` for "Chapter 01".
         """
 
 

devine/core/tracks/__init__.py

@@ -1,9 +1,8 @@
 from .audio import Audio
 from .chapter import Chapter
-from .chapters import Chapters
 from .subtitle import Subtitle
 from .track import Track
 from .tracks import Tracks
 from .video import Video
 
-__all__ = ("Audio", "Chapter", "Chapters", "Subtitle", "Track", "Tracks", "Video")
+__all__ = ("Audio", "Chapter", "Subtitle", "Track", "Tracks", "Video")

devine/core/tracks/chapter.py

@@ -1,82 +1,95 @@
 from __future__ import annotations
 
 import re
+from pathlib import Path
 from typing import Optional, Union
-from zlib import crc32
-
-TIMESTAMP_FORMAT = re.compile(r"^(?P<hour>\d{2}):(?P<minute>\d{2}):(?P<second>\d{2})(?P<ms>.\d{3}|)$")
 
 
 class Chapter:
-    def __init__(self, timestamp: Union[str, int, float], name: Optional[str] = None):
+    line_1 = re.compile(r"^CHAPTER(?P<number>\d+)=(?P<timecode>[\d\\.]+)$")
-        """
+    line_2 = re.compile(r"^CHAPTER(?P<number>\d+)NAME=(?P<title>[\d\\.]+)$")
-        Create a new Chapter with a Timestamp and optional name.
 
-        The timestamp may be in the following formats:
+    def __init__(self, number: int, timecode: str, title: Optional[str] = None):
-        - "HH:MM:SS" string, e.g., `25:05:23`.
+        self.id = f"chapter-{number}"
-        - "HH:MM:SS.mss" string, e.g., `25:05:23.120`.
+        self.number = number
-        - a timecode integer in milliseconds, e.g., `90323120` is `25:05:23.120`.
+        self.timecode = timecode
-        - a timecode float in seconds, e.g., `90323.12` is `25:05:23.120`.
+        self.title = title
 
-        If you have a timecode integer in seconds, just multiply it by 1000.
+        if "." not in self.timecode:
-        If you have a timecode float in milliseconds (no decimal value), just convert
+            self.timecode += ".000"
-        it to an integer.
-        """
-        if timestamp is None:
-            raise ValueError("The timestamp must be provided.")
 
-        if not isinstance(timestamp, (str, int, float)):
+    def __bool__(self) -> bool:
-            raise TypeError(f"Expected timestamp to be {str}, {int} or {float}, not {type(timestamp)}")
+        return self.number and self.number >= 0 and self.timecode
-        if not isinstance(name, (str, type(None))):
-            raise TypeError(f"Expected name to be {str}, not {type(name)}")
-
-        if not isinstance(timestamp, str):
-            if isinstance(timestamp, int):  # ms
-                hours, remainder = divmod(timestamp, 1000 * 60 * 60)
-                minutes, remainder = divmod(remainder, 1000 * 60)
-                seconds, ms = divmod(remainder, 1000)
-            elif isinstance(timestamp, float):  # seconds.ms
-                hours, remainder = divmod(timestamp, 60 * 60)
-                minutes, remainder = divmod(remainder, 60)
-                seconds, ms = divmod(int(remainder * 1000), 1000)
-            else:
-                raise TypeError
-            timestamp = f"{hours:02}:{minutes:02}:{seconds:02}.{str(ms).zfill(3)[:3]}"
-
-        timestamp_m = TIMESTAMP_FORMAT.match(timestamp)
-        if not timestamp_m:
-            raise ValueError(f"The timestamp format is invalid: {timestamp}")
-
-        hour, minute, second, ms = timestamp_m.groups()
-        if not ms:
-            timestamp += ".000"
-
-        self.timestamp = timestamp
-        self.name = name
 
     def __repr__(self) -> str:
-        return "{name}({items})".format(
+        """
-            name=self.__class__.__name__,
+        OGM-based Simple Chapter Format intended for use with MKVToolNix.
-            items=", ".join([f"{k}={repr(v)}" for k, v in self.__dict__.items()])
+
+        This format is not officially part of the Matroska spec. This was a format
+        designed for OGM tools that MKVToolNix has since re-used. More Information:
+        https://mkvtoolnix.download/doc/mkvmerge.html#mkvmerge.chapters.simple
+        """
+        return "CHAPTER{num}={time}\nCHAPTER{num}NAME={name}".format(
+            num=f"{self.number:02}",
+            time=self.timecode,
+            name=self.title or ""
         )
 
     def __str__(self) -> str:
         return " | ".join(filter(bool, [
             "CHP",
-            self.timestamp,
+            f"[{self.number:02}]",
-            self.name
+            self.timecode,
+            self.title
         ]))
 
-    @property
-    def id(self) -> str:
-        """Compute an ID from the Chapter data."""
-        checksum = crc32(str(self).encode("utf8"))
-        return hex(checksum)
-
     @property
     def named(self) -> bool:
         """Check if Chapter is named."""
-        return bool(self.name)
+        return bool(self.title)
+
+    @classmethod
+    def loads(cls, data: str) -> Chapter:
+        """Load chapter data from a string."""
+        lines = [x.strip() for x in data.strip().splitlines(keepends=False)]
+        if len(lines) > 2:
+            return cls.loads("\n".join(lines))
+        one, two = lines
+
+        one_m = cls.line_1.match(one)
+        two_m = cls.line_2.match(two)
+        if not one_m or not two_m:
+            raise SyntaxError(f"An unexpected syntax error near:\n{one}\n{two}")
+
+        one_str, timecode = one_m.groups()
+        two_str, title = two_m.groups()
+        one_num, two_num = int(one_str.lstrip("0")), int(two_str.lstrip("0"))
+
+        if one_num != two_num:
+            raise SyntaxError(f"The chapter numbers ({one_num},{two_num}) does not match.")
+        if not timecode:
+            raise SyntaxError("The timecode is missing.")
+        if not title:
+            title = None
+
+        return cls(number=one_num, timecode=timecode, title=title)
+
+    @classmethod
+    def load(cls, path: Union[Path, str]) -> Chapter:
+        """Load chapter data from a file."""
+        if isinstance(path, str):
+            path = Path(path)
+        return cls.loads(path.read_text(encoding="utf8"))
+
+    def dumps(self) -> str:
+        """Return chapter data as a string."""
+        return repr(self)
+
+    def dump(self, path: Union[Path, str]) -> int:
+        """Write chapter data to a file."""
+        if isinstance(path, str):
+            path = Path(path)
+        return path.write_text(self.dumps(), encoding="utf8")
 
 
 __all__ = ("Chapter",)

devine/core/tracks/chapters.py

@@ -1,156 +0,0 @@
-from __future__ import annotations
-
-import re
-from abc import ABC
-from pathlib import Path
-from typing import Any, Iterable, Optional, Union
-from zlib import crc32
-
-from sortedcontainers import SortedKeyList
-
-from devine.core.tracks import Chapter
-
-OGM_SIMPLE_LINE_1_FORMAT = re.compile(r"^CHAPTER(?P<number>\d+)=(?P<timestamp>\d{2,}:\d{2}:\d{2}\.\d{3})$")
-OGM_SIMPLE_LINE_2_FORMAT = re.compile(r"^CHAPTER(?P<number>\d+)NAME=(?P<name>.*)$")
-
-
-class Chapters(SortedKeyList, ABC):
-    def __init__(self, iterable: Optional[Iterable[Chapter]] = None):
-        super().__init__(key=lambda x: x.timestamp or 0)
-        for chapter in iterable or []:
-            self.add(chapter)
-
-    def __repr__(self) -> str:
-        return "{name}({items})".format(
-            name=self.__class__.__name__,
-            items=", ".join([f"{k}={repr(v)}" for k, v in self.__dict__.items()])
-        )
-
-    def __str__(self) -> str:
-        return "\n".join([
-            " | ".join(filter(bool, [
-                "CHP",
-                f"[{i:02}]",
-                chapter.timestamp,
-                chapter.name
-            ]))
-            for i, chapter in enumerate(self, start=1)
-        ])
-
-    @classmethod
-    def loads(cls, data: str) -> Chapters:
-        """Load chapter data from a string."""
-        lines = [
-            line.strip()
-            for line in data.strip().splitlines(keepends=False)
-        ]
-
-        if len(lines) % 2 != 0:
-            raise ValueError("The number of chapter lines must be even.")
-
-        chapters = []
-
-        for line_1, line_2 in zip(lines[::2], lines[1::2]):
-            line_1_match = OGM_SIMPLE_LINE_1_FORMAT.match(line_1)
-            if not line_1_match:
-                raise SyntaxError(f"An unexpected syntax error occurred on: {line_1}")
-            line_2_match = OGM_SIMPLE_LINE_2_FORMAT.match(line_2)
-            if not line_2_match:
-                raise SyntaxError(f"An unexpected syntax error occurred on: {line_2}")
-
-            line_1_number, timestamp = line_1_match.groups()
-            line_2_number, name = line_2_match.groups()
-
-            if line_1_number != line_2_number:
-                raise SyntaxError(
-                    f"The chapter numbers {line_1_number} and {line_2_number} do not match on:\n{line_1}\n{line_2}")
-
-            if not timestamp:
-                raise SyntaxError(f"The timestamp is missing on: {line_1}")
-
-            chapters.append(Chapter(timestamp, name))
-
-        return cls(chapters)
-
-    @classmethod
-    def load(cls, path: Union[Path, str]) -> Chapters:
-        """Load chapter data from a file."""
-        if isinstance(path, str):
-            path = Path(path)
-        return cls.loads(path.read_text(encoding="utf8"))
-
-    def dumps(self, fallback_name: str = "") -> str:
-        """
-        Return chapter data in OGM-based Simple Chapter format.
-        https://mkvtoolnix.download/doc/mkvmerge.html#mkvmerge.chapters.simple
-
-        Parameters:
-            fallback_name: Name used for Chapters without a Name set.
-
-        The fallback name can use the following variables in f-string style:
-
-        - {i}: The Chapter number starting at 1.
-               E.g., `"Chapter {i}"`: "Chapter 1", "Intro", "Chapter 3".
-        - {j}: A number starting at 1 that increments any time a Chapter has no name.
-               E.g., `"Chapter {j}"`: "Chapter 1", "Intro", "Chapter 2".
-
-        These are formatted with f-strings, directives are supported.
-        For example, `"Chapter {i:02}"` will result in `"Chapter 01"`.
-        """
-        chapters = []
-        j = 0
-
-        for i, chapter in enumerate(self, start=1):
-            if not chapter.name:
-                j += 1
-            chapters.append("CHAPTER{num}={time}\nCHAPTER{num}NAME={name}".format(
-                num=f"{i:02}",
-                time=chapter.timestamp,
-                name=chapter.name or fallback_name.format(
-                    i=i,
-                    j=j
-                )
-            ))
-
-        return "\n".join(chapters)
-
-    def dump(self, path: Union[Path, str], *args: Any, **kwargs: Any) -> int:
-        """
-        Write chapter data in OGM-based Simple Chapter format to a file.
-
-        Parameters:
-            path: The file path to write the Chapter data to, overwriting
-                any existing data.
-
-        See `Chapters.dumps` for more parameter documentation.
-        """
-        if isinstance(path, str):
-            path = Path(path)
-        path.parent.mkdir(parents=True, exist_ok=True)
-
-        ogm_text = self.dumps(*args, **kwargs)
-        return path.write_text(ogm_text, encoding="utf8")
-
-    def add(self, value: Chapter) -> None:
-        if not isinstance(value, Chapter):
-            raise TypeError(f"Can only add {Chapter} objects, not {type(value)}")
-
-        if any(chapter.timestamp == value.timestamp for chapter in self):
-            raise ValueError(f"A Chapter with the Timestamp {value.timestamp} already exists")
-
-        super().add(value)
-
-        if not any(chapter.timestamp == "00:00:00.000" for chapter in self):
-            self.add(Chapter(0))
-
-    @property
-    def id(self) -> str:
-        """Compute an ID from the Chapter data."""
-        checksum = crc32("\n".join([
-            chapter.id
-            for chapter in self
-        ]).encode("utf8"))
-        return hex(checksum)
-
-
-__all__ = ("Chapters", "Chapter")

devine/core/tracks/subtitle.py

@@ -136,9 +136,6 @@ class Subtitle(Track):
         if (self.cc or self.sdh) and self.forced:
             raise ValueError("A text track cannot be CC/SDH as well as Forced.")
 
-        # Called after Track has been converted to another format
-        self.OnConverted: Optional[Callable[[Subtitle.Codec], None]] = None
-
     def get_track_name(self) -> Optional[str]:
         """Return the base Track Name."""
         track_name = super().get_track_name() or ""
@@ -149,12 +146,7 @@ class Subtitle(Track):
             track_name += flag
         return track_name or None
 
-    def download(
+    def download(self, session: requests.Session, prepare_drm: Callable, progress: partial) -> None:
-        self,
-        session: requests.Session,
-        prepare_drm: partial,
-        progress: Optional[partial] = None
-    ):
         super().download(session, prepare_drm, progress)
         if not self.path:
             return
@@ -201,16 +193,14 @@ class Subtitle(Track):
                 Subtitle.Codec.SubStationAlphav4: "AdvancedSubStationAlpha",
                 Subtitle.Codec.TimedTextMarkupLang: "TimedText1.0"
             }.get(codec, codec.name)
-            sub_edit_args = [
+            subprocess.run(
+                [
                     sub_edit_executable,
                     "/Convert", self.path, sub_edit_format,
                     f"/outputfilename:{output_path.name}",
+                    f"/outputfolder:{output_path.parent}",
                     "/encoding:utf8"
-            ]
+                ],
-            if codec == Subtitle.Codec.SubRip:
-                sub_edit_args.append("/ConvertColorsToDialog")
-            subprocess.run(
-                sub_edit_args,
                 check=True,
                 stdout=subprocess.DEVNULL,
                 stderr=subprocess.DEVNULL
@@ -231,12 +221,9 @@ class Subtitle(Track):
 
             output_path.write_text(subtitle_text, encoding="utf8")
 
-        self.path = output_path
+        self.swap(output_path)
         self.codec = codec
 
-        if callable(self.OnConverted):
-            self.OnConverted(codec)
-
         return output_path
 
     @staticmethod
@@ -524,6 +511,27 @@ class Subtitle(Track):
             stdout=subprocess.DEVNULL
         )
 
+    def remove_multi_lang_srt_header(self) -> None:
+        """
+        Remove Multi-Language SRT Header from Subtitle.
+
+        Sometimes a SubRip (SRT) format Subtitle has a "MULTI-LANGUAGE SRT" line,
+        when it shouldn't. This can cause Subtitle format/syntax errors in some
+        programs including mkvmerge/MKVToolNix.
+
+        This should only be used if it truly is a normal SubRip (SRT) subtitle
+        just with this line added by mistake.
+        """
+        if not self.path or not self.path.exists():
+            raise ValueError("You must download the subtitle track first.")
+
+        if self.codec != Subtitle.Codec.SubRip:
+            raise ValueError("Only SubRip (SRT) format Subtitles have the 'MULTI-LANGUAGE SRT' header.")
+
+        srt_text = self.path.read_text("utf8")
+        fixed_srt_text = srt_text.replace("MULTI-LANGUAGE SRT\n", "")
+        self.path.write_text(fixed_srt_text, "utf8")
+
     def __str__(self) -> str:
         return " | ".join(filter(bool, [
             "SUB",

devine/core/tracks/track.py

@@ -4,167 +4,237 @@ import logging
 import re
 import shutil
 import subprocess
-from copy import copy
 from enum import Enum
 from functools import partial
 from pathlib import Path
 from typing import Any, Callable, Iterable, Optional, Union
 from uuid import UUID
-from zlib import crc32
 
-import m3u8
+import requests
 from langcodes import Language
-from requests import Session
 
 from devine.core.config import config
-from devine.core.constants import DOWNLOAD_CANCELLED, DOWNLOAD_LICENCE_ONLY
+from devine.core.constants import DOWNLOAD_CANCELLED, DOWNLOAD_LICENCE_ONLY, TERRITORY_MAP
-from devine.core.downloaders import aria2c, curl_impersonate, requests
+from devine.core.downloaders import downloader
 from devine.core.drm import DRM_T, Widevine
 from devine.core.utilities import get_binary_path, get_boxes, try_ensure_utf8
 from devine.core.utils.subprocess import ffprobe
 
 
 class Track:
+    class DRM(Enum):
+        pass
+
     class Descriptor(Enum):
         URL = 1  # Direct URL, nothing fancy
-        HLS = 2  # https://en.wikipedia.org/wiki/HTTP_Live_Streaming
+        M3U = 2  # https://en.wikipedia.org/wiki/M3U (and M3U8)
-        DASH = 3  # https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP
+        MPD = 3  # https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP
 
     def __init__(
         self,
+        id_: str,
         url: Union[str, list[str]],
         language: Union[Language, str],
         is_original_lang: bool = False,
         descriptor: Descriptor = Descriptor.URL,
         needs_repack: bool = False,
-        name: Optional[str] = None,
         drm: Optional[Iterable[DRM_T]] = None,
         edition: Optional[str] = None,
-        downloader: Optional[Callable] = None,
+        extra: Optional[Any] = None
-        data: Optional[dict] = None,
-        id_: Optional[str] = None,
     ) -> None:
-        if not isinstance(url, (str, list)):
-            raise TypeError(f"Expected url to be a {str}, or list of {str}, not {type(url)}")
-        if not isinstance(language, (Language, str)):
-            raise TypeError(f"Expected language to be a {Language} or {str}, not {type(language)}")
-        if not isinstance(is_original_lang, bool):
-            raise TypeError(f"Expected is_original_lang to be a {bool}, not {type(is_original_lang)}")
-        if not isinstance(descriptor, Track.Descriptor):
-            raise TypeError(f"Expected descriptor to be a {Track.Descriptor}, not {type(descriptor)}")
-        if not isinstance(needs_repack, bool):
-            raise TypeError(f"Expected needs_repack to be a {bool}, not {type(needs_repack)}")
-        if not isinstance(name, (str, type(None))):
-            raise TypeError(f"Expected name to be a {str}, not {type(name)}")
-        if not isinstance(id_, (str, type(None))):
-            raise TypeError(f"Expected id_ to be a {str}, not {type(id_)}")
-        if not isinstance(edition, (str, type(None))):
-            raise TypeError(f"Expected edition to be a {str}, not {type(edition)}")
-        if not isinstance(downloader, (Callable, type(None))):
-            raise TypeError(f"Expected downloader to be a {Callable}, not {type(downloader)}")
-        if not isinstance(data, (dict, type(None))):
-            raise TypeError(f"Expected data to be a {dict}, not {type(data)}")
-
-        invalid_urls = ", ".join(set(type(x) for x in url if not isinstance(x, str)))
-        if invalid_urls:
-            raise TypeError(f"Expected all items in url to be a {str}, but found {invalid_urls}")
-
-        if drm is not None:
-            try:
-                iter(drm)
-            except TypeError:
-                raise TypeError(f"Expected drm to be an iterable, not {type(drm)}")
-
-        if downloader is None:
-            downloader = {
-                "aria2c": aria2c,
-                "curl_impersonate": curl_impersonate,
-                "requests": requests
-            }[config.downloader]
-
-        self.path: Optional[Path] = None
-        self.url = url
-        self.language = Language.get(language)
-        self.is_original_lang = is_original_lang
-        self.descriptor = descriptor
-        self.needs_repack = needs_repack
-        self.name = name
-        self.drm = drm
-        self.edition: str = edition
-        self.downloader = downloader
-        self.data = data or {}
-
-        if self.name is None:
-            lang = Language.get(self.language)
-            if (lang.language or "").lower() == (lang.territory or "").lower():
-                lang.territory = None  # e.g. en-en, de-DE
-            reduced = lang.simplify_script()
-            extra_parts = []
-            if reduced.script is not None:
-                script = reduced.script_name(max_distance=25)
-                if script and script != "Zzzz":
-                    extra_parts.append(script)
-            if reduced.territory is not None:
-                territory = reduced.territory_name(max_distance=25)
-                if territory and territory != "ZZ":
-                    territory = territory.removesuffix(" SAR China")
-                    extra_parts.append(territory)
-            self.name = ", ".join(extra_parts) or None
-
-        if not id_:
-            this = copy(self)
-            this.url = self.url.rsplit("?", maxsplit=1)[0]
-            checksum = crc32(repr(this).encode("utf8"))
-            id_ = hex(checksum)[2:]
-
         self.id = id_
+        self.url = url
+        # required basic metadata
+        self.language = Language.get(language)
+        self.is_original_lang = bool(is_original_lang)
+        # optional io metadata
+        self.descriptor = descriptor
+        self.needs_repack = bool(needs_repack)
+        # drm
+        self.drm = drm
+        # extra data
+        self.edition: str = edition
+        self.extra: Any = extra or {}  # allow anything for extra, but default to a dict
 
-        # TODO: Currently using OnFoo event naming, change to just segment_filter
+        # events
         self.OnSegmentFilter: Optional[Callable] = None
-
-        # Called after one of the Track's segments have downloaded
-        self.OnSegmentDownloaded: Optional[Callable[[Path], None]] = None
-        # Called after the Track has downloaded
         self.OnDownloaded: Optional[Callable] = None
-        # Called after the Track or one of its segments have been decrypted
+        self.OnDecrypted: Optional[Callable] = None
-        self.OnDecrypted: Optional[Callable[[DRM_T, Optional[m3u8.Segment]], None]] = None
-        # Called after the Track has been repackaged
         self.OnRepacked: Optional[Callable] = None
-        # Called before the Track is multiplexed
         self.OnMultiplex: Optional[Callable] = None
 
+        # should only be set internally
+        self.path: Optional[Path] = None
+
     def __repr__(self) -> str:
         return "{name}({items})".format(
             name=self.__class__.__name__,
             items=", ".join([f"{k}={repr(v)}" for k, v in self.__dict__.items()])
         )
 
-    def __eq__(self, other: Any) -> bool:
+    def __eq__(self, other: object) -> bool:
         return isinstance(other, Track) and self.id == other.id
 
-    def download(
+    def get_track_name(self) -> Optional[str]:
-        self,
+        """Return the base Track Name. This may be enhanced in sub-classes."""
-        session: Session,
+        if (self.language.language or "").lower() == (self.language.territory or "").lower():
-        prepare_drm: partial,
+            self.language.territory = None  # e.g. en-en, de-DE
-        progress: Optional[partial] = None
+        if self.language.territory == "US":
-    ):
+            self.language.territory = None
-        """Download and optionally Decrypt this Track."""
+        reduced = self.language.simplify_script()
-        from devine.core.manifests import DASH, HLS
+        extra_parts = []
+        if reduced.script is not None:
+            extra_parts.append(reduced.script_name(max_distance=25))
+        if reduced.territory is not None:
+            territory = reduced.territory_name(max_distance=25)
+            extra_parts.append(TERRITORY_MAP.get(territory, territory))
+        return ", ".join(extra_parts) or None
 
+    def get_key_id(self, init_data: Optional[bytes] = None, *args, **kwargs) -> Optional[UUID]:
+        """
+        Probe the DRM encryption Key ID (KID) for this specific track.
+
+        It currently supports finding the Key ID by probing the track's stream
+        with ffprobe for `enc_key_id` data, as well as for mp4 `tenc` (Track
+        Encryption) boxes.
+
+        It explicitly ignores PSSH information like the `PSSH` box, as the box
+        is likely to contain multiple Key IDs that may or may not be for this
+        specific track.
+
+        To retrieve the initialization segment, this method calls :meth:`get_init_segment`
+        with the positional and keyword arguments. The return value of `get_init_segment`
+        is then used to determine the Key ID.
+
+        Returns:
+            The Key ID as a UUID object, or None if the Key ID could not be determined.
+        """
+        if not init_data:
+            init_data = self.get_init_segment(*args, **kwargs)
+        if not isinstance(init_data, bytes):
+            raise TypeError(f"Expected init_data to be bytes, not {init_data!r}")
+
+        # try get via ffprobe, needed for non mp4 data e.g. WEBM from Google Play
+        probe = ffprobe(init_data)
+        if probe:
+            for stream in probe.get("streams") or []:
+                enc_key_id = stream.get("tags", {}).get("enc_key_id")
+                if enc_key_id:
+                    return UUID(bytes=base64.b64decode(enc_key_id))
+
+        # look for track encryption mp4 boxes
+        for tenc in get_boxes(init_data, b"tenc"):
+            if tenc.key_ID.int != 0:
+                return tenc.key_ID
+
+        # look for UUID mp4 boxes holding track encryption mp4 boxes
+        for uuid_box in get_boxes(init_data, b"uuid"):
+            if uuid_box.extended_type == UUID("8974dbce-7be7-4c51-84f9-7148f9882554"):
+                tenc = uuid_box.data
+                if tenc.key_ID.int != 0:
+                    return tenc.key_ID
+
+    def get_init_segment(
+        self,
+        maximum_size: int = 20000,
+        url: Optional[str] = None,
+        byte_range: Optional[str] = None,
+        session: Optional[requests.Session] = None
+    ) -> bytes:
+        """
+        Get the Track's Initial Segment Data Stream.
+
+        HLS and DASH tracks must explicitly provide a URL to the init segment or file.
+        Providing the byte-range for the init segment is recommended where possible.
+
+        If `byte_range` is not set, it will make a HEAD request and check the size of
+        the file. If the size could not be determined, it will download up to the first
+        20KB only, which should contain the entirety of the init segment. You may
+        override this by changing the `maximum_size`.
+
+        The default maximum_size of 20000 (20KB) is a tried-and-tested value that
+        seems to work well across the board.
+
+        Parameters:
+            maximum_size: Size to assume as the content length if byte-range is not
+                used, the content size could not be determined, or the content size
+                is larger than it. A value of 20000 (20KB) or higher is recommended.
+            url: Explicit init map or file URL to probe from.
+            byte_range: Range of bytes to download from the explicit or implicit URL.
+            session: Session context, e.g., authorization and headers.
+        """
+        if not session:
+            session = requests.Session()
+
+        if self.descriptor != self.Descriptor.URL and not url:
+            # We cannot know which init map from the HLS or DASH playlist is actually used.
+            # For DASH this could be from any adaptation set, any period, e.t.c.
+            # For HLS we could make some assumptions, but it's best that it is explicitly provided.
+            raise ValueError(
+                f"An explicit URL to an init map or file must be provided for {self.descriptor.name} tracks."
+            )
+
+        url = url or self.url
+        if not url:
+            raise ValueError("The track must have an URL to point towards it's data.")
+
+        content_length = maximum_size
+
+        if byte_range:
+            if not isinstance(byte_range, str):
+                raise TypeError(f"Expected byte_range to be a str, not {byte_range!r}")
+            if not re.match(r"^\d+-\d+$", byte_range):
+                raise ValueError(f"The value of byte_range is unrecognized: '{byte_range}'")
+            start, end = byte_range.split("-")
+            if start > end:
+                raise ValueError(f"The start range cannot be greater than the end range: {start}>{end}")
+        else:
+            size_test = session.head(url)
+            if "Content-Length" in size_test.headers:
+                content_length_header = int(size_test.headers["Content-Length"])
+                if content_length_header > 0:
+                    # use whichever is smaller in case this is a large file
+                    content_length = min(content_length_header, maximum_size)
+            range_test = session.head(url, headers={"Range": "bytes=0-1"})
+            if range_test.status_code == 206:
+                byte_range = f"0-{content_length-1}"
+
+        if byte_range:
+            res = session.get(
+                url=url,
+                headers={
+                    "Range": f"bytes={byte_range}"
+                }
+            )
+            res.raise_for_status()
+            init_data = res.content
+        else:
+            # Take advantage of streaming support to take just the first n bytes
+            # This is a hacky alternative to HTTP's Range on unsupported servers
+            init_data = None
+            with session.get(url, stream=True) as s:
+                for chunk in s.iter_content(content_length):
+                    init_data = chunk
+                    break
+            if not init_data:
+                raise ValueError(f"Failed to read {content_length} bytes from the track URI.")
+
+        return init_data
+
+    def download(self, session: requests.Session, prepare_drm: Callable, progress: partial) -> None:
         if DOWNLOAD_LICENCE_ONLY.is_set():
             progress(downloaded="[yellow]SKIPPING")
 
         if DOWNLOAD_CANCELLED.is_set():
-            progress(downloaded="[yellow]SKIPPED")
+            progress(downloaded="[yellow]CANCELLED")
             return
 
-        log = logging.getLogger("track")
+        log = logging.getLogger("Track")
 
         proxy = next(iter(session.proxies.values()), None)
 
-        track_type = self.__class__.__name__
+        save_path = config.directories.temp / f"{self.__class__.__name__}_{self.id}.mp4"
-        save_path = config.directories.temp / f"{track_type}_{self.id}.mp4"
+        if self.__class__.__name__ == "Subtitle":
-        if track_type == "Subtitle":
             save_path = save_path.with_suffix(f".{self.codec.extension}")
 
         if self.descriptor != self.Descriptor.URL:
@@ -175,15 +245,14 @@ class Track:
         def cleanup():
             # track file (e.g., "foo.mp4")
             save_path.unlink(missing_ok=True)
-            # aria2c control file (e.g., "foo.mp4.aria2" or "foo.mp4.aria2__temp")
+            # aria2c control file (e.g., "foo.mp4.aria2")
             save_path.with_suffix(f"{save_path.suffix}.aria2").unlink(missing_ok=True)
-            save_path.with_suffix(f"{save_path.suffix}.aria2__temp").unlink(missing_ok=True)
             if save_dir.exists() and save_dir.name.endswith("_segments"):
                 shutil.rmtree(save_dir)
 
         if not DOWNLOAD_LICENCE_ONLY.is_set():
             if config.directories.temp.is_file():
-                raise ValueError(f"Temp Directory '{config.directories.temp}' must be a Directory, not a file")
+                raise EnvironmentError(f"Temp Directory '{config.directories.temp}' must be a Directory, not a file")
 
             config.directories.temp.mkdir(parents=True, exist_ok=True)
 
@@ -194,7 +263,8 @@ class Track:
             cleanup()
 
         try:
-            if self.descriptor == self.Descriptor.HLS:
+            if self.descriptor == self.Descriptor.M3U:
+                from devine.core.manifests import HLS
                 HLS.download_track(
                     track=self,
                     save_path=save_path,
@@ -204,7 +274,8 @@ class Track:
                     proxy=proxy,
                     license_widevine=prepare_drm
                 )
-            elif self.descriptor == self.Descriptor.DASH:
+            elif self.descriptor == self.Descriptor.MPD:
+                from devine.core.manifests import DASH
                 DASH.download_track(
                     track=self,
                     save_path=save_path,
@@ -214,9 +285,10 @@ class Track:
                     proxy=proxy,
                     license_widevine=prepare_drm
                 )
-            elif self.descriptor == self.Descriptor.URL:
+            # no else-if as DASH may convert the track to URL descriptor
+            if self.descriptor == self.Descriptor.URL:
                 try:
-                    if not self.drm and track_type in ("Video", "Audio"):
+                    if not self.drm and self.__class__.__name__ in ("Video", "Audio"):
                         # the service might not have explicitly defined the `drm` property
                         # try find widevine DRM information from the init data of URL
                         try:
@@ -241,40 +313,29 @@ class Track:
                     if DOWNLOAD_LICENCE_ONLY.is_set():
                         progress(downloaded="[yellow]SKIPPED")
                     else:
-                        for status_update in self.downloader(
+                        downloader(
-                            urls=self.url,
+                            uri=self.url,
-                            output_dir=save_path.parent,
+                            out=save_path,
-                            filename=save_path.name,
                             headers=session.headers,
                             cookies=session.cookies,
-                            proxy=proxy
+                            proxy=proxy,
-                        ):
+                            progress=progress
-                            file_downloaded = status_update.get("file_downloaded")
+                        )
-                            if not file_downloaded:
-                                progress(**status_update)
-
-                        # see https://github.com/devine-dl/devine/issues/71
-                        save_path.with_suffix(f"{save_path.suffix}.aria2__temp").unlink(missing_ok=True)
 
                         self.path = save_path
-                        if callable(self.OnDownloaded):
-                            self.OnDownloaded()
 
                         if drm:
                             progress(downloaded="Decrypting", completed=0, total=100)
                             drm.decrypt(save_path)
                             self.drm = None
                             if callable(self.OnDecrypted):
-                                self.OnDecrypted(drm)
+                                self.OnDecrypted(self)
                             progress(downloaded="Decrypted", completed=100)
 
-                        if track_type == "Subtitle" and self.codec.name not in ("fVTT", "fTTML"):
+                        if self.__class__.__name__ == "Subtitle":
                             track_data = self.path.read_bytes()
                             track_data = try_ensure_utf8(track_data)
-                            track_data = track_data.decode("utf8"). \
+                            track_data = html.unescape(track_data.decode("utf8")).encode("utf8")
-                                replace("‎", html.unescape("‎")). \
-                                replace("‏", html.unescape("‏")). \
-                                encode("utf8")
                             self.path.write_bytes(track_data)
 
                         progress(downloaded="Downloaded")
@@ -307,170 +368,6 @@ class Track:
             self.path.unlink()
             self.path = None
 
-    def move(self, target: Union[Path, str]) -> Path:
-        """
-        Move the Track's file from current location, to target location.
-        This will overwrite anything at the target path.
-
-        Raises:
-            TypeError: If the target argument is not the expected type.
-            ValueError: If track has no file to move, or the target does not exist.
-            OSError: If the file somehow failed to move.
-
-        Returns the new location of the track.
-        """
-        if not isinstance(target, (str, Path)):
-            raise TypeError(f"Expected {target} to be a {Path} or {str}, not {type(target)}")
-
-        if not self.path:
-            raise ValueError("Track has no file to move")
-
-        if not isinstance(target, Path):
-            target = Path(target)
-
-        if not target.exists():
-            raise ValueError(f"Target file {repr(target)} does not exist")
-
-        moved_to = Path(shutil.move(self.path, target))
-        if moved_to.resolve() != target.resolve():
-            raise OSError(f"Failed to move {self.path} to {target}")
-
-        self.path = target
-        return target
-
-    def get_track_name(self) -> Optional[str]:
-        """Get the Track Name."""
-        return self.name
-
-    def get_key_id(self, init_data: Optional[bytes] = None, *args, **kwargs) -> Optional[UUID]:
-        """
-        Probe the DRM encryption Key ID (KID) for this specific track.
-
-        It currently supports finding the Key ID by probing the track's stream
-        with ffprobe for `enc_key_id` data, as well as for mp4 `tenc` (Track
-        Encryption) boxes.
-
-        It explicitly ignores PSSH information like the `PSSH` box, as the box
-        is likely to contain multiple Key IDs that may or may not be for this
-        specific track.
-
-        To retrieve the initialization segment, this method calls :meth:`get_init_segment`
-        with the positional and keyword arguments. The return value of `get_init_segment`
-        is then used to determine the Key ID.
-
-        Returns:
-            The Key ID as a UUID object, or None if the Key ID could not be determined.
-        """
-        if not init_data:
-            init_data = self.get_init_segment(*args, **kwargs)
-        if not isinstance(init_data, bytes):
-            raise TypeError(f"Expected init_data to be bytes, not {init_data!r}")
-
-        probe = ffprobe(init_data)
-        if probe:
-            for stream in probe.get("streams") or []:
-                enc_key_id = stream.get("tags", {}).get("enc_key_id")
-                if enc_key_id:
-                    return UUID(bytes=base64.b64decode(enc_key_id))
-
-        for tenc in get_boxes(init_data, b"tenc"):
-            if tenc.key_ID.int != 0:
-                return tenc.key_ID
-
-        for uuid_box in get_boxes(init_data, b"uuid"):
-            if uuid_box.extended_type == UUID("8974dbce-7be7-4c51-84f9-7148f9882554"):  # tenc
-                tenc = uuid_box.data
-                if tenc.key_ID.int != 0:
-                    return tenc.key_ID
-
-    def get_init_segment(
-        self,
-        maximum_size: int = 20000,
-        url: Optional[str] = None,
-        byte_range: Optional[str] = None,
-        session: Optional[Session] = None
-    ) -> bytes:
-        """
-        Get the Track's Initial Segment Data Stream.
-
-        HLS and DASH tracks must explicitly provide a URL to the init segment or file.
-        Providing the byte-range for the init segment is recommended where possible.
-
-        If `byte_range` is not set, it will make a HEAD request and check the size of
-        the file. If the size could not be determined, it will download up to the first
-        20KB only, which should contain the entirety of the init segment. You may
-        override this by changing the `maximum_size`.
-
-        The default maximum_size of 20000 (20KB) is a tried-and-tested value that
-        seems to work well across the board.
-
-        Parameters:
-            maximum_size: Size to assume as the content length if byte-range is not
-                used, the content size could not be determined, or the content size
-                is larger than it. A value of 20000 (20KB) or higher is recommended.
-            url: Explicit init map or file URL to probe from.
-            byte_range: Range of bytes to download from the explicit or implicit URL.
-            session: Session context, e.g., authorization and headers.
-        """
-        if not isinstance(maximum_size, int):
-            raise TypeError(f"Expected maximum_size to be an {int}, not {type(maximum_size)}")
-        if not isinstance(url, (str, type(None))):
-            raise TypeError(f"Expected url to be a {str}, not {type(url)}")
-        if not isinstance(byte_range, (str, type(None))):
-            raise TypeError(f"Expected byte_range to be a {str}, not {type(byte_range)}")
-        if not isinstance(session, (Session, type(None))):
-            raise TypeError(f"Expected session to be a {Session}, not {type(session)}")
-
-        if not url:
-            if self.descriptor != self.Descriptor.URL:
-                raise ValueError(f"An explicit URL must be provided for {self.descriptor.name} tracks")
-            if not self.url:
-                raise ValueError("An explicit URL must be provided as the track has no URL")
-            url = self.url
-
-        if not session:
-            session = Session()
-
-        content_length = maximum_size
-
-        if byte_range:
-            if not isinstance(byte_range, str):
-                raise TypeError(f"Expected byte_range to be a str, not {byte_range!r}")
-            if not re.match(r"^\d+-\d+$", byte_range):
-                raise ValueError(f"The value of byte_range is unrecognized: '{byte_range}'")
-            start, end = byte_range.split("-")
-            if start > end:
-                raise ValueError(f"The start range cannot be greater than the end range: {start}>{end}")
-        else:
-            size_test = session.head(url)
-            if "Content-Length" in size_test.headers:
-                content_length_header = int(size_test.headers["Content-Length"])
-                if content_length_header > 0:
-                    content_length = min(content_length_header, maximum_size)
-            range_test = session.head(url, headers={"Range": "bytes=0-1"})
-            if range_test.status_code == 206:
-                byte_range = f"0-{content_length-1}"
-
-        if byte_range:
-            res = session.get(
-                url=url,
-                headers={
-                    "Range": f"bytes={byte_range}"
-                }
-            )
-            res.raise_for_status()
-            init_data = res.content
-        else:
-            init_data = None
-            with session.get(url, stream=True) as s:
-                for chunk in s.iter_content(content_length):
-                    init_data = chunk
-                    break
-            if not init_data:
-                raise ValueError(f"Failed to read {content_length} bytes from the track URI.")
-
-        return init_data
-
     def repackage(self) -> None:
         if not self.path or not self.path.exists():
             raise ValueError("Cannot repackage a Track that has not been downloaded.")
@@ -509,7 +406,36 @@ class Track:
             else:
                 raise
 
-        self.path = output_path
+        self.swap(output_path)
+        self.move(original_path)
+
+    def move(self, target: Union[str, Path]) -> bool:
+        """
+        Move the Track's file from current location, to target location.
+        This will overwrite anything at the target path.
+        """
+        if not self.path:
+            return False
+        target = Path(target)
+
+        ok = Path(shutil.move(self.path, target)).resolve() == target.resolve()
+        if ok:
+            self.path = target
+        return ok
+
+    def swap(self, target: Union[str, Path]) -> bool:
+        """
+        Swaps the Track's file with the Target file. The current Track's file is deleted.
+        Returns False if the Track is not yet downloaded, or the target path does not exist.
+        """
+        target = Path(target)
+        if not target.exists() or not self.path:
+            return False
+        self.path.unlink()
+        ok = Path(shutil.move(target, self.path)).resolve() == self.path.resolve()
+        if not ok:
+            return False
+        return self.move(target)
 
 
 __all__ = ("Track",)

devine/core/tracks/tracks.py

@@ -6,6 +6,7 @@ from functools import partial
 from pathlib import Path
 from typing import Callable, Iterator, Optional, Sequence, Union
 
+from Cryptodome.Random import get_random_bytes
 from langcodes import Language, closest_supported_match
 from rich.progress import BarColumn, Progress, SpinnerColumn, TextColumn, TimeRemainingColumn
 from rich.table import Table
@@ -13,9 +14,9 @@ from rich.tree import Tree
 
 from devine.core.config import config
 from devine.core.console import console
-from devine.core.constants import LANGUAGE_MAX_DISTANCE, AnyTrack, TrackT
+from devine.core.constants import LANGUAGE_MAX_DISTANCE, LANGUAGE_MUX_MAP, AnyTrack, TrackT
 from devine.core.tracks.audio import Audio
-from devine.core.tracks.chapters import Chapter, Chapters
+from devine.core.tracks.chapter import Chapter
 from devine.core.tracks.subtitle import Subtitle
 from devine.core.tracks.track import Track
 from devine.core.tracks.video import Video
@@ -36,11 +37,11 @@ class Tracks:
         Chapter: 3
     }
 
-    def __init__(self, *args: Union[Tracks, Sequence[Union[AnyTrack, Chapter, Chapters]], Track, Chapter, Chapters]):
+    def __init__(self, *args: Union[Tracks, list[Track], Track]):
         self.videos: list[Video] = []
         self.audio: list[Audio] = []
         self.subtitles: list[Subtitle] = []
-        self.chapters = Chapters()
+        self.chapters: list[Chapter] = []
 
         if args:
             self.add(args)
@@ -51,13 +52,6 @@ class Tracks:
     def __len__(self) -> int:
         return len(self.videos) + len(self.audio) + len(self.subtitles)
 
-    def __add__(
-        self,
-        other: Union[Tracks, Sequence[Union[AnyTrack, Chapter, Chapters]], Track, Chapter, Chapters]
-    ) -> Tracks:
-        self.add(other)
-        return self
-
     def __repr__(self) -> str:
         return "{name}({items})".format(
             name=self.__class__.__name__,
@@ -143,7 +137,7 @@ class Tracks:
 
     def add(
         self,
-        tracks: Union[Tracks, Sequence[Union[AnyTrack, Chapter, Chapters]], Track, Chapter, Chapters],
+        tracks: Union[Tracks, Sequence[Union[AnyTrack, Chapter]], Track, Chapter],
         warn_only: bool = False
     ) -> None:
         """Add a provided track to its appropriate array and ensuring it's not a duplicate."""
@@ -172,7 +166,7 @@ class Tracks:
             elif isinstance(track, Subtitle):
                 self.subtitles.append(track)
             elif isinstance(track, Chapter):
-                self.chapters.add(track)
+                self.chapters.append(track)
             else:
                 raise ValueError("Track type was not set or is invalid.")
 
@@ -249,6 +243,13 @@ class Tracks:
                 continue
             self.subtitles.sort(key=lambda x: is_close_match(language, [x.language]), reverse=True)
 
+    def sort_chapters(self) -> None:
+        """Sort chapter tracks by chapter number."""
+        if not self.chapters:
+            return
+        # number
+        self.chapters.sort(key=lambda x: x.number)
+
     def select_video(self, x: Callable[[Video], bool]) -> None:
         self.videos = list(filter(x, self.videos))
 
@@ -288,6 +289,16 @@ class Tracks:
             ][:per_language or None])
         return selected
 
+    def export_chapters(self, to_file: Optional[Union[Path, str]] = None) -> str:
+        """Export all chapters in order to a string or file."""
+        self.sort_chapters()
+        data = "\n".join(map(repr, self.chapters))
+        if to_file:
+            to_file = Path(to_file)
+            to_file.parent.mkdir(parents=True, exist_ok=True)
+            to_file.write_text(data, encoding="utf8")
+        return data
+
     def mux(self, title: str, delete: bool = True, progress: Optional[partial] = None) -> tuple[Path, int]:
         """
         Multiplex all the Tracks into a Matroska Container file.
@@ -311,9 +322,11 @@ class Tracks:
             if not vt.path or not vt.path.exists():
                 raise ValueError("Video Track must be downloaded before muxing...")
             if callable(vt.OnMultiplex):
-                vt.OnMultiplex()
+                vt.OnMultiplex(vt)
             cl.extend([
-                "--language", f"0:{vt.language}",
+                "--language", "0:{}".format(LANGUAGE_MUX_MAP.get(
+                    str(vt.language), str(vt.language)
+                )),
                 "--default-track", f"0:{i == 0}",
                 "--original-flag", f"0:{vt.is_original_lang}",
                 "--compression", "0:none",  # disable extra compression
@@ -324,10 +337,12 @@ class Tracks:
             if not at.path or not at.path.exists():
                 raise ValueError("Audio Track must be downloaded before muxing...")
             if callable(at.OnMultiplex):
-                at.OnMultiplex()
+                at.OnMultiplex(at)
             cl.extend([
                 "--track-name", f"0:{at.get_track_name() or ''}",
-                "--language", f"0:{at.language}",
+                "--language", "0:{}".format(LANGUAGE_MUX_MAP.get(
+                    str(at.language), str(at.language)
+                )),
                 "--default-track", f"0:{i == 0}",
                 "--visual-impaired-flag", f"0:{at.descriptive}",
                 "--original-flag", f"0:{at.is_original_lang}",
@@ -339,11 +354,13 @@ class Tracks:
             if not st.path or not st.path.exists():
                 raise ValueError("Text Track must be downloaded before muxing...")
             if callable(st.OnMultiplex):
-                st.OnMultiplex()
+                st.OnMultiplex(st)
             default = bool(self.audio and is_close_match(st.language, [self.audio[0].language]) and st.forced)
             cl.extend([
                 "--track-name", f"0:{st.get_track_name() or ''}",
-                "--language", f"0:{st.language}",
+                "--language", "0:{}".format(LANGUAGE_MUX_MAP.get(
+                    str(st.language), str(st.language)
+                )),
                 "--sub-charset", "0:UTF-8",
                 "--forced-track", f"0:{st.forced}",
                 "--default-track", f"0:{default}",
@@ -356,9 +373,9 @@ class Tracks:
         if self.chapters:
             chapters_path = config.directories.temp / config.filenames.chapters.format(
                 title=sanitize_filename(title),
-                random=self.chapters.id
+                random=get_random_bytes(16).hex()
             )
-            self.chapters.dump(chapters_path, fallback_name=config.chapter_fallback_name)
+            self.export_chapters(chapters_path)
             cl.extend(["--chapter-charset", "UTF-8", "--chapters", str(chapters_path)])
         else:
             chapters_path = None

devine/core/tracks/video.py

@@ -200,8 +200,8 @@ class Video(Track):
             str(output_path)
         ], check=True)
 
-        self.path = output_path
+        self.swap(output_path)
-        original_path.unlink()
+        self.move(original_path)
 
     def ccextractor(
         self, track_id: Any, out_path: Union[Path, str], language: Language, original: bool = False
@@ -321,12 +321,11 @@ class Video(Track):
         i = file.index(b"x264")
         encoding_settings = file[i: i + file[i:].index(b"\x00")].replace(b":", br"\\:").replace(b",", br"\,").decode()
 
-        original_path = self.path
+        cleaned_path = self.path.with_suffix(f".cleaned{self.path.suffix}")
-        cleaned_path = original_path.with_suffix(f".cleaned{original_path.suffix}")
         subprocess.run([
             executable, "-hide_banner",
             "-loglevel", "panic",
-            "-i", original_path,
+            "-i", self.path,
             "-map_metadata", "-1",
             "-fflags", "bitexact",
             "-bsf:v", f"filter_units=remove_types=6,h264_metadata=sei_user_data={uuid}+{encoding_settings}",
@@ -336,8 +335,7 @@ class Video(Track):
 
         log.info(" + Removed")
 
-        self.path = cleaned_path
+        self.swap(cleaned_path)
-        original_path.unlink()
 
         return True
 

devine/core/utilities.py

@@ -1,10 +1,8 @@
 import ast
 import contextlib
 import importlib.util
-import os
 import re
 import shutil
-import socket
 import sys
 import time
 import unicodedata
@@ -12,10 +10,11 @@ from collections import defaultdict
 from datetime import datetime
 from pathlib import Path
 from types import ModuleType
-from typing import Optional, Sequence, Union
+from typing import AsyncIterator, Optional, Sequence, Union
-from urllib.parse import ParseResult, urlparse
+from urllib.parse import urlparse
 
 import chardet
+import pproxy
 import requests
 from construct import ValidationError
 from langcodes import Language, closest_match
@@ -245,36 +244,35 @@ def try_ensure_utf8(data: bytes) -> bytes:
                 return data
 
 
-def get_free_port() -> int:
+@contextlib.asynccontextmanager
-    """
+async def start_pproxy(proxy: str) -> AsyncIterator[str]:
-    Get an available port to use between a-b (inclusive).
+    proxy = urlparse(proxy)
 
-    The port is freed as soon as this has returned, therefore, it
+    scheme = {
-    is possible for the port to be taken before you try to use it.
+        "https": "http+ssl",
-    """
+        "socks5h": "socks"
-    with contextlib.closing(socket.socket(socket.AF_INET, socket.SOCK_STREAM)) as s:
+    }.get(proxy.scheme, proxy.scheme)
-        s.bind(("", 0))
-        s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
-        return s.getsockname()[1]
 
+    remote_server = f"{scheme}://{proxy.hostname}"
+    if proxy.port:
+        remote_server += f":{proxy.port}"
+    if proxy.username or proxy.password:
+        remote_server += "#"
+    if proxy.username:
+        remote_server += proxy.username
+    if proxy.password:
+        remote_server += f":{proxy.password}"
 
-def get_extension(value: Union[str, Path, ParseResult]) -> Optional[str]:
+    server = pproxy.Server("http://localhost:0")  # random port
-    """
+    remote = pproxy.Connection(remote_server)
-    Get a URL or Path file extension/suffix.
+    handler = await server.start_server({"rserver": [remote]})
 
-    Note: The returned value will begin with `.`.
+    try:
-    """
+        port = handler.sockets[0].getsockname()[1]
-    if isinstance(value, ParseResult):
+        yield f"http://localhost:{port}"
-        value_parsed = value
+    finally:
-    elif isinstance(value, (str, Path)):
+        handler.close()
-        value_parsed = urlparse(str(value))
+        await handler.wait_closed()
-    else:
-        raise TypeError(f"Expected {str}, {Path}, or {ParseResult}, got {type(value)}")
-
-    if value_parsed.path:
-        ext = os.path.splitext(value_parsed.path)[1]
-        if ext and ext != ".":
-            return ext
 
 
 class FPS(ast.NodeVisitor):

devine/core/utils/click_types.py

@@ -1,8 +1,7 @@
 import re
-from typing import Any, Optional, Union
+from typing import Optional, Union
 
 import click
-from click.shell_completion import CompletionItem
 from pywidevine.cdm import Cdm as WidevineCdm
 
 
@@ -123,62 +122,6 @@ class QualityList(click.ParamType):
         return sorted(resolutions, reverse=True)
 
 
-class MultipleChoice(click.Choice):
-    """
-    The multiple choice type allows multiple values to be checked against
-    a fixed set of supported values.
-
-    It internally uses and is based off of click.Choice.
-    """
-
-    name = "multiple_choice"
-
-    def __repr__(self) -> str:
-        return f"MultipleChoice({list(self.choices)})"
-
-    def convert(
-        self,
-        value: Any,
-        param: Optional[click.Parameter] = None,
-        ctx: Optional[click.Context] = None
-    ) -> list[Any]:
-        if not value:
-            return []
-        if isinstance(value, str):
-            values = value.split(",")
-        elif isinstance(value, list):
-            values = value
-        else:
-            self.fail(
-                f"{value!r} is not a supported value.",
-                param,
-                ctx
-            )
-
-        chosen_values: list[Any] = []
-        for value in values:
-            chosen_values.append(super().convert(value, param, ctx))
-
-        return chosen_values
-
-    def shell_complete(
-        self,
-        ctx: click.Context,
-        param: click.Parameter,
-        incomplete: str
-    ) -> list[CompletionItem]:
-        """
-        Complete choices that start with the incomplete value.
-
-        Parameters:
-            ctx: Invocation context for this command.
-            param: The parameter that is requesting completion.
-            incomplete: Value being completed. May be empty.
-        """
-        incomplete = incomplete.rsplit(",")[-1]
-        return super(self).shell_complete(ctx, param, incomplete)
-
-
 SEASON_RANGE = SeasonRange()
 LANGUAGE_RANGE = LanguageRange()
 QUALITY_LIST = QualityList()

devine/vaults/API.py

@@ -1,214 +0,0 @@
-from typing import Iterator, Optional, Union
-from uuid import UUID
-
-from requests import Session
-
-from devine.core import __version__
-from devine.core.vault import Vault
-
-
-class API(Vault):
-    """Key Vault using a simple RESTful HTTP API call."""
-
-    def __init__(self, name: str, uri: str, token: str):
-        super().__init__(name)
-        self.uri = uri.rstrip("/")
-        self.session = Session()
-        self.session.headers.update({
-            "User-Agent": f"Devine v{__version__}"
-        })
-        self.session.headers.update({
-            "Authorization": f"Bearer {token}"
-        })
-
-    def get_key(self, kid: Union[UUID, str], service: str) -> Optional[str]:
-        if isinstance(kid, UUID):
-            kid = kid.hex
-
-        data = self.session.get(
-            url=f"{self.uri}/{service.lower()}/{kid}",
-            headers={
-                "Accept": "application/json"
-            }
-        ).json()
-
-        code = int(data.get("code", 0))
-        message = data.get("message")
-        error = {
-            0: None,
-            1: Exceptions.AuthRejected,
-            2: Exceptions.TooManyRequests,
-            3: Exceptions.ServiceTagInvalid,
-            4: Exceptions.KeyIdInvalid
-        }.get(code, ValueError)
-
-        if error:
-            raise error(f"{message} ({code})")
-
-        content_key = data.get("content_key")
-        if not content_key:
-            return None
-
-        if not isinstance(content_key, str):
-            raise ValueError(f"Expected {content_key} to be {str}, was {type(content_key)}")
-
-        return content_key
-
-    def get_keys(self, service: str) -> Iterator[tuple[str, str]]:
-        page = 1
-
-        while True:
-            data = self.session.get(
-                url=f"{self.uri}/{service.lower()}",
-                params={
-                    "page": page,
-                    "total": 10
-                },
-                headers={
-                    "Accept": "application/json"
-                }
-            ).json()
-
-            code = int(data.get("code", 0))
-            message = data.get("message")
-            error = {
-                0: None,
-                1: Exceptions.AuthRejected,
-                2: Exceptions.TooManyRequests,
-                3: Exceptions.PageInvalid,
-                4: Exceptions.ServiceTagInvalid,
-            }.get(code, ValueError)
-
-            if error:
-                raise error(f"{message} ({code})")
-
-            content_keys = data.get("content_keys")
-            if content_keys:
-                if not isinstance(content_keys, dict):
-                    raise ValueError(f"Expected {content_keys} to be {dict}, was {type(content_keys)}")
-
-                for key_id, key in content_keys.items():
-                    yield key_id, key
-
-            pages = int(data["pages"])
-            if pages <= page:
-                break
-
-            page += 1
-
-    def add_key(self, service: str, kid: Union[UUID, str], key: str) -> bool:
-        if isinstance(kid, UUID):
-            kid = kid.hex
-
-        data = self.session.post(
-            url=f"{self.uri}/{service.lower()}/{kid}",
-            json={
-                "content_key": key
-            },
-            headers={
-                "Accept": "application/json"
-            }
-        ).json()
-
-        code = int(data.get("code", 0))
-        message = data.get("message")
-        error = {
-            0: None,
-            1: Exceptions.AuthRejected,
-            2: Exceptions.TooManyRequests,
-            3: Exceptions.ServiceTagInvalid,
-            4: Exceptions.KeyIdInvalid,
-            5: Exceptions.ContentKeyInvalid
-        }.get(code, ValueError)
-
-        if error:
-            raise error(f"{message} ({code})")
-
-        # the kid:key was new to the vault (optional)
-        added = bool(data.get("added"))
-        # the key for kid was changed/updated (optional)
-        updated = bool(data.get("updated"))
-
-        return added or updated
-
-    def add_keys(self, service: str, kid_keys: dict[Union[UUID, str], str]) -> int:
-        data = self.session.post(
-            url=f"{self.uri}/{service.lower()}",
-            json={
-                "content_keys": {
-                    str(kid).replace("-", ""): key
-                    for kid, key in kid_keys.items()
-                }
-            },
-            headers={
-                "Accept": "application/json"
-            }
-        ).json()
-
-        code = int(data.get("code", 0))
-        message = data.get("message")
-        error = {
-            0: None,
-            1: Exceptions.AuthRejected,
-            2: Exceptions.TooManyRequests,
-            3: Exceptions.ServiceTagInvalid,
-            4: Exceptions.KeyIdInvalid,
-            5: Exceptions.ContentKeyInvalid
-        }.get(code, ValueError)
-
-        if error:
-            raise error(f"{message} ({code})")
-
-        # each kid:key that was new to the vault (optional)
-        added = int(data.get("added"))
-        # each key for a kid that was changed/updated (optional)
-        updated = int(data.get("updated"))
-
-        return added + updated
-
-    def get_services(self) -> Iterator[str]:
-        data = self.session.post(
-            url=self.uri,
-            headers={
-                "Accept": "application/json"
-            }
-        ).json()
-
-        code = int(data.get("code", 0))
-        message = data.get("message")
-        error = {
-            0: None,
-            1: Exceptions.AuthRejected,
-            2: Exceptions.TooManyRequests,
-        }.get(code, ValueError)
-
-        if error:
-            raise error(f"{message} ({code})")
-
-        service_list = data.get("service_list", [])
-
-        if not isinstance(service_list, list):
-            raise ValueError(f"Expected {service_list} to be {list}, was {type(service_list)}")
-
-        for service in service_list:
-            yield service
-
-
-class Exceptions:
-    class AuthRejected(Exception):
-        """Authentication Error Occurred, is your token valid? Do you have permission to make this call?"""
-
-    class TooManyRequests(Exception):
-        """Rate Limited; Sent too many requests in a given amount of time."""
-
-    class PageInvalid(Exception):
-        """Requested page does not exist."""
-
-    class ServiceTagInvalid(Exception):
-        """The Service Tag is invalid."""
-
-    class KeyIdInvalid(Exception):
-        """The Key ID is invalid."""
-
-    class ContentKeyInvalid(Exception):
-        """The Content Key is invalid."""

pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "devine"
-version = "3.1.0"
+version = "2.2.0"
-description = "Modular Movie, TV, and Music Archival Software."
+description = "Open-Source Movie, TV, and Music Downloading Solution."
 license = "GPL-3.0-only"
 authors = ["rlaphoenix <rlaphoenix@pm.me>"]
 readme = "README.md"
@@ -39,43 +39,39 @@ Brotli = "^1.1.0"
 click = "^8.1.7"
 construct = "^2.8.8"
 crccheck = "^1.3.0"
-jsonpickle = "^3.0.3"
+jsonpickle = "^3.0.2"
 langcodes = { extras = ["data"], version = "^3.3.0" }
-lxml = "^5.1.0"
+lxml = "^4.9.3"
-pproxy = "^2.7.9"
+pproxy = "^2.7.8"
-protobuf = "^4.25.3"
+protobuf = "^4.24.4"
-pycaption = "^2.2.4"
+pycaption = "^2.2.0"
-pycryptodomex = "^3.20.0"
+pycryptodomex = "^3.19.0"
 pyjwt = "^2.8.0"
 pymediainfo = "^6.1.0"
 pymp4 = "^1.4.0"
 pymysql = "^1.1.0"
-pywidevine = { extras = ["serve"], version = "^1.8.0" }
+pywidevine = { extras = ["serve"], version = "^1.7.0" }
 PyYAML = "^6.0.1"
 requests = { extras = ["socks"], version = "^2.31.0" }
-rich = "^13.7.1"
+rich = "^13.7.0"
 "rlaphoenix.m3u8" = "^3.4.0"
-"ruamel.yaml" = "^0.18.6"
+"ruamel.yaml" = "^0.17.40"
 sortedcontainers = "^2.4.0"
 subtitle-filter = "^1.4.8"
-Unidecode = "^1.3.8"
+Unidecode = "^1.3.7"
-urllib3 = "^2.2.1"
+urllib3 = "^2.1.0"
 chardet = "^5.2.0"
-curl-cffi = "^0.6.1"
+curl-cffi = "^0.5.10"
-# Temporary explicit versions of these langcodes dependencies as language-data v1.1
-# uses marisa-trie v0.7.8 which doesn't have Python 3.12 wheels.
-language-data = "^1.2.0.dev3"
-marisa-trie = "^1.1.0"
 
 [tool.poetry.dev-dependencies]
-pre-commit = "^3.6.2"
+pre-commit = "^3.5.0"
-mypy = "^1.8.0"
+mypy = "^1.7.1"
 mypy-protobuf = "^3.5.0"
-types-protobuf = "^4.24.0.20240129"
+types-protobuf = "^4.24.0.4"
 types-PyMySQL = "^1.1.0.1"
-types-requests = "^2.31.0.20240218"
+types-requests = "^2.31.0.10"
-isort = "^5.13.2"
+isort = "^5.12.0"
-ruff = "~0.3.0"
+ruff = "~0.1.6"
 
 [tool.poetry.scripts]
 devine = "devine.core.__main__:main"
@@ -83,8 +79,6 @@ devine = "devine.core.__main__:main"
 [tool.ruff]
 force-exclude = true
 line-length = 120
-
-[tool.ruff.lint]
 select = ["E4", "E7", "E9", "F", "W"]
 
 [tool.isort]