dvs/ap_ds

Files

T

dvs 646c1b11c8 Initial commit: ap_ds 音频播放库 (Audio Player By DVS 原版)

2026-08-27 19:13:25 +08:00

112 KiB

Raw Permalink Blame History

🎉 ap_ds 4.1.0 RC Prerelease — Opus Support, But Please Help Us Test!

⚠️ Important: 4.1.0 is an RC (Release Candidate) prerelease version for collecting feedback and bug reports.

Opus support currently exists in ONLY this one mainline version (4.1.0). Future mainline versions (4.2.0+) will REMOVE Opus support. Opus will migrate to the AFS branch (ap-ds-afs) for independent development.

If you find any issues, please contact us immediately:

📧 Primary email: me@dvsyun.top

📧 Backup email: dvs6666@163.com

⏱️ We promise: Fix within 3 business days

Your feedback is crucial to us! Let's polish Opus support to perfection together.

"We're not just supporting a new format — we're opening a door to higher quality, smaller size, and a freer future." — DVS Development Team, August 19, 2026

ap_ds: Lightweight Python Audio Library

Current Version: v4.1.0 RC – Opus Support Preview

Release Date: August 19, 2026

Version Status: 🚀 Release Candidate — Testing Feedback Welcome

📢 To All Audio Developers

Friends, colleagues, music lovers, and all who have ever wrestled with audio formats late into the night:

Today, we announce — ap_ds 4.1.0 RC is officially released, supporting the Opus audio format in the mainline branch for the very first time!

But before the excitement takes over, we must be honest with you:

This is an RC (Release Candidate) prerelease version.

What does that mean? It means:

✅ All core features are implemented and tested
✅ 229 Opus-specific tests all pass
✅ Cross-platform (Windows/Linux/macOS) playback verified
⚠️ There may still be edge-case bugs we haven't discovered
⚠️ We need real users to verify it in different environments

So we're handing this version to you — our users — to help us test.

"4.1.0 is the only appearance of Opus in the mainline. After this, it will move to the AFS branch."

This is part of ap_ds's dual-track strategy. We'll explain in detail in the following sections.

🚨 Important Notes on 4.1.0's Version Positioning

1. This Is an RC Prerelease

4.1.0 is NOT an LTS (Long-Term Support) version, nor is it the final stable release. It is a Release Candidate, intended for:

Collecting real-world usage feedback
Discovering bugs that we couldn't cover in testing
Verifying cross-platform compatibility
Gathering data for the final release

2. Opus Support Exists ONLY in 4.1.0 Mainline

This is a very important statement:

Opus support currently exists in ONLY version 4.1.0 of the ap_ds mainline branch.

Future mainline versions (4.2.0, 4.3.0, etc.) will REMOVE Opus support.

Opus and all future new formats will migrate to the AFS branch (ap-ds-afs), developing independently with 1.x version numbers.

Here's what this means:

Version	Opus Included?	Description
4.1.0 RC	✅ Yes	The ONLY mainline version with Opus
4.2.0+ (mainline)	❌ No	Opus removed, back to lightweight positioning
ap-ds-afs 1.0.0+	✅ Yes	Permanent home for Opus and all new formats

Why?

Because ap_ds mainline's core promise is 2.5MB lightweight. Opus support (including DLLs) increased the size to 3.87MB. 4.2.0 will directly remove Opus, returning to 2.5MB — more thorough than any optimization.

So, if you need Opus support:

Short-term testing: Use 4.1.0 RC
Long-term use: Switch to the AFS branch (pip install ap-ds-afs)

Both packages use the exact same import method (from ap_ds import AudioLibrary), so migration cost is zero.

3. Issue Reporting and Fix Commitment

If you discover any issues while using 4.1.0 RC:

Contact the author immediately:
- Primary email: me@dvsyun.top
- Backup email: dvs6666@163.com
- Please CC both emails to ensure delivery
We promise:
- Fix within 3 business days
- Patch version (4.1.1) will be released as soon as possible
When reporting, please provide:
- Operating system and version
- Python version (python --version)
- Full error message (if any)
- Steps to reproduce
- Audio file sample (if possible)

Every piece of feedback helps us build a better ap_ds. 🙏

🤔 Why Opus? — The Inevitability of a Technical Choice

1. The "Ceiling" of Quality and Compression

Opus is an open audio codec jointly developed by the Xiph.Org Foundation and IETF, combining SILK (for speech) and CELT (for general audio) algorithms, with adaptive bitrates from 6 kbps to 510 kbps. This means:

At low bitrates (< 32 kbps), Opus voice clarity far exceeds MP3 and AAC
At medium-high bitrates (64–128 kbps), Opus quality rivals or even surpasses MP3 at 320kbps
Ultra-low latency (as low as 5ms), ideal for real-time communication and gaming

2. Open Source and Freedom — A Philosophical Fit

Opus uses a BSD-style license with no patent restrictions, completely free. This aligns perfectly with ap_ds's commitment to openness, freedom, and zero burden.

3. Mature Ecosystem and Clear Demand

Today, Opus is widely used in:

WebRTC (real-time audio/video communication)
Discord, WhatsApp, Signal and other instant messaging apps
Game engines (Unity, Unreal both support Opus)
Audio streaming (Radio, Podcasts)
Embedded devices (low power, high compression)

As more audio content is published in Opus format, ap_ds as a general-purpose audio library must respond to this trend.

4. Paving the Way for the AFS Branch

Opus is the first member of the AFS (All-Format Support) branch. Through the 4.1.0 RC practice, we will validate the technical solution for cross-platform Opus playback, accumulating experience for the official release of the AFS branch.

😅 Why Wasn't Opus Supported Before? — The Story of Upstream Dependencies

That's a great question. To be honest, we always wanted to support Opus, since ap_ds 1.0. But the reality was:

ap_ds's audio playback has always been built on SDL2 and SDL2_mixer.

SDL2 is an excellent cross-platform multimedia library that handles low-level details like audio device abstraction, mixing, and buffering across Windows, macOS, and Linux. SDL2_mixer provides out-of-the-box support for MP3, WAV, OGG, FLAC, and other formats.

But here's the problem — SDL2_mixer's Opus support has never been stable enough.

Specifically:

Issue	Description
Windows	Official SDL2_mixer Windows binaries often omit Opus support or have incomplete compilation flags, causing `Mix_LoadMUS` to return NULL for `.opus` files
macOS	SDL2_mixer's Framework builds often have version mismatches with `libopusfile` and `libogg` dependencies, leading to runtime crashes
Linux	SDL2_mixer depends on system-installed `libopusfile-dev`, but package names and versions vary wildly across distributions
API Inconsistency	Even when Opus loads, SDL2_mixer's metadata extraction (duration, bitrate, etc.) often returns 0 or incorrect values

We tried various approaches:

Compiling our own Opus-enabled SDL2_mixer binaries → size ballooned, maintenance cost too high
Forcing users to install libopusfile-dev on Linux → poor user experience, and Windows/macOS couldn't be solved
Waiting for upstream SDL2_mixer fixes → waited through multiple versions, issues persist

Conclusion: SDL2_mixer's upstream support was insufficient for us to provide stable Opus playback.

So, before 4.1.0, we made a difficult decision — to temporarily not provide Opus support, to avoid giving users an unstable experience.

How Did 4.1.0 Solve It?

Since SDL2 wasn't working, we decided not to use SDL2 at all!

In 4.1.0, we completely bypass SDL2 and SDL2_mixer, building a dedicated playback pipeline for Opus:

User passes .opus file
       ↓
  Detects .opus extension
       ↓
  libopusfile decodes
       ↓
  Cross-platform audio output engine
   ├── Windows → winmm waveOut
   ├── Linux   → ALSA (libasound.so)
   └── macOS   → Core Audio AudioQueue
       ↓
  Audio output to speakers

This pipeline is completely independent of SDL2,不受 any limitations from SDL2_mixer.

Meanwhile, Windows users don't need to manually find DLLs — ap_ds will automatically download the required four DLLs (libopusfile-0.dll, libopus-0.dll, libogg-0.dll, libopusurl-0.dll) from https://dvsyun.top/ap_ds/download/ on first run, with SHA256 hash verification to ensure file integrity.

This is why 4.1.0 can support Opus, and why it couldn't before. It's not that we didn't want to — it's that we had to find a solution that was reliable, stable, and truly cross-platform. Now, we have.

🧬 4.1.0 Technical Deep Dive — What Did We Actually Do?

1. New Modules and Architecture Refactoring

To integrate Opus support without affecting mainline stability, we carefully designed the project structure:

File	Responsibility	Description
`opusplayer.py`	Opus playback core	Contains `OpusAudio` class, encapsulating all Opus playback, control, metadata, and batch processing APIs, completely independent of the SDL2 engine
`_opusdll.py`	Opus library loader	Handles cross-platform loading of `libopusfile` (Windows auto-download, Linux/macOS system detection + install guidance), provides unified `opusfile` handle
`player.py` (modified)	Opus routing integration	Auto-detects `.opus` files in `AudioLibrary` and forwards them to the `OpusAudio` sub-player, enabling completely transparent Opus playback
`__init__.py` (modified)	Exports OpusAudio	Users can directly `from ap_ds import OpusAudio` for standalone player, or use seamlessly via `AudioLibrary`

Key Design: AID 1:1 Mapping

When a user plays via AudioLibrary.play_from_file("song.opus"), the library internally creates an OpusAudio instance and generates a sub-AID, then uses a dictionary _aid_to_opus_aid to map the main library AID to the sub-library AID. This way, all subsequent controls (pause, resume, volume, seek) are correctly routed to the Opus engine, while user code remains completely unaware of this forwarding logic.

# User code — exactly the same as before!
from ap_ds import AudioLibrary

lib = AudioLibrary()
aid = lib.play_from_file("song.opus")  # Auto-routed to Opus engine
lib.pause_audio(aid)                   # Auto-routed to Opus engine
lib.seek_audio(aid, 30.0)              # Auto-routed to Opus engine
lib.stop_audio(aid)                    # Auto-routed to Opus engine

Completely transparent, zero learning curve.

2. Cross-Platform Playback Backends — Three Platform Engines Rewritten for Opus

Opus playback cannot depend on SDL2 (as detailed in the previous section), so we decided to bypass SDL2 entirely and use native OS audio APIs directly, paired with libopusfile decoding.

Platform	Playback Solution	Tech Stack
Windows	`winmm waveOut`	`libopusfile` decode → `waveOutWrite` multi-buffer output (4 buffers, 50ms/block)
Linux	ALSA	`libasound.so.2` → `snd_pcm_open` → `snd_pcm_writei` (direct PCM output)
macOS	Core Audio AudioQueue	Apple official C API → callback fills `AudioQueueBuffer` (consistent with official examples)

Platform Implementation Details:

Windows (_play_worker_windows):

Uses waveOutOpen to open the default audio device
Uses CreateEventW + WaitForSingleObject to synchronize buffer completion events
4 buffers rotating to eliminate stuttering
WAVEHDR struct uses c_void_p for 64-bit compatibility

Linux (_play_worker_linux):

Uses snd_pcm_open to open "default" device
Uses snd_pcm_set_params to set PCM parameters (S16_LE, interleaved mode)
Uses snd_pcm_writei to write PCM data
On -EPIPE (buffer underrun), calls snd_pcm_recover for automatic recovery

macOS (_play_worker_macos):

Uses AudioQueueNewOutput to create output queue
Uses AudioQueueAllocateBuffer to allocate buffers
Callback HandleOutputBuffer decodes Opus and fills mAudioData
Uses AudioQueueStart to start playback, AudioQueueStop to stop

This means that regardless of the user's operating system, Opus playback achieves native-level performance and stability.

3. Opus-Specific Error Code System (2001-2010)

To make Opus-related errors clearer and more traceable, we added 10 dedicated error codes:

Code	Constant	Meaning	Suggested Action
2001	`AP_DS_ERR_OPUS_LIB_LOAD_FAILED`	`libopusfile` load failed	Check if DLL exists or is locked
2002	`AP_DS_ERR_OPUS_DLL_DEPENDENCY`	DLL dependency missing	Ensure `libopus-0.dll` and `libogg-0.dll` exist
2003	`AP_DS_ERR_OPUS_OPEN_FAILED`	Opus file open failed	File may be corrupted or not a valid Opus stream
2004	`AP_DS_ERR_OPUS_HEADER_CORRUPT`	OpusHead header corrupted	Header info invalid or corrupted
2005	`AP_DS_ERR_OPUS_TAGS_PARSE_FAILED`	Tags parse failed	Tag data corrupted or invalid format
2006	`AP_DS_ERR_OPUS_DECODE_FAILED`	Opus decode failed	Audio data corrupted
2007	`AP_DS_ERR_OPUS_SEEK_FAILED`	Opus seek failed	Stream may not support seeking to that position
2008	`AP_DS_ERR_OPUS_BITRATE_UNAVAILABLE`	Bitrate unavailable	Unable to determine bitrate for this Opus stream
2009	`AP_DS_ERR_OPUS_NOT_SEEKABLE`	Stream not seekable	This Opus stream does not support seeking
2010	`AP_DS_ERR_OPUS_CHANNEL_INVALID`	Invalid channel count	Opus stream has invalid channel count

All Opus errors include:

Machine-readable error codes (for programmatic handling)
Human-readable error messages (for developer understanding)
Actionable suggestions (for user problem-solving)

4. Automatic DLL Download and Hash Verification (Windows)

Windows users don't need to manually find DLLs. When ap_ds first detects that Opus support is needed, it will:

Check if the required four DLLs exist in the package directory
If missing or hash verification fails, download from https://dvsyun.top/ap_ds/download/
After download, perform SHA256 hash verification to ensure file integrity
If hash mismatches, automatically retry

DLL File List and Hashes:

Filename	Size	SHA256
`libopusfile-0.dll`	55,884 bytes	`fc8ff75c5e0180e73b0528dc78c51ed0fb493741375cdc227f50c2a33cabf727`
`libopus-0.dll`	500,112 bytes	`90aa25a0a6525d7da48a7ae8dd3306e45b0c28ce09a73d2a02b56cd95418d5be`
`libogg-0.dll`	40,580 bytes	`3038ce8d161324a6349bf7c83b78493857ff6a3501e3adb3d541c6a07bd94a57`
`libopusurl-0.dll`	76,772 bytes	`a6cde968a23f2d0067332a13718c52e265653a2c35d65862e8dff4cf2a0346d9`

5. Cross-Platform Opus Library Loading Explained — Why Can't macOS Auto-Download Like SDL2?

Background: How Does SDL2's Auto-Download Work?

In ap_ds, the SDL2 loader (_sdl2.py) supports automatic download on Windows and macOS:

Windows: Downloads SDL2.dll and SDL2_mixer.dll from CDN
macOS: Downloads SDL2.dmg and SDL2_mixer.dmg from CDN, automatically mounts the DMG, extracts SDL2.framework and SDL2_mixer.framework to the package directory

This mechanism works well because SDL2 officially provides precompiled macOS Framework installers that can be directly downloaded, mounted, and extracted.

So Why Can't Opus Follow the Same Path?

Because Opus officially does not provide macOS precompiled Framework packages.

SDL2 has an official macOS download page with .dmg files containing complete .framework directory structures. But the Opus ecosystem (Xiph.Org Foundation) only provides source tarballs (.tar.gz), no precompiled macOS binaries whatsoever.

We searched carefully:

Xiph.Org official website → source only
Opus Codec official website → source only
libopusfile official website → source only
Major open-source mirror sites → source only
Even third-party maintained precompiled Frameworks — no reliable, verifiable source found

This is not an oversight on ap_ds's part; it's an objective gap in the Opus ecosystem.

What About macOS Users Then?

In 4.1.0, we designed a four-layer fallback mechanism for macOS users:

Layer 1: Detect system-installed Opus libraries
   ├── ctypes.util.find_library("opusfile")
   ├── /opt/homebrew/lib/libopusfile.dylib (Apple Silicon Homebrew)
   ├── /usr/local/lib/libopusfile.dylib (Intel Homebrew)
   └── /opt/local/lib/libopusfile.dylib (MacPorts)
         ↓ If not found
Layer 2: Attempt MacPorts auto-installation
   └── sudo port install opus opusfile libogg
         ↓ If MacPorts doesn't exist or install fails
Layer 3: Attempt Homebrew auto-installation
   └── brew install opus opusfile libogg
         ↓ If Homebrew doesn't exist or install fails
Layer 4: Show manual installation guide (with complete commands and steps)

In other words: ap_ds will do everything it can to automatically install Opus for the user, and only when all automatic options fail will it guide the user to manual operations.

What About Linux? — Consistent with SDL2

Linux Opus loading strategy is fully consistent with SDL2:

Layer 1: User config file (~/.config/ap_ds/opus_paths.conf)
Layer 2: System libraries (ctypes.util.find_library("opusfile"))
Layer 3: Auto-installation (apt-get / dnf / pacman, interactive sudo password)
Layer 4: Interactive setup (guide user to manually specify .so path)

Linux users will feel no surprises, as this flow is almost identical to _sdl2.py's Linux loading logic.

Why Do macOS Users See an "Apology" Rather Than a "Normal Prompt"?

Because macOS users are accustomed to SDL2's smooth "auto-download → extract → ready-to-use" experience. When they see that Opus requires extra installation steps, their first reaction may be "has ap_ds regressed?"

So in _macos_apology(), we candidly state:

"We sincerely apologize. On macOS, SDL2 libraries can be downloaded automatically, but we could NOT find any precompiled Opus framework packages for macOS. This is a limitation of the Opus ecosystem, not of ap_ds."

This is honesty, not excuse-making. We don't want users to mistakenly think we "got lazy" and didn't implement auto-download — we're telling the truth: Opus officially provides no macOS precompiled packages; we cannot download something that doesn't exist.

Technical Comparison Summary

Dimension	SDL2 (Windows/macOS)	Opus (Windows)	Opus (macOS)	Opus (Linux)
Official precompiled packages	✅ Yes (.dll / .dmg)	✅ Yes (.dll)	❌ None	❌ None (but system repos have them)
Auto-download solution	✅ Direct download	✅ Direct download	❌ Cannot download	❌ N/A (uses system libs)
Auto-installation solution	❌ N/A	❌ N/A	✅ MacPorts/Homebrew	✅ apt/dnf/pacman
User experience	Out-of-box	Out-of-box	Extra steps (auto + guided)	Extra steps (auto + guided)

What Have We Done to Compensate?

Prioritize auto-installation: On macOS, we try MacPorts first, then Homebrew, minimizing manual steps
Clear error messages: If auto-install fails, we provide complete, copy-paste-ready commands
Humble and honest: Directly state this is a limitation of the Opus ecosystem, not ap_ds's fault
SDL2-consistent Linux experience: Linux users see no difference

Quick Guide for macOS Users

If you're using ap_ds 4.1.0 to play Opus files on macOS, the easiest approach is:

# Using Homebrew (recommended)
brew install opus opusfile libogg

# Or using MacPorts
sudo port install opus opusfile libogg

After installation, ap_ds will automatically detect the system libraries without any extra configuration.

We understand this isn't as convenient as "download and use," but until Opus officially provides macOS precompiled packages, this is the best solution available.

6. OpusAudio Class — Complete API

The OpusAudio class provides an API almost identical to AudioLibrary, but entirely based on libopusfile and native audio output:

Method	Function
`play_from_file(file_path, loops=0, start_pos=0.0)`	Play Opus file
`play_from_memory(file_path, loops=0, start_pos=0.0)`	Play from cache
`new_aid(file_path)`	Preload Opus file
`play_audio(aid)`	Resume playback
`pause_audio(aid)`	Pause playback
`stop_audio(aid)`	Stop playback, return elapsed time
`seek_audio(aid, position)`	Seek to position (seconds)
`set_volume(aid, volume)`	Set volume (0-128)
`get_volume(aid)`	Get volume
`fadein_music(aid, loops=-1, ms=0)`	Fade in playback
`fadein_music_pos(aid, loops=-1, ms=0, position=0.0)`	Fade in from position
`fadeout_music(ms=0)`	Fade out and stop
`is_music_playing()`	Whether currently playing
`is_music_paused()`	Whether currently paused
`get_music_fading()`	Get fade in/out status
`get_audio_metadata(file_path)`	Get Opus metadata
`get_audio_duration(file_path)`	Get Opus duration
`get_audio_extended_metadata(file_path)`	Get extended tags (title/artist/album etc.)
`batch_get_metadata(file_paths, max_workers=None)`	Batch parse Opus metadata
`batch_get_duration(file_paths, max_workers=None)`	Batch get Opus durations
`cleanup_function()`	Release all resources

7. `_opusdll.py` — Cross-Platform Library Loader

Following the same design pattern as _sdl2.py, _opusdll.py provides:

Windows Loading Flow:

Check DLL files in package directory
If missing → auto-download
If exists but hash verification fails → re-download
Load into ctypes.CDLL and set function prototypes

Linux Loading Flow:

Check user config (~/.config/ap_ds/opus_paths.conf)
Check system libraries (ctypes.util.find_library("opusfile"))
Auto-installation (apt-get/dnf/pacman, interactive sudo password)
Interactive setup (guide user to manually specify .so path)

macOS Loading Flow:

Detect system libraries (find_library + common Homebrew/MacPorts paths)
Attempt MacPorts installation (sudo port install opus opusfile libogg)
Attempt Homebrew installation (brew install opus opusfile libogg)
Manual installation guidance

8. Library Size Changes — 4.1.0's "Temporary Weight Gain" and 4.2.0's "Slim-Down Return"

Due to the addition of opusplayer.py (~45KB) and _opusdll.py (~25KB), plus Windows DLLs (total ~673KB), this 4.1.0 RC release temporarily grows to 3.87 MB (4,059,251 bytes).

But — please note — this size increase is "one-time," existing ONLY in this single 4.1.0 release.

Why?

Because our strategy is very clear:

Version	Opus Support	Size	Description
4.0.x	❌	~2.5MB	Stable release, no Opus
4.1.0 RC	✅	~3.87MB	Only mainline with Opus, for feedback collection
4.2.0+ (mainline)	❌	~2.5MB	Remove Opus, return to lightweight
ap-ds-afs 1.0.0+	✅	~2.8-3.5MB	Opus's permanent home

That is to say:

4.1.0 RC's size increase is temporary, intentional, and one-release-only
4.2.0 mainline will completely remove Opus code and DLLs, naturally returning to 2.5MB
All users needing Opus should switch to the AFS branch (ap-ds-afs)

Our goal remains unchanged: keep ap_ds mainline extremely lightweight. 2.5MB is our brand promise, and we won't abandon it.

No need for 4.1.1 to "optimize size" — because 4.2.0 will simply cut Opus out directly, more thoroughly than any optimization.

⚠️ IMPORTANT: Windows Multiprocessing Warning

If you're using ap_ds 4.1.0 RC on Windows…

Please read this carefully. Your program's ability to run depends on it.

What's the problem?

On Windows, Python's multiprocessing module uses the spawn method to create new processes. This means each child process will re-import your main module.

If you call batch_get_metadata() or any other batch parsing function that uses ProcessPoolExecutor directly at the top level of your script, child processes will execute those calls again when re-importing the main module, causing infinite recursion and ultimately a BrokenProcessPool error.

Your program will crash. Directly.

Which APIs are affected?

All batch parsing functions that use ProcessPoolExecutor:

batch_get_metadata()
batch_get_duration()
batch_get_metadata_by_type()

The new Opus batch parsing in 4.1.0 RC is affected as well.

How to fix it?

It's simple — wrap your batch parsing code inside if __name__ == "__main__":.

❌ Wrong (Crashes on Windows):

from ap_ds import batch_get_metadata

# This will crash directly on Windows!
results = batch_get_metadata("/music/", max_workers=4)
print(f"Parsed {len(results)} files")

✅ Correct:

from ap_ds import batch_get_metadata

def main():
    results = batch_get_metadata("/music/", max_workers=4)
    print(f"Parsed {len(results)} files")

if __name__ == "__main__":
    main()

✅ Works with config functions too:

from ap_ds import batch_get_metadata

def load_config():
    return {"audio_dir": "/music/"}

def main():
    config = load_config()
    results = batch_get_metadata(config["audio_dir"], max_workers=4)
    print(f"Parsed {len(results)} files")

if __name__ == "__main__":
    main()

✅ Jupyter Notebook Users:

Put the batch call inside a function, then execute it in a cell:

def run_batch():
    from ap_ds import batch_get_metadata
    return batch_get_metadata("/music/", max_workers=4)

results = run_batch()

Why don't Linux and macOS have this problem?

Linux and macOS use fork by default to create child processes, which copy the parent process's memory space without re-executing the main module code.

However, we still recommend using if __name__ == "__main__" entry point protection on all platforms. It's good programming practice and ensures cross-platform compatibility.

Frequently Asked Questions

Q: I only called batch_get_metadata() once in my script. Why does it recurse?

A: Because on Windows, each child process re-imports your script. Without if __name__ == "__main__" protection, the import itself executes the top-level code again, causing infinite recursion.

Q: What should I set max_workers to?

A: We recommend os.cpu_count() or None (auto). On Windows, spawn process overhead is relatively high, so it's not recommended to exceed the number of CPU cores.

Q: Does this issue exist in 4.0.x as well?

A: Yes. All batch_* functions in all versions have the same issue on Windows. This is an inherent limitation of Python's multiprocessing on Windows, not a bug in ap_ds.

Q: Can ap_ds fix this automatically for me?

A: No. This is behavior at the Python runtime level that ap_ds cannot intervene in. We can only remind you in documentation and try to provide clear error messages in the code.

When using batch_get_metadata() or any batch parsing API on Windows, you MUST protect your entry point with if __name__ == "__main__":. Otherwise, your program will crash when child processes start.

🌿 AFS Branch: ap_ds's "Dual-Track" Future

Why "Split the Family"?

With Opus joining, the mainline version size grew from 2.5MB to 3.87MB. This sparked a deeper consideration:

In the future, we'll add more formats. Each new format increases size. If we cram all formats into mainline, ap_ds will eventually become a bloated monster,背离 the original "lightweight" vision.

So we decided: split the family.

Mainline 4.1.0's Special Status

4.1.0 is the ONLY mainline version that includes Opus support.

This is an intentional design decision:

4.1.0 RC: Opus's "swan song" in mainline — collect feedback, validate solutions
4.2.0+ (mainline): Remove Opus, return to lightweight positioning
ap-ds-afs 1.0.0+: Opus's permanent home, independent development

This gives users a clear choice:

Your Need	Recommended Version
Need Opus and willing to test	4.1.0 RC (mainline)
Need Opus and追求 long-term stability	ap-ds-afs 1.0.0 (AFS branch)
Don't need Opus,追求 extreme lightweight	4.2.0+ (mainline)

AFS Branch (All-Format Support)

AFS (All-Format Support) is a brand-new independent branch that will carry all new format support "beyond the 2.5MB limit."

Dimension	Mainline (ap-ds)	AFS Branch (ap-ds-afs)
PyPI package name	`ap-ds`	`ap-ds-afs`
Import name	`ap_ds`	`ap_ds` (identical!)
Version number	4.x (continuing)	1.0.0 (fresh start)
Opus support	❌ (except 4.1.0)	✅ Permanent
Core formats	MP3 / WAV / FLAC / OGG / AAC	Above + Opus + all future new formats
Size	~2.5MB	~2.8-3.5MB
Update strategy	Security fixes only	Syncs all mainline updates + own new formats
Target users	Minimalist developers	Developers needing special formats

Key Design: Import Name Identical

# Regardless of whether user installed ap-ds or ap-ds-afs
# The import method is exactly the same!
from ap_ds import AudioLibrary

This means users can switch seamlessly between the two packages by simply changing the package name in requirements.txt or pip install.

Which Should Users Choose?

Your Need	Recommendation
Only need MP3/WAV/FLAC/OGG/AAC	Mainline (ap-ds 4.2.0+) — lightweight, stable
Need Opus and willing to test RC	Mainline (ap-ds 4.1.0 RC) — early adopter, feedback
Need Opus and追求 long-term stability	AFS (ap-ds-afs 1.0.0) — full-featured, long-term support
Unsure if you'll need new formats later	Install mainline first, switch to AFS when needed (import name identical, painless switching)

Foolproof Design: Conflict Detection

If a user installs both ap-ds and ap-ds-afs simultaneously, ap_ds's __init__.py will detect the conflict and hard-exit:

# When both packages are installed, import triggers hard exit
>>> import ap_ds
Checking for package conflicts...
WARNING: Package conflict detected!
The following packages exist simultaneously:
  - ap-ds and ap-ds-afs
Please uninstall one of them.
# Python process exits directly 💀

This is not a bug — it's "foolproof design" — ensuring users don't encounter hard-to-debug import issues from package conflicts.

📖 Technical Manual Update

show_tech_manual() has been updated to 4.1.0, adding the following sections:

2.1 OPUS Support (New Section)

Opus format introduction (IETF standard, WebRTC, low-bitrate high quality)
Opus playback backend architecture (Windows waveOut / Linux ALSA / macOS Core Audio)
Opus auto-DLL download (Windows) and system library detection (Linux/macOS)
OpusAudio class complete API reference
AID 1:1 mapping mechanism explanation
Cross-platform Opus library loading deep dive (including macOS special notes)
4.1.0 version positioning notes (RC prerelease + Opus uniqueness statement)

10.4 Opus Error Codes (New Section)

Complete 10 Opus error codes list (2001-2010)
Meaning and suggested action for each error code
Error trigger scenario examples

Version History

Added 4.1.0 RC version entry
Recorded Opus support, cross-platform playback backends, AFS branch establishment
Marked 4.1.0 as RC prerelease
Clearly stated Opus uniqueness in mainline

🚀 Getting Started

Installation

# Mainline 4.1.0 RC (includes Opus, for testing feedback)
pip install ap-ds==4.1.0rc1

# Mainline 4.2.0+ (no Opus, extreme lightweight)
pip install ap-ds>=4.2.0

# AFS branch (includes Opus + all future new formats, permanent support)
pip install ap-ds-afs==1.0.0

💡 Why Python 3.15t / 3.14t?

Python's Free-Threading versions (filenames with t) remove the GIL (Global Interpreter Lock), enabling true multi-core parallelism. Combined with ap_ds's batch parsing, 120 MP3 files can be parsed in just 0.33 seconds.

⚠️ Version Selection Note: Python 3.15t is currently beta (b4) with potential unknown issues. For a more stable environment, we recommend Python 3.14t (stable). Both support GIL-free free-threading mode.

Windows Users

Python 3.14t (Stable) Downloads:

Architecture	Download Link
Windows 64-bit	python-3.14.4t-amd64.zip
Windows 32-bit	python-3.14.4t-win32.zip
ARM64	python-3.14.4t-arm64.zip

Python 3.15t (Beta) Downloads:

Architecture	Download Link
Windows 64-bit	python-3.15.0b4t-amd64.zip
Windows 32-bit	python-3.15.0b4t-win32.zip
ARM64	python-3.15.0b4t-arm64.zip

📦 ZIP — Extract and Use: Download and extract to any directory, add python.exe path to system PATH, and you're ready. No EXE installer needed — deployment in seconds.

Linux Users

Option 1: Use Package Manager

Fedora:

sudo dnf install python3.14-freethreading

After installation, interpreter is at /usr/bin/python3.14t.

Ubuntu/Debian (using deadsnakes PPA):

sudo add-apt-repository ppa:deadsnakes
sudo apt-get update
sudo apt-get install python3.14-nogil

This PPA provides the -nogil version, also a GIL-disabled build.

Option 2: Use Conda (Cross-Platform)

Install from conda-forge channel:

conda create -n nogil -c conda-forge python-freethreading
mamba create -n nogil -c conda-forge python-freethreading

Option 3: Build from Source (Universal)

# Download Python 3.14 source
wget https://www.python.org/ftp/python/3.14.0/Python-3.14.0.tgz
tar -xzf Python-3.14.0.tgz
cd Python-3.14.0

# Configure: --disable-gil is the key parameter
./configure --disable-gil

# Compile and install
make -j$(nproc)
sudo make install

macOS Users

Option 1: Official Installer (GUI)

Download the macOS installer from python.org
Run the installer, click the "Customize" button in the "Installation Type" screen
In the component list that appears, check the "Free-threaded Python" option, continue installation

Option 2: Using Homebrew

brew install python-freethreading

After installation, interpreter is at $(brew --prefix)/bin/python3.14t.

Verify Installation

Run the following commands to verify free-threading is working properly:

# Check version info (should include "free-threading build")
python3.14t --version

# Check GIL status (output False means GIL is disabled)
python3.14t -c "import sys; print(sys._is_gil_enabled())"

Create Virtual Environment

python3.14t -m venv my_env
source my_env/bin/activate  # Linux/macOS
my_env\Scripts\activate     # Windows

💡 Tip: Using python3.14t -m venv creates a GIL-free isolated environment.

Quick Start

from ap_ds import AudioLibrary

# Initialize library
lib = AudioLibrary()

# Play audio file
aid = lib.play_from_file("music/song.mp3")

# Control playback
lib.pause_audio(aid)      # Pause
lib.play_audio(aid)       # Resume
lib.seek_audio(aid, 30.5) # Seek to 30.5 seconds

# Stop and get elapsed time
duration = lib.stop_audio(aid)
print(f"Played {duration:.2f} seconds")

Playing Opus Files (via AudioLibrary, Recommended)

from ap_ds import AudioLibrary

lib = AudioLibrary()
aid = lib.play_from_file("song.opus")  # Auto-routed to Opus engine

# All control methods are identical
lib.pause_audio(aid)
lib.seek_audio(aid, 30.0)
lib.set_volume(aid, 80)
lib.stop_audio(aid)

Playing Opus Files (Directly Using OpusAudio)

from ap_ds import OpusAudio

opus = OpusAudio()
aid = opus.play_from_file("song.opus")
opus.pause_audio(aid)
opus.seek_audio(aid, 30.0)
opus.stop_audio(aid)

Getting Opus Metadata

from ap_ds import get_audio_metadata

meta = get_audio_metadata("song.opus")
print(meta["duration"])       # 265
print(meta["sample_rate"])    # 48000
print(meta["channels"])       # 2
print(meta["bitrate"])        # 105351

Batch Parsing

from ap_ds import batch_get_metadata

# Batch parse entire folder (120 MP3s in just 0.33 seconds!)
results = batch_get_metadata("/music/playlist/", max_workers=8)

for meta in results:
    print(f"{meta['path']}: {meta['duration']}s, {meta['bitrate']}bps")

Batch Parsing API Overview:

API	Description
`batch_get_metadata()`	Batch parse, returns full metadata list
`batch_get_duration()`	Batch get durations, returns `{path: duration}`
`batch_get_metadata_by_type()`	Batch parse filtered by format

⚠️ CRITICAL: Windows Batch Parsing and BrokenProcessPool

If you're using batch parsing APIs on Windows, you MUST protect your entry point with if __name__ == "__main__".

Why?

On Windows, ProcessPoolExecutor uses spawn to create new processes. This means each subprocess will re-import your main module. Without entry point protection, this creates an infinite recursion loop, crashing your program with a BrokenProcessPool error.

This is not optional. This is mandatory.

❌ Wrong (Will Crash on Windows):

from ap_ds import batch_get_metadata

# This will crash with BrokenProcessPool on Windows!
results = batch_get_metadata("/music/", max_workers=4)

✅ Correct (Always Works):

from ap_ds import batch_get_metadata

def main():
    results = batch_get_metadata("/music/", max_workers=4)
    print(f"Parsed {len(results)} files")

if __name__ == "__main__":
    main()

✅ Correct (Script with Configuration):

from ap_ds import batch_get_metadata

def get_config():
    # ... config logic ...
    return config

def main():
    config = get_config()
    results = batch_get_metadata(config["audio_dir"], max_workers=4)
    print(f"Parsed {len(results)} files")

if __name__ == "__main__":
    main()

This applies to:

✅ Any script that imports ap_ds and uses batch parsing on Windows
✅ Jupyter notebooks (if running on Windows, wrap batch calls in a function with if __name__ == "__main__")
✅ Any test scripts (e.g., CI-CD-TEST.py)

What about Linux/macOS?

Not required. Linux and macOS use fork by default and don't have this issue. However, for cross-platform compatibility, using entry point protection is still good practice.

DAP Playlist System

Audio files are automatically recorded to DAP (Dvs Audio Playlist) when played:

# Files are recorded automatically
aid1 = lib.play_from_file("song1.mp3")
aid2 = lib.play_from_file("song2.ogg")

# Get all recordings
recordings = lib.get_dap_recordings()
print(f"Recorded {len(recordings)} files")

# Save as JSON
success = lib.save_dap_to_json("my_playlist.ap-ds-dap")

DAP stores only metadata (path, duration, bitrate, channels), not audio data. Each record takes approximately 150 bytes of memory.

Platform Support

Windows

Automatically downloads SDL2.dll and SDL2_mixer.dll with hash verification
Opus support auto-downloads four DLLs (libopusfile-0.dll, libopus-0.dll, libogg-0.dll, libopusurl-0.dll)
No manual configuration needed
Supports Windows 7 and above

macOS

Automatically downloads SDL2.framework and SDL2_mixer.framework with hash verification
Opus support via Homebrew or MacPorts system library installation (auto-detection + auto-install attempt)
Supports macOS 10.9 and above

Linux

Intelligent Multi-Layer Import System:

System library check: Uses system-installed SDL2 libraries
User configuration: Checks paths saved from previous runs
Auto-installation: Detects package manager and installs required packages
Interactive guidance: Provides manual options if all above fail

Opus Package Manager Support:

# Ubuntu/Debian
sudo apt-get install libopusfile-dev libopus-dev libogg-dev

# Fedora
sudo dnf install opusfile-devel opus-devel libogg-devel

# Arch
sudo pacman -S opusfile opus libogg

SDL2 Package Manager Support:

# Ubuntu/Debian
sudo apt-get install libsdl2-dev libsdl2-mixer-dev

# Fedora
sudo dnf install SDL2-devel SDL2_mixer-devel

# Arch
sudo pacman -S sdl2 sdl2_mixer

Embedded ARM64

Tested on the following platforms:

Orange Pi 4 Pro (Allwinner A733, 2xA76 + 6xA55 @ 2.0GHz)
Raspberry Pi 5 (BCM2712, 4xA76 @ 2.4GHz)

Both running Ubuntu 22.04 with full audio functionality via 3.5mm output. Memory growth ~4MB after extensive testing.

🧪 CI/CD Testing — Comprehensive Coverage, All Passed

Opus Test Suite (OPUS_TEST.py)

4.1.0 adds a complete Opus test suite covering 16 test categories:

Test Category	Test Items	Description
OP-1	Module import/constants/error codes	Verify all Opus modules, constants, error codes correctly exported
OP-2	DLL loading and auto-download	Verify `_opusdll.py` loading, DLL file existence, hash verification
OP-3	Metadata parsing	Verify Opus duration, sample rate, channels, bitrate, extended tags
OP-4	Playback functionality	Verify `play_from_file`, `play_from_memory`, `new_aid`
OP-5	Playback control	Verify `pause_audio`, `play_audio`, `stop_audio`
OP-6	Volume control	Verify `set_volume` (0-128 boundary values) and `get_volume`
OP-7	Seeking functionality	Verify `seek_audio` for various positions and boundary values
OP-8	Fade in/out	Verify `fadein_music`, `fadein_music_pos`, `fadeout_music`
OP-9	Opus vs native format distinction	Verify `_is_opus_file` and AudioLibrary auto-routing
OP-10	Batch parsing	Verify `batch_get_metadata`, `batch_get_duration`, `batch_by_type`
OP-11	Error code triggering	Verify all 10 Opus error codes trigger correctly
OP-12	AID 1:1 mapping	Verify main AID ↔ Opus sub-AID full lifecycle
OP-13	Resource management	Verify `cleanup_function` correctly releases resources
OP-14	Boundary values and error testing	Verify invalid parameter types, out-of-range values handled correctly
OP-15	DLL-specific testing	Verify DLL file sizes, hashes, load idempotency
OP-16	Error code trigger specific testing	Verify each Opus error code triggers in real scenarios

Test Results

==================================================================
 Opus Test Summary
==================================================================
   Passed : 229
   Failed : 0
   Skipped: 0
==================================================================

All 229 tests passed, 0 failed, 0 skipped.

CI/CD Integration

OPUS_TEST.py is integrated into the CI/CD pipeline, automatically running before each release. Test coverage includes:

Automated tests (no human intervention): OP-1 ~ OP-16
Listening tests (requires human confirmation): OP-L (5 listening tests)

Listening tests include:

Normal playback is audible
Fade-in effect is smooth
Fade-out effect is smooth
Volume 0 is silent
Seek position is correct

Running Tests

# Full test (including listening tests)
python OPUS_TEST.py

# Automated tests only
python OPUS_TEST.py --auto

# Listening tests only
python OPUS_TEST.py --listen

🧪 CI/CD Comprehensive Testing — All Passed (Green)

Below is the complete output summary of the full CICD test suite (CI,CD_TEST.py) run on a Windows machine with Python 3.13.4. It covers all public APIs, error handling, edge cases, boundary values, and interactive listening tests — 650+ tests passed, 0 failed, 0 skipped. The same tests have been verified on macOS and Ubuntu with identical results.

==================================================================
 Opus Test Summary
==================================================================
   Passed : 229
   Failed : 0
   Skipped: 0
==================================================================

==================================================================
 CICD Test Summary
==================================================================
   Passed : 650+
   Failed : 0
   Skipped: 0
==================================================================

You can run the test suite yourself by executing CI,CD_TEST.py (included in the package):

python CI,CD_TEST.py --auto      # Automated tests only
python CI,CD_TEST.py --listen    # Interactive listening tests only
python CI,CD_TEST.py --full      # Everything (default)

🧪 API Import Verification Test

We provide a comprehensive test script IMPORT_TEST.py to verify that all documented APIs exist and are correctly exported.

Test Results

============================================================
🧪 Testing All ap_ds API Exports
============================================================

📦 Testing top-level imports:
----------------------------------------
  ✅ AudioLibrary
  ✅ OpusAudio
  ✅ batch_get_metadata
  ✅ batch_get_duration
  ✅ batch_get_metadata_by_type
  ✅ get_audio_duration
  ✅ get_audio_metadata
  ✅ auto_check_runtime
  ✅ check_runtime_mode
  ✅ show_tech_manual

----------------------------------------
🎯 Testing AudioLibrary methods:
----------------------------------------
  ✅ AudioLibrary.__init__
  ✅ AudioLibrary.play_from_file
  ✅ AudioLibrary.play_from_memory
  ✅ AudioLibrary.new_aid
  ✅ AudioLibrary.play_audio
  ✅ AudioLibrary.pause_audio
  ✅ AudioLibrary.stop_audio
  ✅ AudioLibrary.seek_audio
  ✅ AudioLibrary.set_volume
  ✅ AudioLibrary.get_volume
  ✅ AudioLibrary.fadein_music
  ✅ AudioLibrary.fadein_music_pos
  ✅ AudioLibrary.fadeout_music
  ✅ AudioLibrary.is_music_playing
  ✅ AudioLibrary.is_music_paused
  ✅ AudioLibrary.get_music_fading
  ✅ AudioLibrary.get_audio_duration
  ✅ AudioLibrary.get_audio_metadata
  ✅ AudioLibrary.get_audio_metadata_by_path
  ✅ AudioLibrary.get_audio_metadata_by_aid
  ✅ AudioLibrary.batch_get_metadata
  ✅ AudioLibrary.batch_get_duration
  ✅ AudioLibrary.batch_get_metadata_by_type
  ✅ AudioLibrary.save_dap_to_json
  ✅ AudioLibrary.get_dap_recordings
  ✅ AudioLibrary.clear_dap_recordings
  ✅ AudioLibrary.clear_memory_cache
  ✅ AudioLibrary.cleanup_function
  ✅ AudioLibrary._find_channel_by_aid
  ✅ AudioLibrary._get_file_path_by_aid
  ✅ AudioLibrary._is_music_file
  ✅ AudioLibrary._seek_audio
  ✅ AudioLibrary._get_duration_by_filepath
  ✅ AudioLibrary._get_file_duration

----------------------------------------
🔍 Checking extra APIs:
----------------------------------------
  ✅ is_full_performance
  ✅ get_runtime_info

============================================================
📊 FINAL SUMMARY
============================================================
🎉 ALL APIs EXIST! Documentation is accurate.
============================================================
✅ Passed: 47
❌ Failed: 0

All 47 API tests passed!

Category	Count	Status
Top-level imports	10	✅ All passed
AudioLibrary methods	34	✅ All passed
Extra APIs	3	✅ All passed
Total	47	✅ 47/47 passed

📊 Version Comparison Overview

Dimension	4.0.1	4.1.0 RC	4.2.0+ (planned)	ap-ds-afs 1.0.0
Opus support	❌	✅	❌	✅
Playback formats	MP3/WAV/FLAC/OGG/AAC	+Opus	MP3/WAV/FLAC/OGG/AAC	+Opus + future formats
Cross-platform playback engine	SDL2 only	SDL2 + Opus native	SDL2 only	SDL2 + Opus native
Opus error codes	❌	✅ 2001-2010	❌	✅ 2001-2010
Auto-DLL download	SDL2	SDL2 + Opus DLL	SDL2	SDL2 + Opus DLL
AFS branch	❌	✅	❌	✅ (AFS itself)
Test coverage	421 tests	650+ tests	421 tests	650+ tests
Library size	~2.5MB	3.87MB	~2.5MB	~2.8-3.5MB
Version status	Stable	RC prerelease	Planned	Planned

📦 Version Relationships

Version	Type	Support Period	Use Case
v3.0.0 LTS	Long-Term Support	Until March 2031	Production environments
v3.1.x	LFV (Latest Feature Version)	~6 months	Early adopters (superseded)
v4.0.x	LFV (Latest Feature Version)	~6 months	Early adopters, new features
v4.1.0 RC	RC prerelease	~2 months	Opus testing and feedback collection
v4.2.0+ (mainline)	LFV (Latest Feature Version)	~6 months	Remove Opus, return to lightweight
ap-ds-afs 1.0.0+	Long-Term Support	TBD	Permanent home for Opus and new formats

📌 Version Upgrade Recommendations

User Type	Recommendation
Production environments	Continue using v3.0.0 LTS, wait for v4.0.0 LTS
Development/testing	Upgrade to v4.1.0 RC to try Opus, help test feedback
Need Opus and willing to test	Use 4.1.0 RC, report bugs
Need Opus and追求 long-term stability	Wait for ap-ds-afs 1.0.0
Don't need Opus,追求 extreme lightweight	Wait for 4.2.0+ or continue with 4.0.x
Previously affected by v3.1.x metadata bug	Must upgrade to v4.0.0+

# 4.1.0 RC installation
pip install ap-ds==4.1.0rc1

# AFS branch (after release)
pip install ap-ds-afs==1.0.0

🌐 apds.top Is Now Live!

The ap_ds official project homepage is now live with TLS encryption!

🎉 Visit: https://apds.top

Website Features

📄 Full documentation: API reference, user guide, FAQ
📦 Version distribution: All version download links and changelogs
🔗 Repository navigation: GitCode (primary), Gitee (China mirror)
✉️ Feedback system: Users can submit feedback directly via the website
🔒 Full-site TLS encryption: All pages served via HTTPS

⚠️ Repository Migration Notice

GitHub Deprecated, GitLab Abandoned, Migrated to GitCode — apds.top Is the Permanent Home

1. Why GitHub Was Deprecated

The developer's GitHub account was locked due to loss of two-factor authentication (2FA) device. After multiple attempts to contact GitHub support, only automated bot responses were received. Due to the complete lack of human assistance, the developer decided to permanently abandon that GitHub account and will not create a new one in the foreseeable future. The old dvs-web/ap_ds repository is now officially deprecated and no longer receives any updates.

2. Why GitLab (JiHu) Was Abandoned

After the GitHub issues, the project migrated the primary repository to Gitee and GitLab (JiHu). However, due to platform policy changes where basic account login became a paid feature, GitLab was recently abandoned. Since the project relies on free open-source collaboration, this change created an unacceptable barrier for contributors and users. The developer attempted to find alternatives but found none within the free tier. Therefore, the GitLab repository is no longer actively maintained.

3. Why Gitee Is Now a Backup (Not Primary)

Gitee is an excellent platform, particularly for Chinese developers, offering fast and stable access. It remains a strongly recommended choice. However, its role has been adjusted to a backup or China-facing mirror for two main reasons:

International accessibility: Gitee's servers are primarily located in China. For developers outside mainland China, access can be slow, unstable, and in some cases, completely blocked due to international network policies. This creates a poor experience for a large portion of the user base.
UI and workflow: While feature-rich, Gitee's UI and workflows are often considered outdated and not as aligned with modern Git workflows that many international developers are accustomed to.

For these reasons, while Gitee is by no means "bad" and will continue to be fully supported as a China-facing mirror, it is no longer suitable as the sole primary repository for a globally-oriented project.

4. Solution: GitCode Becomes the New Primary Repository

After surveying the landscape of free Git hosting platforms, GitCode emerged as the ideal solution. GitCode offers a modern interface, strong feature set, and, most importantly, excellent accessibility for developers both within China and abroad. It has become the new official primary repository for the ap_ds project.

New primary repository located at: https://gitcode.com/dvsxt/ap_ds

5. Permanent Home: apds.top

apds.top is now officially live and fully operational!

This is not just another repository mirror — it is the permanent official home of ap_ds. It solves all platform dependency issues:

✅ Full independent control: No longer affected by third-party platform policy changes
✅ TLS encryption: Full-site HTTPS secure access
✅ Permanent stability: Even if all third-party platforms fail, apds.top remains available
✅ One-stop service: Documentation, downloads, feedback, and repository navigation all integrated

Official Project Homepage: https://apds.top

🔄 NEW: GitHub Repository (Compatibility Mirror)

A Word from the Developer

Recently, US-based developer Clint Shepherd raised a legitimate concern:

"Why isn't there a GitHub repository? GitCode is in China, I'm not used to it. apds.top doesn't have issues, pull requests, forks, or other collaboration features..."

After careful consideration, we realized he was absolutely right.

While we have no intention of making GitHub the primary platform — their customer support was terrible when our account was locked, and we're still bitter about it — the reality is undeniable:

GitHub has a massive user base
The ecosystem is very mature
Many developers are simply more comfortable with it
Collaboration features (issues, PRs, forks) are standard expectations

So we made a pragmatic decision: We created a new GitHub account (dvs-dvsxt) and set up a compatibility mirror at:

🔗 https://github.com/dvs-dvsxt/ap_ds

Important: This Is a Compatibility Mirror, Not the Primary Repository

Aspect	Details
Purpose	Compatibility mirror for GitHub developers
Permanent Home	apds.top — permanent official source
Primary Mirror	GitCode — global access
China Mirror	Gitee — for Chinese users
GitHub Status	⚠️ Compatibility mirror — updated periodically, may lag
Best For	✅ GitHub users who prefer familiar workflows to star, watch, or clone

What This Means for You

If you're a GitHub user: You can now clone, fork, and file issues on GitHub. We will respond to GitHub issues, but please be patient — response times may be slower than on GitCode or apds.top.
If you're a developer in China: Gitee remains your fastest and most stable choice. Use it with confidence.
If you want the latest updates: Always check apds.top first. It has the latest releases, changelogs, and announcements.
If you want to contribute: We welcome contributions on any platform — GitCode, Gitee, or GitHub. PRs are reviewed regardless of origin.

Final Repository Strategy (Updated)

Platform	Status	Use
apds.top	✅ Permanent home	Official source, documentation, downloads
GitCode	✅ Primary mirror	Code hosting for global users
GitHub	✅ Compatibility mirror	For GitHub developers (NEW!)
Gitee	✅ China mirror	Fast access for Chinese users
GitLab (JiHu)	❌ Abandoned	No longer maintained

Special thanks to Clint Shepherd for asking the hard questions and pushing us to make ap_ds more accessible to the global Python community. Your feedback made this better. 🙏

Overview

ap_ds is a lightweight (2.5MB) Python audio library for playing and high-precision metadata parsing of MP3, FLAC, OGG, and WAV files. It has no external Python dependencies, uses only the Python standard library, and provides non-blocking playback suitable for GUI applications.

Core Features:

Extremely lightweight: 2.5MB on Windows / 3.36MB full solution on macOS
Zero Python dependencies: Standard library only
High-precision metadata: WAV/FLAC 100%, OGG 99.99%, MP3 >98%
Batch parsing: Process hundreds of files in parallel using batch_get_metadata()
Non-blocking playback: Ideal for GUI applications
Cross-platform: Windows, macOS, Linux, embedded ARM64
DAP recording system: Metadata-only automatic playback history
Python 3.15t support: GIL-free true parallelism, full multi-core performance
LTS support: First long-term support version with 5-year maintenance commitment

Contact and Support

📧 Licensing Inquiries

me@dvsyun.top or dvs6666@163.com · Response within 7 business days

🛠️ Technical Support

apds.top Issues · GitCode Issues · GitHub Issues · Gitee Issues · Email (completely free)

Official Repositories and Project Sources

✅ Official Primary Repository (Preferred): https://apds.top — ap_ds's permanent official home, fully independent control, TLS encrypted,不受 third-party platform policy changes.

✅ Primary Mirror (Global Access): GitCode — Globally accessible, modern UI, actively maintained as a public mirror.

✅ Compatibility Mirror (GitHub Users): GitHub — For GitHub developers. May lag behind primary sources. Issues and PRs welcome, but response times may be slower.

✅ China Mirror (Fast and Stable): Gitee — Full mirror for Chinese developers, fast access, classic stable UI.

❌ Deprecated and Abandoned:

GitHub (dvs-web/ap_ds) — Deprecated due to permanent account lockout, no longer maintained.
GitLab (JiHu) — Abandoned due to platform policy changes (basic account login became a paid feature), no longer maintained.

ℹ️ ap_ds v3.0.0 LTS – First Long-Term Support Version This release consolidates all previous improvements, adds deterministic resource cleanup, hash-verified downloads, and a 5-year support commitment. The old GitHub repository (dvs-web/ap_ds) is deprecated and no longer updated. The GitLab repository has been abandoned. For future updates and contributions, please use the official primary repository apds.top or the public mirrors GitCode, GitHub (dvs-dvsxt/ap_ds), and Gitee.

Developer Personal Homepage and Blog: https://dvsx.top — Blog is currently under maintenance and upgrades. Stay tuned. ap_ds Project Homepage: https://apds.top (Official documentation, releases, and licensing center).

🔗 Canonical URL: https://apds.top/ 📖 Blog and Author: dvsx.top — Blog is currently under maintenance and upgrades. Stay tuned.

About the Author

Developer: Dvs (DvsXT) Personal Homepage and Blog: https://dvsx.top — Blog is currently under maintenance and upgrades. Stay tuned. Author Bio: https://dvsyun.top/me/dvs Email: me@dvsyun.top · dvs6666@163.com

ap_ds Official Portal

🎵 ap_ds Official Website (Primary): https://apds.top — Permanent official home, full documentation, version releases, licensing center 📦 PyPI Project Page: https://pypi.org/project/ap_ds/ — Installable via pip 🌐 Mirror Documentation Site: https://www.dvsyun.top/ap_ds — Backup documentation access

👉 The ap_ds project homepage (apds.top) is the preferred recommendation for official sources, hosting full documentation, license details, version changelogs, and official releases. The author's personal blog (dvsx.top) is currently under maintenance and upgrades. Stay tuned.

ap_ds Audio Library — Complete API Reference

Version: v4.1.0 RC Document Date: August 2026 Project Homepage: https://apds.top

AudioLibrary Class — Complete API
- Initialization
- Playback Methods
- Control Methods
- Volume Methods
- Fade and Transition Methods
- Metadata Methods
- Batch Parsing Methods
- DAP System Methods
- Resource Management
- Internal Helper Methods
Top-Level Convenience Functions
AudioParser Module — Metadata API
SDL2 Integration Layer
Constants Reference
Unified Error Handling and Error Code Reference
Environment Variables Reference
Technical Manual (show_tech_manual())

AudioLibrary Class — Complete API

The AudioLibrary class is the main interface for audio playback, control, and metadata management. Every method that can fail returns a unified error tuple:

(code: int, message: str, suggestion: str)

On success, methods return their documented success value. On failure, they never raise exceptions (__init__ excepted, as it is a constructor) — they return the error tuple above. See the Unified Error Handling and Error Code Reference section for details.

Initialization

`init(frequency: int = 44100, format: int = MIX_DEFAULT_FORMAT, channels: int = 2, chunksize: int = 2048) -> None`

Description:

Initializes the SDL2 audio subsystem and SDL2_mixer library. Must be called before any playback operations. Sets up the audio device with the specified parameters and registers an exit handler for automatic resource cleanup.

Note: __init__ is a constructor — it cannot return error tuples. If SDL2 or mixer initialization fails, it raises RuntimeError (the only intentional exception in the library).

Parameters:

Parameter	Type	Default	Description
`frequency`	`int`	`44100`	Audio sample rate (Hz). Common values: 44100 (CD quality), 48000 (DVD/video), 22050 (voice).
`format`	`int`	`MIX_DEFAULT_FORMAT`	Audio sample format. Typically `AUDIO_S16SYS` (16-bit signed, system endianness).
`channels`	`int`	`2`	Number of audio channels. `1` = mono, `2` = stereo.
`chunksize`	`int`	`2048`	Buffer size (in samples). Larger values reduce CPU but increase latency.

Raises (constructor only):

RuntimeError: If SDL2 initialization fails (e.g., no audio device available).
RuntimeError: If mixer initialization fails (e.g., unsupported format).

Examples:

from ap_ds import AudioLibrary

lib = AudioLibrary()                              # Default (CD quality, stereo)
lib_voice = AudioLibrary(frequency=22050, channels=1, chunksize=1024)

Playback Methods

`play_from_file(file_path: str, loops: int = 0, start_pos: float = 0.0) -> Union[int, Tuple[int, str, str]]`

Description:

Loads and plays an audio file directly from disk. .ap-ds-dap files are DAP export records (output format) and are not supported as playback input — passing one returns (1003, msg, suggestion).

Parameters:

Parameter	Type	Default	Description
`file_path`	`str` / `bytes` / `os.PathLike`	Required	Full path to the audio file. Supported formats: MP3, WAV, FLAC, OGG, Opus (4.1.0 RC).
`loops`	`int`	`0`	Number of loops after the first playback. `0` = once, `-1` = infinite, `>0` = count.
`start_pos`	`float`	`0.0`	Start position in seconds. Not supported for sound effects.

Returns:

Success: int — Unique Audio ID (AID) identifying this playback instance.
Failure: Tuple[int, str, str] — (error_code, error_message, suggestion).

Error Codes:

Code	Condition
`1001`	`file_path` does not exist, or `file_path` type is invalid (`None`, `int`, `list`, `dict`, `tuple`, etc.).
`1999`	`loops` type is invalid (must be `int`).
`1003`	Audio loading failed (e.g., `.ap-ds-dap` file, corrupted file, or unsupported format).
`1004`	Playback failed (e.g., no available channel or audio device issue).
`2003`	Opus file open failed (Opus-specific).

Behavior by File Type:

File Type	Mode	Seek Support	Fade Support
MP3, OGG, FLAC	Music (Mix_PlayMusic)	Yes	Yes
Opus (4.1.0 RC)	OpusAudio (libopusfile + native API)	Yes	Yes
WAV (duration ≥ threshold)	Music (Mix_PlayMusic)	Yes	Yes
WAV (duration < threshold)	Sound effect (Mix_PlayChannel)	No	No
Other formats	Sound effect (Mix_PlayChannel)	No	No

Examples:

aid = lib.play_from_file("song.mp3")                  # Play once
aid = lib.play_from_file("beep.wav", loops=5)         # Loop 5 times
aid = lib.play_from_file("podcast.mp3", start_pos=30.0)
aid = lib.play_from_file("ambient.ogg", loops=-1)     # Infinite loop
aid = lib.play_from_file("song.opus")                 # Opus auto-routed (4.1.0 RC)

# Invalid input returns error tuple (no exception):
result = lib.play_from_file(None)          # (1001, 'Invalid file path type: NoneType', ...)
result = lib.play_from_file("x.mp3", loops="2")  # (1999, 'Invalid loops type: str...', ...)

`play_from_memory(file_path: str, loops: int = 0, start_pos: float = 0.0) -> Union[int, Tuple[int, str, str]]`

Description:

Plays an audio file that has been preloaded into memory via new_aid(). Faster than play_from_file() because the file is already cached.

Parameters: Same as play_from_file().

Returns:

Success: int (AID).
Failure: Tuple[int, str, str] — (error_code, error_message, suggestion).

Error Codes:

Code	Condition
`1013`	`file_path` not loaded into memory, or `file_path` type invalid (`None`, `list`, `dict`, etc.).
`1999`	`loops` type invalid (must be `int`).
`1004`	Playback from cache failed.

Example:

lib.new_aid("gunshot.wav")
lib.new_aid("explosion.wav")
aid = lib.play_from_memory("gunshot.wav")   # Instant, no disk I/O

`new_aid(file_path: str) -> Union[int, Tuple[int, str, str]]`

Description:

Preloads an audio file into memory without playing it. Useful for caching sounds or tracks that will be played multiple times.

Parameters:

Parameter	Type	Description
`file_path`	`str` / `bytes` / `os.PathLike`	Path to the audio file to cache.

Returns:

Success: int (AID).
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1001`	`file_path` does not exist or type is invalid.
`1003`	Audio loading failed.

Example:

sounds = {
    'hit':  lib.new_aid("hit.wav"),
    'jump': lib.new_aid("jump.wav"),
    'coin': lib.new_aid("coin.wav"),
}
lib.play_from_memory(sounds['hit'])

Control Methods

`play_audio(aid: int) -> Tuple[int, str, str]`

Description:

Resumes a paused audio instance.

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.

Example:

lib.pause_audio(aid)
lib.play_audio(aid)      # Resume
result = lib.play_audio(99999)   # (1002, 'Invalid AID: 99999', ...)

`pause_audio(aid: int) -> Tuple[int, str, str]`

Description:

Pauses an audio instance. Can be resumed with play_audio().

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.

`stop_audio(aid: int) -> Union[float, Tuple[int, str, str]]`

Description:

Stops playback and returns the elapsed playback time in seconds.

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.

Returns:

Success: float — Elapsed playback time in seconds.
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.

Example:

played = lib.stop_audio(aid)          # e.g., 3.42
result = lib.stop_audio(99999)        # (1002, 'Invalid AID: 99999', ...)

`seek_audio(aid: int, position: float) -> Tuple[int, str, str]`

Description:

Seeks to the specified position in seconds. Only supported for music-mode files (MP3, OGG, FLAC, long WAV, Opus).

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.
`position`	`int` / `float`	Position in seconds.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.
`1999`	`position` type is invalid.
`2007`	Opus seek failed (Opus-specific).
`2009`	Opus stream not seekable (Opus-specific).

Example:

lib.seek_audio(aid, 30.5)
result = lib.seek_audio(aid, None)    # (1999, 'Invalid position type: NoneType...', ...)

Volume Methods

`set_volume(aid: int, volume: int) -> Tuple[int, str, str]`

Description:

Sets the volume. Range 0–128.

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.
`volume`	`int`	Volume value 0–128.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.
`1015`	`volume` is not `int`, or out of 0–128 range.
`1004`	Applying volume failed.

Example:

lib.set_volume(aid, 64)
result = lib.set_volume(aid, "loud")   # (1015, 'Invalid volume type: str...', ...)
result = lib.set_volume(aid, 200)      # (1015, 'Invalid volume: 200 (must be 0-128)', ...)

`get_volume(aid: int) -> Union[int, Tuple[int, str, str]]`

Description:

Returns the current volume (0–128).

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.

Returns:

Success: int — Current volume (0–128).
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid.

Fade and Transition Methods

`fadein_music(aid: int, loops: int = -1, ms: int = 0) -> Tuple[int, str, str]`

Description:

Fades music in from silence to full volume over ms milliseconds.

Parameters:

Parameter	Type	Default	Description
`aid`	`int`	—	AID of the music file.
`loops`	`int`	`-1`	`-1` = infinite, `0` = once, `>0` = count.
`ms`	`int`	`0`	Fade duration in milliseconds.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid or not a music file.
`1999`	`ms` / `loops` type is invalid.
`1003`	Music loading failed.
`1004`	SDL_mixer fade failed.

`fadein_music_pos(aid: int, loops: int = -1, ms: int = 0, position: float = 0.0) -> Tuple[int, str, str]`

Description:

Fades music in from the specified position.

Parameters:

Parameter	Type	Default	Description
`aid`	`int`	—	AID of the music file.
`loops`	`int`	`-1`	Loop count.
`ms`	`int`	`0`	Fade duration in milliseconds.
`position`	`int` / `float`	`0.0`	Start position in seconds.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1002`	`aid` is invalid or not a music file.
`1999`	`ms` / `loops` / `position` type is invalid.
`1012`	`Mix_FadeInMusicPos` not supported by current SDL_mixer.
`1003`	Music loading failed.
`1004`	SDL_mixer fade failed.

`fadeout_music(ms: int = 0) -> Tuple[int, str, str]`

Description:

Fades out the currently playing music over ms milliseconds.

Parameters:

Parameter	Type	Default	Description
`ms`	`int`	`0`	Fade duration in milliseconds.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1004`	No music is playing, or fade failed.

`is_music_playing() -> bool`

Description:

Returns whether music is currently playing.

Returns: bool — True if playing, False otherwise.

`is_music_paused() -> bool`

Description:

Returns whether music is currently paused.

Returns: bool — True if paused, False otherwise.

`get_music_fading() -> int`

Description:

Returns the current fade status.

Returns: int:

0 (MUS_NO_FADING): No fade in progress
1 (MUS_FADING_IN): Fading in
2 (MUS_FADING_OUT): Fading out

Metadata Methods

`get_audio_duration(source: Union[str, int], is_file: bool = False) -> Union[int, Tuple[int, str, str]]`

Description:

Returns the duration of an audio file in seconds. Accepts a file path (str) or an AID (int).

Parameters:

Parameter	Type	Default	Description
`source`	`str` or `int`	—	File path or AID.
`is_file`	`bool`	`False`	If `True`, treats `source` as a file path.

Returns:

Success: int — Duration in seconds (floored).
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1001`	File does not exist.
`1002`	Invalid AID.
`1011`	Metadata parsing failed.
`1999`	Unknown error.

`get_audio_metadata(source: Union[str, int], is_file: bool = False) -> Union[Dict, Tuple[int, str, str]]`

Description:

Returns full metadata for an audio file. Accepts a file path (str) or an AID (int).

Parameters:

Parameter	Type	Default	Description
`source`	`str` or `int`	—	File path or AID.
`is_file`	`bool`	`False`	If `True`, treats `source` as a file path.

Returns:

Success: Dict — Contains path, format, duration, length, sample_rate, channels, bitrate.
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1001`	File does not exist.
`1002`	Invalid AID.
`1011`	Metadata parsing failed.
`1014`	`source` type is invalid.
`2003`	Opus file open failed (Opus-specific).
`2004`	OpusHead corrupted (Opus-specific).
`2008`	Opus bitrate unavailable (Opus-specific).

`get_audio_metadata_by_path(file_path: str) -> Union[Dict, Tuple[int, str, str]]`

Description:

Returns full metadata for an audio file by file path.

Parameters:

Parameter	Type	Description
`file_path`	`str`	Audio file path.

Returns:

Success: Dict.
Failure: Tuple[int, str, str].

`get_audio_metadata_by_aid(aid: int) -> Union[Dict, Tuple[int, str, str]]`

Description:

Returns full metadata for an audio file by AID.

Parameters:

Parameter	Type	Description
`aid`	`int`	Audio ID.

Returns:

Success: Dict.
Failure: Tuple[int, str, str].

Batch Parsing Methods

Windows Note: When using batch APIs, protect your entry point with if __name__ == "__main__":.

`batch_get_metadata(file_paths: Union[List[str], str], max_workers: Optional[int] = None, show_progress: bool = False) -> List[Dict]`

Description:

Parses multiple audio files in parallel using ProcessPoolExecutor. Accepts a list of file paths or a directory path (recursively scans supported formats).

Parameters:

Parameter	Type	Default	Description
`file_paths`	`List[str]` or `str`	—	List of files or directory path.
`max_workers`	`int`	`None`	Number of worker processes. Defaults to CPU core count. Must be a positive integer.
`show_progress`	`bool`	`False`	Prints progress to stdout.

Returns:

Success: List[Dict] — Metadata dicts for successfully parsed files (failed files are omitted).
Failure: Tuple[int, str, str] — When max_workers is invalid.

Error Codes:

Code	Condition
`1999`	`max_workers` is invalid (0, negative, non-integer, or exceeds platform limits).

`batch_get_duration(file_paths: Union[List[str], str], max_workers: Optional[int] = None) -> Dict[str, int]`

Description:

Returns durations for multiple files in parallel.

Returns:

Success: Dict[str, int] — {path: duration_seconds}.
Failure: Tuple[int, str, str] — Invalid max_workers.

`batch_get_metadata_by_type(file_paths: Union[List[str], str], file_type: str, max_workers: Optional[int] = None) -> List[Dict]`

Description:

Parses multiple files but only returns results matching a specific format (e.g., "mp3", "opus").

Returns:

Success: List[Dict] — Metadata for matching files.
Failure: Tuple[int, str, str] — Invalid max_workers.

DAP System Methods

`save_dap_to_json(save_path: str) -> Tuple[int, str, str]`

Description:

Saves DAP records to a .ap-ds-dap JSON file.

Parameters:

Parameter	Type	Description
`save_path`	`str`	Output path. Must end with `.ap-ds-dap`.

Returns:

Success: (0, "", "").
Failure: Tuple[int, str, str].

Error Codes:

Code	Condition
`1009`	Extension is not `.ap-ds-dap`.
`1010`	Write failed (bad path, permission denied, disk full).

`get_dap_recordings() -> List[Dict]`

Description:

Returns the current DAP records (a copy).

Returns: List[Dict] — Each record contains path, duration, bitrate, channels.

`clear_dap_recordings() -> None`

Description:

Clears all DAP records from memory.

Returns: None.

Resource Management

`clear_memory_cache() -> None`

Description:

Releases all cached Mix_Chunk and Mix_Music objects and clears the cache.

Returns: None.

`cleanup_function() -> None`

Description:

Cleans up all resources: clears cache, closes mixer, quits SDL. Automatically registered with atexit.

Returns: None.

Top-Level Convenience Functions

The following functions are exported at the package top level (from ap_ds import ...).

`batch_get_metadata(file_paths, max_workers=None, show_progress=False) -> List[Dict]`

Description: Parses multiple audio files in parallel.

`batch_get_duration(file_paths, max_workers=None) -> Dict[str, int]`

Description: Returns durations for multiple files in parallel.

`batch_get_metadata_by_type(file_paths, file_type, max_workers=None) -> List[Dict]`

Description: Parses in parallel and filters by format.

`get_audio_duration(file_path: str) -> int`

Description: Returns the duration of a single audio file in seconds.

`get_audio_metadata(file_path: str) -> Optional[Dict]`

Description: Returns full metadata for a single audio file.

`auto_check_runtime() -> Optional[Dict]`

Description: Runs a runtime self-check.

`check_runtime_mode() -> bool`

Description: Checks whether the GIL is enabled.

`show_tech_manual() -> None`

Description: Prints the complete built-in technical manual to stdout.

`is_full_performance() -> bool`

Description: Returns whether running in full-performance mode.

`get_runtime_info() -> Dict`

Description: Returns a runtime information dictionary.

AudioParser Module — Metadata API

The audio_parser module provides pure-Python metadata parsers with zero external dependencies.

Format-Specific Parser Classes

Class	Format	Accuracy	Parsing Method
`WAVFile`	WAV	100%	RIFF chunk structure
`FLACFile`	FLAC	100%	STREAMINFO metadata block
`MP3File`	MP3	>98%	Per-frame sync word scanning
`AACFile`	AAC (ADTS)	>99%	ADTS frame parsing
`OGGFile`	OGG Vorbis	99.99%	Granule position + Vorbis headers

All parser classes inherit from FileType, exposing properties length, sample_rate, channels, and bitrate.

`open_audio(filename: str) -> FileType`

Description:

Factory function that returns the appropriate parser instance for the file.

Raises:

ValueError: If the file format is unsupported.

SDL2 Integration Layer

The _sdl2 module provides the cross-platform SDL2 loader, constants, structures, and ctypes bindings. This is an internal module; users interact with it indirectly through AudioLibrary.

Global SDL2 Functions

Function	Description
`SDL_Init(flags)`	Initializes SDL subsystems. Returns 0 on success.
`SDL_Quit()`	Shuts down SDL.
`SDL_GetError()`	Returns the last SDL error message.
`SDL_RWFromFile(file, mode)`	Opens a file as an SDL RWops handle.
`SDL_Delay(ms)`	Suspends the calling thread for `ms` milliseconds.

Global SDL2_mixer Functions

Function	Description
`Mix_OpenAudio(freq, format, channels, chunksize)`	Opens the audio mixer. Returns 0 on success.
`Mix_CloseAudio()`	Closes the mixer.
`Mix_LoadWAV(file)`	Loads a WAV sound effect into a `Mix_Chunk`.
`Mix_LoadMUS(file)`	Loads a music file into a `Mix_Music`.
`Mix_FreeChunk(chunk)`	Frees a sound effect chunk.
`Mix_FreeMusic(music)`	Frees a music object.
`Mix_PlayChannel(channel, chunk, loops)`	Plays a chunk on a channel.
`Mix_PlayMusic(music, loops)`	Plays music.
`Mix_Pause(channel)` / `Mix_PauseMusic()`	Pauses a channel / music.
`Mix_Resume(channel)` / `Mix_ResumeMusic()`	Resumes a channel / music.
`Mix_HaltChannel(channel)` / `Mix_HaltMusic()`	Stops a channel / music.
`Mix_SetMusicPosition(position)`	Seeks music to position (seconds).
`Mix_Volume(channel, volume)`	Sets/gets channel volume (0–128, `-1` = get).
`Mix_VolumeMusic(volume)`	Sets/gets music volume (0–128, `-1` = get).
`Mix_FadeInMusic(music, loops, ms)`	Fades music in.
`Mix_FadeOutMusic(ms)`	Fades music out.
`Mix_FadeInMusicPos(music, loops, ms, position)`	Fades music in from position.

Constants Reference

SDL Initialization Flags

Constant	Value	Description
`SDL_INIT_TIMER`	`0x00000001`	Timer subsystem.
`SDL_INIT_AUDIO`	`0x00000010`	Audio subsystem.
`SDL_INIT_VIDEO`	`0x00000020`	Video subsystem.
`SDL_INIT_JOYSTICK`	`0x00000200`	Joystick subsystem.
`SDL_INIT_HAPTIC`	`0x00001000`	Haptic subsystem.
`SDL_INIT_GAMECONTROLLER`	`0x00002000`	Game controller subsystem.
`SDL_INIT_EVENTS`	`0x00004000`	Events subsystem.
`SDL_INIT_EVERYTHING`	`0x00007231`	All subsystems combined.

Audio Formats

Constant	Value	Description
`AUDIO_U8`	`0x0008`	Unsigned 8-bit.
`AUDIO_S8`	`0x8008`	Signed 8-bit.
`AUDIO_U16LSB`	`0x0010`	Unsigned 16-bit little-endian.
`AUDIO_S16LSB`	`0x8010`	Signed 16-bit little-endian.
`AUDIO_U16MSB`	`0x1010`	Unsigned 16-bit big-endian.
`AUDIO_S16MSB`	`0x9010`	Signed 16-bit big-endian.
`AUDIO_S32LSB`	`0x8020`	Signed 32-bit little-endian.
`AUDIO_S32MSB`	`0x9020`	Signed 32-bit big-endian.
`AUDIO_F32LSB`	`0x8120`	Float 32-bit little-endian.
`AUDIO_F32MSB`	`0x9120`	Float 32-bit big-endian.
`MIX_DEFAULT_FORMAT`	`AUDIO_S16SYS`	Default mixer format.

Mixer Initialization Flags

Constant	Value	Description
`MIX_INIT_FLAC`	`0x00000001`	FLAC support.
`MIX_INIT_MOD`	`0x00000002`	MOD support.
`MIX_INIT_MP3`	`0x00000008`	MP3 support.
`MIX_INIT_OGG`	`0x00000010`	OGG support.
`MIX_INIT_MID`	`0x00000020`	MIDI support.
`MIX_INIT_OPUS`	`0x00000040`	Opus support (unused in SDL2_mixer, replaced by OpusAudio).

Music Type Constants

Constant	Value	Description
`MUS_NONE`	`0`	No music.
`MUS_CMD`	`1`	Command-based.
`MUS_WAV`	`2`	WAV.
`MUS_MOD`	`3`	MOD.
`MUS_MID`	`4`	MIDI.
`MUS_OGG`	`5`	OGG.
`MUS_MP3`	`6`	MP3.
`MUS_FLAC`	`7`	FLAC.
`MUS_OPUS`	`8`	Opus (unused in SDL2_mixer).

Fade Status Constants

Constant	Value	Description
`MUS_NO_FADING`	`0`	No fade in progress.
`MUS_FADING_IN`	`1`	Fading in.
`MUS_FADING_OUT`	`2`	Fading out.

Unified Error Handling and Error Code Reference

Overview

Starting from v4.0.0, ap_ds uses a unified error handling system. All methods that can fail return a consistent tuple format:

(error_code: int, error_message: str, suggestion: str)

Success: (0, "", "")

Failure: (error_code, error_message, suggestion)

Complete Error Code Reference

Code	Constant	Meaning	Suggestion
0	`AP_DS_SUCCESS`	Operation completed successfully	No action needed
1001	`AP_DS_ERR_FILE_NOT_FOUND`	File does not exist	Verify file path exists and is accessible
1002	`AP_DS_ERR_INVALID_AID`	Invalid or expired Audio ID	Check AID is valid and audio is loaded
1003	`AP_DS_ERR_AUDIO_LOAD_FAILED`	Audio file loading failed	Check file format and integrity
1004	`AP_DS_ERR_PLAYBACK_FAILED`	Audio playback failed	Check audio device and file format
1005	`AP_DS_ERR_SDL_INIT_FAILED`	SDL2 initialization failed	Check SDL2 installation and audio driver
1006	`AP_DS_ERR_MIXER_INIT_FAILED`	SDL2_mixer initialization failed	Check audio device and available formats
1007	`AP_DS_ERR_UNSUPPORTED_FORMAT`	Audio format not supported	Use a supported format
1008	`AP_DS_ERR_NOT_MUSIC_FILE`	Operation only supports music files	Sound effects do not support this operation
1009	`AP_DS_ERR_DAP_INVALID_EXT`	Invalid DAP file extension	Use `.ap-ds-dap` extension when saving DAP records
1010	`AP_DS_ERR_DAP_SAVE_FAILED`	DAP record saving failed	Check write permissions and disk space
1011	`AP_DS_ERR_METADATA_PARSE_FAILED`	Metadata parsing failed	File may be corrupted or uses unsupported variant
1012	`AP_DS_ERR_FADE_NOT_SUPPORTED`	Fade operation not supported by SDL_mixer	Update SDL_mixer or use alternative method
1013	`AP_DS_ERR_AUDIO_NOT_LOADED`	Audio file not loaded into memory	Call `new_aid()` first to load file into cache
1014	`AP_DS_ERR_INVALID_SOURCE`	Invalid source type for metadata query	Use file path (str) or AID (int)
1015	`AP_DS_ERR_INVALID_VOLUME`	Volume value out of range	Volume must be between 0 and 128
1016	`AP_DS_ERR_SEEK_NOT_SUPPORTED`	Seeking not supported for this audio	Sound effects (short WAVs) do not support seeking
2001	`AP_DS_ERR_OPUS_LIB_LOAD_FAILED`	libopusfile loading failed	Check if DLL exists
2002	`AP_DS_ERR_OPUS_DLL_DEPENDENCY`	DLL dependency missing	Ensure libopus-0.dll and libogg-0.dll exist
2003	`AP_DS_ERR_OPUS_OPEN_FAILED`	Opus file open failed	File may be corrupted or not a valid Opus stream
2004	`AP_DS_ERR_OPUS_HEADER_CORRUPT`	OpusHead header corrupted	Header info invalid or corrupted
2005	`AP_DS_ERR_OPUS_TAGS_PARSE_FAILED`	Tags parsing failed	Tag data corrupted or invalid format
2006	`AP_DS_ERR_OPUS_DECODE_FAILED`	Opus decoding failed	Audio data corrupted
2007	`AP_DS_ERR_OPUS_SEEK_FAILED`	Opus seek failed	Stream may not support seeking to that position
2008	`AP_DS_ERR_OPUS_BITRATE_UNAVAILABLE`	Bitrate unavailable	Unable to determine bitrate for this Opus stream
2009	`AP_DS_ERR_OPUS_NOT_SEEKABLE`	Stream not seekable	This Opus stream does not support seeking
2010	`AP_DS_ERR_OPUS_CHANNEL_INVALID`	Invalid channel count	Opus stream has invalid channel count
1999	`AP_DS_ERR_UNKNOWN`	An unexpected error occurred	Check file integrity and retry

Environment Variables Reference

Variable	Default	Description
`AP_DS_HIDE_SUPPORT_PROMPT`	Not set	Set to `1` to hide startup banner
`AP_DS_WAV_THRESHOLD`	`6`	WAV mode switch threshold (seconds)
`AP_DS_SUPPRESS_WARNINGS`	Not set	Set to `1` to suppress deprecation warnings
`AP_DS_SHOW_CONGRATS`	Not set	Set to `0` to hide full-performance congratulatory message
`AP_DS_SKIP_AUTO_CHECK`	`1`	Set to `0` to enable runtime self-check on import

Technical Manual (show_tech_manual())

ap_ds 4.1.0 RC includes a built-in technical manual that can be displayed by calling show_tech_manual(). This manual contains:

Library overview
Supported audio formats (including Opus chapter)
Core components (AudioLibrary, metadata parser, SDL2 loader, Opus player)
DAP system documentation
WAV smart mode documentation
Environment variables reference
Performance optimization tips
Cross-platform notes (including macOS Opus special notes)
Troubleshooting guide
API reference
Version history
Contribution and support information

View the manual:

from ap_ds import show_tech_manual
show_tech_manual()

Version History

v4.1.0 RC (August 19, 2026) — Opus Support Preview

⚠️ This is an RC (Release Candidate) prerelease version for collecting feedback and bug reports.

🚨 Important: Opus support exists ONLY in version 4.1.0 of the mainline. Mainline 4.2.0+ will REMOVE Opus support, migrating to the AFS branch (ap-ds-afs).

🎯 New: Opus Support

Added opusplayer.py: OpusAudio class, complete Opus playback engine
Added _opusdll.py: Cross-platform Opus library loader (Windows auto-DLL download, Linux/macOS system library detection + installation)
Added 10 Opus-specific error codes (2001-2010)
Windows: Auto-download libopusfile-0.dll, libopus-0.dll, libogg-0.dll, libopusurl-0.dll with SHA256 hash verification
Linux: apt/dnf/pacman auto-installation + interactive guidance
macOS: Homebrew/MacPorts auto-detection + auto-install attempt + manual guidance
Cross-platform playback backends: Windows (winmm waveOut) / Linux (ALSA) / macOS (Core Audio AudioQueue)

🎯 New: AFS Branch

Announced AFS (All-Format Support) branch (ap-ds-afs)
Opus will migrate to AFS branch after 4.1.0 RC
Mainline 4.2.0+ will remove Opus, returning to 2.5MB lightweight positioning
Added package conflict detection (foolproof design)

🧪 Testing

Added OPUS_TEST.py: 229 Opus-specific tests, all passed
CI/CD integration: 650+ comprehensive tests, all passed
5 listening tests (normal playback, fade-in, fade-out, volume 0, Seek)

📚 Documentation

show_tech_manual() added 2.1 OPUS Support section
Added 10.4 Opus Error Codes section
Version history added 4.1.0 RC entry
Updated README complete version with API reference, error codes, environment variables, and all other content

📦 Size

Current 3.87 MB (4,059,251 bytes) — 4.1.0 RC only
4.2.0+ will remove Opus, returning to ~2.5MB

v4.0.0 (August 2026) — Architecture Refactor and Performance Edition

This is a complete rewrite of the core architecture. This is the largest refactor in ap_ds history, addressing all technical debt accumulated in the v3.1.x series.

🔧 Architecture Changes

Complete module refactoring:
- Merged audio_parser.py (wrapper) with audio_info.py (actual parsers) into a single audio_parser.py, eliminating circular dependencies
- Split player.py into player.py (AudioLibrary only) and _sdl2.py (loader + constants + bindings)
- Each file now has a single responsibility, easy to maintain and debug

🐛 Bug Fixes

Fixed v3.1.x circular import bug: get_audio_duration() and get_audio_metadata() returning 0 or None in some environments is now completely resolved
Fixed inconsistent import behavior across Python versions
Eliminated redundant wrapper layers that caused confusion

🚀 New Features

Unified error handling system: All methods return (error_code, error_message, suggestion) tuples,告别 the era of exception chaos and inconsistent return values
Batch parsing API: batch_get_metadata(), batch_get_duration(), batch_get_metadata_by_type() using ProcessPoolExecutor to process hundreds of files in parallel
Python 3.15t free-threading support: GIL-free true parallelism, batch parsing scales linearly on multi-core CPUs
O(1) DAP deduplication: _add_to_dap_recordings() uses set-based deduplication, near-instantaneous duplicate checking
Lazy imports (Python 3.15+): Heavy modules loaded on demand, speeding up import ap_ds
Runtime self-check: Automatic environment diagnosis on import (disable with AP_DS_SKIP_AUTO_CHECK=1)
Smart WAV mode: WAV files shorter than AP_DS_WAV_THRESHOLD (default 6 seconds) play as sound effects; longer files stream as music, supporting seeking and fade effects
Fade control: fadein_music(), fadein_music_pos(), fadeout_music() and status checks
Cross-platform SDL2 loader: Auto-downloads and hash-verifies SDL2 binaries on Windows and macOS; intelligent fallback on Linux
Parameter type validation: Every public entry point performs strict type checking, programming errors return clear error tuples instead of obscure ctypes.ArgumentError

📚 Documentation

Complete API reference: covers all AudioLibrary methods, parameters, return values, error codes
Unified error code reference: list of all error codes from 1001 to 1999
Environment variables reference: AP_DS_WAV_THRESHOLD, AP_DS_SUPPRESS_WARNINGS, etc.
show_tech_manual() built-in technical manual

🔗 Repository

Added GitHub compatibility mirror: dvs-dvsxt/ap_ds, providing convenience for GitHub users
Primary repositories remain apds.top and GitCode

⚡ Performance

120 MP3 files batch parsing: 0.331 seconds (8 processes), 2.94x faster than Mutagen (0.973 seconds)
3.88x faster than v3.0.0 multi-threaded solution (1.285 seconds)

v3.1.2 (July 2026) — Performance and Batch Parsing Edition (LFV)

🚨 Known Issue: Circular import bug causing get_audio_duration() and get_audio_metadata() to return 0 or None in some environments. This is fixed in v4.0.0. Users are advised to skip this version and upgrade directly to v4.0.0.

🚀 New Features

Batch parsing API:
- batch_get_metadata(): Batch parse audio files, returns full metadata list
- batch_get_duration(): Batch get audio durations, returns {path: duration}
- batch_get_metadata_by_type(): Batch parse filtered by format
Python 3.15t free-threading support: Full adaptation to GIL-free environment, runtime auto-detection of GIL status
DAP deduplication optimization: Upgraded from O(n) linear scan to O(1) set-based deduplication, with O(n) fallback retained
Lazy imports (Python 3.15+): Heavy modules loaded on demand
Runtime diagnostic functions: is_full_performance() and get_runtime_info()
Runtime self-check: Auto-executed on import (skip with AP_DS_SKIP_AUTO_CHECK=1)

⚡ Performance

120 MP3 files: Serial 1.367s → 8-process parallel 0.331s (4.13x speedup)
2.94x faster than Mutagen, 3.88x faster than v3.0.0

📦 Version Relationship

This version is an LFV (Latest Feature Version), not LTS
Next LTS is v4.0.0 LTS (after Python 3.15 stabilizes)

v3.0.0 LTS (March 22, 2026) — First Long-Term Support Version

This is ap_ds's first LTS version. After years of refinement, extensive real-world testing, and thorough internal resource management refactoring, this version is ready for mission-critical applications, enterprise deployments, and personal projects.

🎯 New Features

Deterministic resource cleanup: Replaced unreliable __del__ finalizers with explicit exit handlers
Hash-verified downloads: Each downloaded SDL2 library is verified against hardcoded SHA-256 hashes before use
Complete test coverage: Tested on all platforms with zero memory leaks
5-year support period: Until March 22, 2031, with free technical support

🔄 Compatibility

No breaking changes, fully backward compatible with v2.x

v2.4.2 (March 22, 2026) — Development Mishap Version

⚠️ This version was accidentally uploaded with a development-stage player.py file. While technically usable, it may contain subtle issues and is not recommended for use in any real-world project.

This version is for curious exploration only — do not use in production environments.

v2.4.1 (March 1, 2026) — Documentation Update

Updated PyPI documentation to fully reflect v2.4.0's new features.

📚 Changes

Updated PyPI project description
Added detailed examples for all new fade functions
Documented AP_DS_HIDE_SUPPORT_PROMPT environment variable
Improved quick-start guide

Note: No code changes in this release — documentation only.

v2.4.0 (March 1, 2026) — Audio Effects and Engineering Improvements

Introduced professional audio transitions and important internal engineering upgrades.

🎵 New Audio Control Functions

Function	Description
`fadein_music(aid, loops=-1, ms=0)`	Fade music in over specified milliseconds
`fadein_music_pos(aid, loops=-1, ms=0, position=0.0)`	Fade music in from specified position
`fadeout_music(ms=0)`	Fade out currently playing music
`is_music_playing()`	Check if music is playing
`is_music_paused()`	Check if music is paused
`get_music_fading()`	Get current fade status

🧠 Engineering Improvements

Cleaner startup banner, controllable via AP_DS_HIDE_SUPPORT_PROMPT
Centralized version management
Robust import system (dual-layer fallback)
Unified project URLs

🔄 Compatibility

No breaking changes, all existing code continues to work

v2.3.6 (February 27, 2026) — Documentation Update

Updated PyPI documentation with detailed license information and version history, added more examples.

v2.3.5 (February 26, 2026) — Stability Optimization and Embedded Verification

Six-dimensional test coverage:

Library loading and initialization
Playback testing (MP3, FLAC, OGG, WAV)
Seek testing
Memory pressure and leak detection (~4MB growth)
Metadata parsing accuracy
DAP system validation

Embedded Platform Support:

Orange Pi 4 Pro (Allwinner A733)
Raspberry Pi 5 (BCM2712)

🐛 Bug Fixes:

Fixed WAV files being incorrectly treated as sound effects — configurable via AP_DS_WAV_THRESHOLD

v2.3.4 (February 10, 2026) — Linux Smart Import System

Revolutionary Linux support improvement with four-layer fallback strategy:

System library check
User configuration check
Automatic package manager installation (apt-get, dnf, pacman)
Interactive guidance

Automatic Configuration Saving:

Environment variables (AP_DS_SDL2_PATH, AP_DS_SDL2_MIXER_PATH)
Persistent config file (~/.config/ap_ds/sdl_paths.conf)

v2.3.3 (February 9, 2026) — Critical Bug Fix and Platform Stabilization

🚨 Critical Update: Fixed a severe segmentation fault that caused the library to fail on macOS and Linux.

Root Cause: Earlier versions only defined C function prototypes (ctypes argtypes/restype) on Windows, causing memory access violations on other operating systems.

Solution: All necessary C function bindings are now unconditionally defined after loading the SDL2 library.

v2.3.2 (February 9, 2026) — Linux Support Enhancement

Expanded Linux support with interactive setup.

Interactive Linux Support:

Use system-installed libraries
Specify compiled .so file paths
Get detailed compilation instructions

v2.3.1 (February 9, 2026) — Documentation Update

Improved README.md with better examples and explanations. Fixed minor errors in documentation examples.

v2.3.0 (January 31, 2026) — DAP Recording System

Introducing the DAP (Dvs Audio Playlist) system.

Core Features:

Smart automatic recording: Auto-triggered in play_from_file(), play_from_memory()
Lightweight design: Metadata only, no audio data
Standardized file format: .ap-ds-dap extension, JSON format
Intelligent deduplication: Automatically avoids duplicate recordings of the same file

New APIs:

_add_to_dap_recordings(file_path) — internal use
save_dap_to_json(save_path) — save as JSON
get_dap_recordings() — get all records
clear_dap_recordings() — clear records

v2.2.0 (January 19, 2026) — Cross-Platform Revolution

From single-platform to cross-platform.

Major New Features:

1. Full macOS Support

Automatically downloads and installs SDL2.framework, SDL2_mixer.framework
Intelligent .dmg file extraction and framework loading
Maintains extreme lightweight: only 3.36MB (vs Windows 2.5MB)

2. Enhanced Automatic Dependency Management

Cross-platform intelligent download strategy
Complete error handling and retry mechanisms
Local caching of dependency files

v2.1.4 (January 18, 2026) — Stable Release

Production-ready stable release.

Core stability: Extensively tested, no known critical bugs
Extreme lightweight: Only 2.5MB complete solution
Full documentation: Detailed technical manual and examples

v2.1.0 (December 26, 2025) — Feature Enhancement

Professional feature expansion.

New Features:

Metadata enhancement: More precise audio information parsing
Playback precision improvements: Better time control and seeking

v2.0.0 (November 5, 2025) — Architecture Refactor

Introducing modern audio management system.

Major Improvements:

AID system: Unified audio instance management
Architecture refactor: Modular design, improved maintainability
Smart memory management: Automatic cleanup of unused audio resources
State management: Unified playback state tracking

v1.0.0 (July 8, 2025) — Initial Release

Project birth, basic functionality.

Core Features:

Basic audio playback: MP3, WAV, FLAC, OGG formats
Playback controls: Basic play, pause, stop, seek APIs
Volume control: Real-time volume adjustment (0-100%)
Lightweight design: ~2MB initial version

License

This project is licensed under the DVS Audio Library (ap_ds) Open Source License v2.0. The full license text is below. Use, copying, modification, or distribution of this software constitutes acceptance of all terms and conditions of this license.

DVS Audio Library (ap_ds) Open Source License v2.0

Version: 2.0 Effective Date: March 22, 2026 Applies to: ap_ds 2.4.1 and above (except for subsequent license updates) Project Homepage: https://apds.top

1. Definitions

1.1. "Software" means the DVS Audio Library (ap_ds) project and all its components, source code, object code, and related documentation.

1.2. "Source Code" means the human-readable form of the Software.

1.3. "Modified Version" means any derivative work created by modifying, supplementing, translating, or otherwise altering the Software.

1.4. "Distribution" means making the Software or a Modified Version available to any third party by any means or medium.

1.5. "You" means any individual or legal entity exercising rights granted by this License.

1.6. "Independent Brand" means a completely new project name, logo, and brand identity that does not create a confusing association with the Software's official names (including but not limited to "ap_ds", "AP_DS", "Audio Library By DVS", "DVS Audio Player", and any variations thereof).

2. License Grant

Subject to the terms and conditions of this License, the author grants You a perpetual, worldwide, royalty-free, non-exclusive, irrevocable right to:

2.1. Use and Run: Run the Software on any computer system for any lawful purpose.

2.2. Copy and Distribute: Make any number of copies of the Software and distribute them.

2.3. Study and Modify: Study the Source Code of the Software and make any modifications to suit Your needs.

2.4. Integrate and Use Commercially: Integrate the Software into Your products or projects and use it in any commercial environment.

3. Obligations and Restrictions

3.1. Attribution and Source Identification

Whenever using, distributing, or integrating the Software or a Modified Version, You must:

a) Retain Original Copyright Notices: Retain all original copyright, patent, and trademark notices in all copies of the Software.

b) Provide Prominent Source Attribution: Clearly and prominently state in the Software's documentation, official website, user interface, or related materials:

Based on DVS Audio Library (ap_ds) v[version number]
Original Author: Dvs (DvsXT)
Project Homepage: https://apds.top

c) Add Notice for Modified Versions: If You distribute a Modified Version, in addition to the above attribution, You must add the following notice:

This is a modified version maintained by [Your name/organization].
Support: [Your contact information].
This version is not official and is not affiliated with the original author.

3.2. Brand Protection

To prevent brand confusion and project fragmentation, Modified Versions must comply with the following strict rules:

a) No Use of Original Brand Names: You may not name a Modified Version "ap_ds", "AP_DS", "Audio Library By DVS", "DVS Audio Player", or any confusingly similar variation, combination, or derivative name.

b) Independent Brand Requirement: Modified Versions must use a completely independent project name and establish their own independent project identity, documentation, and community.

c) Maintainer Responsibility Statement: Distributors of Modified Versions must state on their project homepage or in a prominent location:

This project is based on DVS Audio Library (ap_ds) but has independently evolved and is entirely maintained by [Your name].
For the original version, please visit: https://apds.top.
The maintainer is solely responsible for any issues related to this project.

3.3. Modified Version Quality Commitment

If You distribute a Modified Version, You must:

a) Clearly State Modifications: Clearly indicate that this is a Modified Version and list key modifications and compatibility notes compared to the original version.

b) Provide Technical Support: Provide valid technical support contact information for the Modified Version You distribute, and define the scope of support.

c) Do Not Mislead Users: You may not imply in any way that Your Modified Version is officially endorsed, supported, or a continuation of the original project.

3.4. Prohibited Uses

You may not use the Software for any illegal activities, malicious purposes, or in violation of local laws and regulations.

4. Patent Grant

4.1. Patent License: The author grants You a worldwide, royalty-free, non-exclusive, non-transferable patent license to make, use, sell, offer for sale, import, or otherwise transfer the Software.

4.2. Patent Defense Termination: If You or Your affiliates file a patent infringement lawsuit against the author regarding the Software, all rights granted to You under this License will automatically and immediately terminate.

5. Technical Transparency and Security

5.1. Security Review Right: Any user has the right to conduct security audits of the Software's Source Code.

5.2. Security Reporting: Reporting of discovered security issues to the original author (me@dvsyun.top) is encouraged, and public disclosure after resolution is supported.

5.3. No Backdoors Commitment: Official releases commit to containing no malicious code, backdoors, or functionality that collects user data without explicit user consent.

6. Disclaimer of Warranties and Limitation of Liability

6.1. Disclaimer of Warranties: The Software is provided "AS IS", without warranty of any kind, express or implied.

6.2. Limitation of Liability: To the maximum extent permitted by applicable law, the author or copyright holder shall not be liable for any direct, indirect, incidental, special, consequential, or punitive damages.

7. License Management and Termination

7.1. Version Control: This License is version 2.0. Subsequent versions will be published on the project homepage.

7.2. Compatibility: This License is compatible with the MIT, BSD 3-Clause, and Apache 2.0 licenses.

7.3. Automatic Termination: If You fail to comply with the terms of this License, Your rights will automatically terminate.

8. Governing Law and Dispute Resolution

8.1. Governing Law: This License shall be governed by the laws of the People's Republic of China.

8.2. Dispute Resolution: Any dispute arising out of or in connection with this License shall first be resolved through friendly negotiation. If negotiation fails, either party may submit the dispute to the competent people's court in the place of the project author's domicile.

9. Contact Information

9.1. Licensing Inquiries:

Email: me@dvsyun.top or dvs6666@163.com
Project Homepage: https://apds.top
Response Time: Within 7 business days

9.2. Technical Support:

Priority: File issues via GitCode Issues
Urgent matters: Send to the above email addresses

Use, copying, modification, or distribution of this software constitutes acceptance of all terms and conditions of this license.

SDL2 Acknowledgements

Sincere Gratitude

ap_ds would not exist without the extraordinary work of the SDL2 development team. We owe them a debt of gratitude that words cannot adequately express.

To Sam Lantinga and the entire SDL development community:

Thank you. From the bottom of our hearts, thank you.

You have built something truly remarkable. For over twenty years, SDL has been the backbone of countless games, multimedia applications, and creative projects worldwide. It runs on everything — Windows, macOS, Linux, Android, iOS, game consoles, and embedded devices. It is stable, efficient, and beautifully designed. It is one of the most important open-source projects of our time.

We are just one small library among thousands that depend on your work. But we are deeply grateful. Every time a user plays an audio file through ap_ds, it's SDL2 doing the heavy lifting — decoding, mixing, and streaming audio with low latency and rock-solid reliability. We just provide the Python wrapper. You provide the magic.

Legal Compliance

This library uses the Simple DirectMedia Layer (SDL2) and SDL2_mixer libraries.

SDL2 Website: https://www.libsdl.org/
SDL2 License: zlib/libpng license
SDL2_mixer Website: https://www.libsdl.org/projects/SDL_mixer/
SDL2_mixer License: zlib/libpng license

The zlib/libpng license is a permissive free software license that allows free use, modification, and distribution of the software in commercial products with minimal attribution requirements.

Closing

ap_ds is built on a simple philosophy: Focus on playback and parsing, stay lightweight, and let developers build great applications.

We welcome feedback, bug reports, and contributions. If you have questions or concerns, please contact us through the official channels.

Thank you for using ap_ds!

112 KiB Raw Permalink Blame History Unescape Escape

🎉 ap_ds 4.1.0 RC Prerelease — Opus Support, But Please Help Us Test!

ap_ds: Lightweight Python Audio Library

📢 To All Audio Developers

🚨 Important Notes on 4.1.0's Version Positioning

1. This Is an RC Prerelease

2. Opus Support Exists ONLY in 4.1.0 Mainline

3. Issue Reporting and Fix Commitment

🤔 Why Opus? — The Inevitability of a Technical Choice

1. The "Ceiling" of Quality and Compression

2. Open Source and Freedom — A Philosophical Fit

3. Mature Ecosystem and Clear Demand

4. Paving the Way for the AFS Branch

😅 Why Wasn't Opus Supported Before? — The Story of Upstream Dependencies

How Did 4.1.0 Solve It?

🧬 4.1.0 Technical Deep Dive — What Did We Actually Do?

1. New Modules and Architecture Refactoring

2. Cross-Platform Playback Backends — Three Platform Engines Rewritten for Opus

3. Opus-Specific Error Code System (2001-2010)

4. Automatic DLL Download and Hash Verification (Windows)

5. Cross-Platform Opus Library Loading Explained — Why Can't macOS Auto-Download Like SDL2?

Background: How Does SDL2's Auto-Download Work?

So Why Can't Opus Follow the Same Path?

What About macOS Users Then?

What About Linux? — Consistent with SDL2

Why Do macOS Users See an "Apology" Rather Than a "Normal Prompt"?

Technical Comparison Summary

What Have We Done to Compensate?

Quick Guide for macOS Users

6. OpusAudio Class — Complete API

7. _opusdll.py — Cross-Platform Library Loader

8. Library Size Changes — 4.1.0's "Temporary Weight Gain" and 4.2.0's "Slim-Down Return"

⚠️ IMPORTANT: Windows Multiprocessing Warning

If you're using ap_ds 4.1.0 RC on Windows…

What's the problem?

Which APIs are affected?

How to fix it?

❌ Wrong (Crashes on Windows):

✅ Correct:

✅ Works with config functions too:

✅ Jupyter Notebook Users:

Why don't Linux and macOS have this problem?

Frequently Asked Questions

🌿 AFS Branch: ap_ds's "Dual-Track" Future

Why "Split the Family"?

Mainline 4.1.0's Special Status

AFS Branch (All-Format Support)

Which Should Users Choose?

Foolproof Design: Conflict Detection

📖 Technical Manual Update

2.1 OPUS Support (New Section)

10.4 Opus Error Codes (New Section)

Version History

🚀 Getting Started

Installation

💡 Why Python 3.15t / 3.14t?

Windows Users

Linux Users

macOS Users

Verify Installation

Create Virtual Environment

Quick Start

Playing Opus Files (via AudioLibrary, Recommended)

Playing Opus Files (Directly Using OpusAudio)

Getting Opus Metadata

Batch Parsing

⚠️ CRITICAL: Windows Batch Parsing and BrokenProcessPool

❌ Wrong (Will Crash on Windows):

✅ Correct (Always Works):

✅ Correct (Script with Configuration):

DAP Playlist System

Platform Support

Windows

macOS

Linux

Embedded ARM64

🧪 CI/CD Testing — Comprehensive Coverage, All Passed

Opus Test Suite (OPUS_TEST.py)

Test Results

CI/CD Integration

112 KiB

Raw Permalink Blame History

7. `_opusdll.py` — Cross-Platform Library Loader

`init(frequency: int = 44100, format: int = MIX_DEFAULT_FORMAT, channels: int = 2, chunksize: int = 2048) -> None`

`play_from_file(file_path: str, loops: int = 0, start_pos: float = 0.0) -> Union[int, Tuple[int, str, str]]`

`play_from_memory(file_path: str, loops: int = 0, start_pos: float = 0.0) -> Union[int, Tuple[int, str, str]]`

`new_aid(file_path: str) -> Union[int, Tuple[int, str, str]]`

`play_audio(aid: int) -> Tuple[int, str, str]`

`pause_audio(aid: int) -> Tuple[int, str, str]`

`stop_audio(aid: int) -> Union[float, Tuple[int, str, str]]`

`seek_audio(aid: int, position: float) -> Tuple[int, str, str]`

`set_volume(aid: int, volume: int) -> Tuple[int, str, str]`

`get_volume(aid: int) -> Union[int, Tuple[int, str, str]]`

`fadein_music(aid: int, loops: int = -1, ms: int = 0) -> Tuple[int, str, str]`

`fadein_music_pos(aid: int, loops: int = -1, ms: int = 0, position: float = 0.0) -> Tuple[int, str, str]`

`fadeout_music(ms: int = 0) -> Tuple[int, str, str]`

`is_music_playing() -> bool`

`is_music_paused() -> bool`

`get_music_fading() -> int`

`get_audio_duration(source: Union[str, int], is_file: bool = False) -> Union[int, Tuple[int, str, str]]`

`get_audio_metadata(source: Union[str, int], is_file: bool = False) -> Union[Dict, Tuple[int, str, str]]`

`get_audio_metadata_by_path(file_path: str) -> Union[Dict, Tuple[int, str, str]]`

`get_audio_metadata_by_aid(aid: int) -> Union[Dict, Tuple[int, str, str]]`

`batch_get_metadata(file_paths: Union[List[str], str], max_workers: Optional[int] = None, show_progress: bool = False) -> List[Dict]`

`batch_get_duration(file_paths: Union[List[str], str], max_workers: Optional[int] = None) -> Dict[str, int]`

`batch_get_metadata_by_type(file_paths: Union[List[str], str], file_type: str, max_workers: Optional[int] = None) -> List[Dict]`

`save_dap_to_json(save_path: str) -> Tuple[int, str, str]`

`get_dap_recordings() -> List[Dict]`

`clear_dap_recordings() -> None`

`clear_memory_cache() -> None`

`cleanup_function() -> None`

`batch_get_metadata(file_paths, max_workers=None, show_progress=False) -> List[Dict]`