数字我 Digital Me：全量入库（大文件走 Git LFS）

2026-10-10 16:24:43 +08:00
commit 0593ff7e62
1893 changed files with 361723 additions and 0 deletions
@@ -0,0 +1,97 @@
+from __future__ import annotations
+
+import argparse
+import json
+from pathlib import Path
+
+
+BITRATES = {
+    (1, 1): [0, 32, 40, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256, 320, 0],
+    (1, 2): [0, 32, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256, 320, 384, 0],
+    (1, 3): [0, 32, 40, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256, 320, 0],
+    (2, 1): [0, 32, 48, 56, 64, 80, 96, 112, 128, 144, 160, 176, 192, 224, 256, 0],
+    (2, 2): [0, 8, 16, 24, 32, 40, 48, 56, 64, 80, 96, 112, 128, 144, 160, 0],
+    (2, 3): [0, 8, 16, 24, 32, 40, 48, 56, 64, 80, 96, 112, 128, 144, 160, 0],
+}
+SAMPLE_RATES = {1: [44100, 48000, 32000], 2: [22050, 24000, 16000], 25: [11025, 12000, 8000]}
+
+
+def synchsafe(b: bytes) -> int:
+    return (b[0] << 21) | (b[1] << 14) | (b[2] << 7) | b[3]
+
+
+def analyze(path: Path) -> dict:
+    data = path.read_bytes()
+    pos = 0
+    if data[:3] == b"ID3" and len(data) >= 10:
+        pos = 10 + synchsafe(data[6:10])
+        if data[5] & 0x10:
+            pos += 10
+
+    frames = 0
+    duration = 0.0
+    bitrates = []
+    sample_rates = []
+    channel_modes = []
+    first_frame = None
+    while pos + 4 <= len(data):
+        h = int.from_bytes(data[pos : pos + 4], "big")
+        if (h >> 21) & 0x7FF != 0x7FF:
+            pos += 1
+            continue
+        version_bits = (h >> 19) & 0x3
+        layer_bits = (h >> 17) & 0x3
+        bitrate_idx = (h >> 12) & 0xF
+        sr_idx = (h >> 10) & 0x3
+        padding = (h >> 9) & 0x1
+        channel_mode = (h >> 6) & 0x3
+        if version_bits == 1 or layer_bits == 0 or bitrate_idx in (0, 15) or sr_idx == 3:
+            pos += 1
+            continue
+        version = {3: 1, 2: 2, 0: 25}[version_bits]
+        layer = {3: 1, 2: 2, 1: 3}[layer_bits]
+        br_version = 1 if version == 1 else 2
+        bitrate = BITRATES[(br_version, layer)][bitrate_idx] * 1000
+        sr = SAMPLE_RATES[version][sr_idx]
+        if layer == 1:
+            samples = 384
+            frame_len = ((12 * bitrate // sr) + padding) * 4
+        elif layer == 3 and version != 1:
+            samples = 576
+            frame_len = (72 * bitrate // sr) + padding
+        else:
+            samples = 1152
+            frame_len = (144 * bitrate // sr) + padding
+        if frame_len <= 4 or pos + frame_len > len(data):
+            pos += 1
+            continue
+        if first_frame is None:
+            first_frame = pos
+        frames += 1
+        duration += samples / sr
+        bitrates.append(bitrate)
+        sample_rates.append(sr)
+        channel_modes.append(channel_mode)
+        pos += frame_len
+
+    return {
+        "path": str(path),
+        "bytes": len(data),
+        "frames": frames,
+        "duration_seconds": round(duration, 6),
+        "duration_hms": f"{int(duration//60):02d}:{duration%60:05.2f}",
+        "sample_rate_hz": max(set(sample_rates), key=sample_rates.count) if sample_rates else None,
+        "channels": "mono" if channel_modes and all(x == 3 for x in channel_modes) else "stereo",
+        "average_bitrate_kbps": round(sum(bitrates) / len(bitrates) / 1000, 2) if bitrates else None,
+        "min_bitrate_kbps": min(bitrates) // 1000 if bitrates else None,
+        "max_bitrate_kbps": max(bitrates) // 1000 if bitrates else None,
+        "vbr": len(set(bitrates)) > 1,
+        "first_audio_frame_offset": first_frame,
+    }
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("path", type=Path)
+    args = parser.parse_args()
+    print(json.dumps(analyze(args.path), ensure_ascii=False, indent=2))