You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

78 lines
2.9 KiB

#!/usr/bin/env python3
import subprocess
import tempfile
import os
import os.path
import collections
import json
import sys
import multiprocessing
import struct
ProcessedTrack = collections.namedtuple("ProcessedTrack", ["dfpwm_file", "metadata"])
def convert_wav_dfpwm(infile, outfile):
subprocess.run(["java", "-jar", "LionRay.jar", infile, outfile])
def convert_any_wav(infile, outfile):
subprocess.run(["ffmpeg", "-hide_banner", "-i", infile, "-ar", "48000", "-ac", "1", outfile], stderr=subprocess.PIPE)
def read_meta(path):
proc = subprocess.run(["ffprobe", "-v", "quiet", "-print_format", "json", "-show_format", "-show_streams", path], stdout=subprocess.PIPE)
data = json.loads(proc.stdout)
meta = {}
# These are the two locations I've found tags in in my not very thorough testing
try:
meta.update(data["format"]["tags"])
except KeyError: pass
try:
meta.update(data["streams"][0]["tags"])
except KeyError: pass
# lowercase all keys because in Opus files these seem to be uppercase sometimes
return { k.lower(): v for k, v in meta.items() }
def process_file(filename):
meta = read_meta(filename)
wav_dest = tempfile.mktemp(".wav")
convert_any_wav(filename, wav_dest)
dfpwm_dest = tempfile.mktemp(".dfpwm")
convert_wav_dfpwm(wav_dest, dfpwm_dest)
os.remove(wav_dest)
print(filename)
return ProcessedTrack(dfpwm_dest, {
"title": meta["title"],
"artist": meta.get("artist", None) or meta.get("artists", None),
"album": meta.get("album", None)
})
def read_binary(filename):
with open(filename, "rb") as f:
return f.read()
def process_dir(dirname):
files = list(map(lambda file: os.path.join(dirname, file), os.listdir(dirname)))
with multiprocessing.Pool(8) as p:
tracks = p.map(process_file, files)
tape_image = b""
tracks_meta = []
for track in tracks:
track.metadata["start"] = len(tape_image)
data = read_binary(track.dfpwm_file)
os.remove(track.dfpwm_file)
track.metadata["end"] = track.metadata["start"] + len(data)
tape_image += data
tracks_meta.append(track.metadata)
# dump in a compact format to save space
meta = json.dumps({ "tracks": tracks_meta }, separators=(',', ':')).encode("utf-8")
assert(len(meta) < 65536)
# new format - 0x54 marker byte, then metadata length as 2-byte big endian integer, then metadata, then concatenated DFPWM files
# start is now not an absolute position but just how far after the metadata it is
tape_image = b"\x54" + struct.pack(">H", len(meta)) + meta + tape_image
with open("tape.bin", "wb") as f:
f.write(tape_image)
# Tape lengths are measured in minutes. 6000 bytes are played per second because they use a 48000Hz sample rate and DFPWM is somehow 1 bit per sample.
length_minutes = len(tape_image) / (6000*60)
print(length_minutes, "minute tape required")
process_dir(sys.argv[1])