fix(transcoding): make piped FLAC transcodes seekable
The FLAC muxer writes STREAMINFO before it knows the stream length, then
rewinds at the end to fill total_samples in. Navidrome pipes ffmpeg's stdout
(-f flac -), which is not seekable, so ffmpeg logs "unable to rewrite FLAC
header" and the field stays 0. A decoder needs total_samples to turn a
timestamp into a byte offset, so it reports an unknown duration and refuses to
seek. Online playback hides this because the client re-requests with a new
offset each time, but an offline copy is permanently unseekable, the symptom
reported against Symfonium where seeking a downloaded track jumps back to the
start.
Transcode now wraps its own output and rewrites total_samples as the first
bytes flow past. This lives in core/ffmpeg because the unseekable pipe is that
package's doing: buildDynamicArgs is what appends the trailing '-'. core/stream
only learns a target format and hands back an io.ReadCloser, so compensating
there leaked a transcoder implementation detail one layer up. TranscodeOptions
grows a Duration field alongside the existing Offset, which also puts the
duration-minus-offset arithmetic in the same function that emits -ss.
The wrapper runs on every transcode rather than only FLAC targets: the format
on a transcoding row is a declared target that nothing validates against the
command's actual -f, so a custom command can emit FLAC under any target_format.
The magic-byte check inside the wrapper is the authoritative test and costs a
26-byte peek. The output sample rate is read back out of the header ffmpeg just
wrote rather than taken from the transcode options, so a resampled (-ar) output
still gets the right count. Anything that is not a FLAC stream with an unset
total_samples passes through byte for byte.
Measured on a 177s source: before, total_samples=0 and ffprobe reported
duration N/A; after, total_samples=7807023 and duration 177.03s, with the audio
payload byte-identical. This affects every piped FLAC regardless of the source
format; only FLAC stores an authoritative "unknown", which is why mp3, opus
and aac survive the same pipe.
No SEEKTABLE is synthesised and the MD5 is left zero: both are optional, and
decoders binary-search using total_samples alone.
2026-09-07 14:19:45 -04:00
|
|
|
package ffmpeg
|
|
|
|
|
|
|
|
|
|
import (
|
|
|
|
|
"encoding/binary"
|
|
|
|
|
"math"
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
const (
|
|
|
|
|
flacPrefixLen = 26 // through the last total_samples byte
|
|
|
|
|
flacMaxTotalSamples = 1<<36 - 1
|
|
|
|
|
)
|
|
|
|
|
|
2026-11-11 19:03:55 -05:00
|
|
|
// flacPrefix fills in the STREAMINFO total_samples that ffmpeg leaves at 0 when
|
|
|
|
|
// writing to a pipe, since a decoder cannot seek a cached FLAC without it.
|
|
|
|
|
func (h *headerPatcher) flacPrefix(buf []byte) ([]byte, error) {
|
|
|
|
|
buf, err := h.fill(buf, flacPrefixLen)
|
|
|
|
|
if err != nil || len(buf) < flacPrefixLen {
|
|
|
|
|
return buf, err
|
fix(transcoding): make piped FLAC transcodes seekable
The FLAC muxer writes STREAMINFO before it knows the stream length, then
rewinds at the end to fill total_samples in. Navidrome pipes ffmpeg's stdout
(-f flac -), which is not seekable, so ffmpeg logs "unable to rewrite FLAC
header" and the field stays 0. A decoder needs total_samples to turn a
timestamp into a byte offset, so it reports an unknown duration and refuses to
seek. Online playback hides this because the client re-requests with a new
offset each time, but an offline copy is permanently unseekable, the symptom
reported against Symfonium where seeking a downloaded track jumps back to the
start.
Transcode now wraps its own output and rewrites total_samples as the first
bytes flow past. This lives in core/ffmpeg because the unseekable pipe is that
package's doing: buildDynamicArgs is what appends the trailing '-'. core/stream
only learns a target format and hands back an io.ReadCloser, so compensating
there leaked a transcoder implementation detail one layer up. TranscodeOptions
grows a Duration field alongside the existing Offset, which also puts the
duration-minus-offset arithmetic in the same function that emits -ss.
The wrapper runs on every transcode rather than only FLAC targets: the format
on a transcoding row is a declared target that nothing validates against the
command's actual -f, so a custom command can emit FLAC under any target_format.
The magic-byte check inside the wrapper is the authoritative test and costs a
26-byte peek. The output sample rate is read back out of the header ffmpeg just
wrote rather than taken from the transcode options, so a resampled (-ar) output
still gets the right count. Anything that is not a FLAC stream with an unset
total_samples passes through byte for byte.
Measured on a 177s source: before, total_samples=0 and ffprobe reported
duration N/A; after, total_samples=7807023 and duration 177.03s, with the audio
payload byte-identical. This affects every piped FLAC regardless of the source
format; only FLAC stores an authoritative "unknown", which is why mp3, opus
and aac survive the same pipe.
No SEEKTABLE is synthesised and the MD5 is left zero: both are optional, and
decoders binary-search using total_samples alone.
2026-09-07 14:19:45 -04:00
|
|
|
}
|
2026-11-11 19:03:55 -05:00
|
|
|
setFLACTotalSamples(buf, h.duration)
|
|
|
|
|
return buf, nil
|
fix(transcoding): make piped FLAC transcodes seekable
The FLAC muxer writes STREAMINFO before it knows the stream length, then
rewinds at the end to fill total_samples in. Navidrome pipes ffmpeg's stdout
(-f flac -), which is not seekable, so ffmpeg logs "unable to rewrite FLAC
header" and the field stays 0. A decoder needs total_samples to turn a
timestamp into a byte offset, so it reports an unknown duration and refuses to
seek. Online playback hides this because the client re-requests with a new
offset each time, but an offline copy is permanently unseekable, the symptom
reported against Symfonium where seeking a downloaded track jumps back to the
start.
Transcode now wraps its own output and rewrites total_samples as the first
bytes flow past. This lives in core/ffmpeg because the unseekable pipe is that
package's doing: buildDynamicArgs is what appends the trailing '-'. core/stream
only learns a target format and hands back an io.ReadCloser, so compensating
there leaked a transcoder implementation detail one layer up. TranscodeOptions
grows a Duration field alongside the existing Offset, which also puts the
duration-minus-offset arithmetic in the same function that emits -ss.
The wrapper runs on every transcode rather than only FLAC targets: the format
on a transcoding row is a declared target that nothing validates against the
command's actual -f, so a custom command can emit FLAC under any target_format.
The magic-byte check inside the wrapper is the authoritative test and costs a
26-byte peek. The output sample rate is read back out of the header ffmpeg just
wrote rather than taken from the transcode options, so a resampled (-ar) output
still gets the right count. Anything that is not a FLAC stream with an unset
total_samples passes through byte for byte.
Measured on a 177s source: before, total_samples=0 and ffprobe reported
duration N/A; after, total_samples=7807023 and duration 177.03s, with the audio
payload byte-identical. This affects every piped FLAC regardless of the source
format; only FLAC stores an authoritative "unknown", which is why mp3, opus
and aac survive the same pipe.
No SEEKTABLE is synthesised and the MD5 is left zero: both are optional, and
decoders binary-search using total_samples alone.
2026-09-07 14:19:45 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// setFLACTotalSamples takes the rate from the header rather than the transcode
|
|
|
|
|
// options, so a resampled (-ar) output still gets the right count.
|
|
|
|
|
func setFLACTotalSamples(prefix []byte, duration float32) {
|
2026-11-11 19:03:55 -05:00
|
|
|
if prefix[4]&0x7F != 0 { // the first metadata block must be STREAMINFO
|
fix(transcoding): make piped FLAC transcodes seekable
The FLAC muxer writes STREAMINFO before it knows the stream length, then
rewinds at the end to fill total_samples in. Navidrome pipes ffmpeg's stdout
(-f flac -), which is not seekable, so ffmpeg logs "unable to rewrite FLAC
header" and the field stays 0. A decoder needs total_samples to turn a
timestamp into a byte offset, so it reports an unknown duration and refuses to
seek. Online playback hides this because the client re-requests with a new
offset each time, but an offline copy is permanently unseekable, the symptom
reported against Symfonium where seeking a downloaded track jumps back to the
start.
Transcode now wraps its own output and rewrites total_samples as the first
bytes flow past. This lives in core/ffmpeg because the unseekable pipe is that
package's doing: buildDynamicArgs is what appends the trailing '-'. core/stream
only learns a target format and hands back an io.ReadCloser, so compensating
there leaked a transcoder implementation detail one layer up. TranscodeOptions
grows a Duration field alongside the existing Offset, which also puts the
duration-minus-offset arithmetic in the same function that emits -ss.
The wrapper runs on every transcode rather than only FLAC targets: the format
on a transcoding row is a declared target that nothing validates against the
command's actual -f, so a custom command can emit FLAC under any target_format.
The magic-byte check inside the wrapper is the authoritative test and costs a
26-byte peek. The output sample rate is read back out of the header ffmpeg just
wrote rather than taken from the transcode options, so a resampled (-ar) output
still gets the right count. Anything that is not a FLAC stream with an unset
total_samples passes through byte for byte.
Measured on a 177s source: before, total_samples=0 and ffprobe reported
duration N/A; after, total_samples=7807023 and duration 177.03s, with the audio
payload byte-identical. This affects every piped FLAC regardless of the source
format; only FLAC stores an authoritative "unknown", which is why mp3, opus
and aac survive the same pipe.
No SEEKTABLE is synthesised and the MD5 is left zero: both are optional, and
decoders binary-search using total_samples alone.
2026-09-07 14:19:45 -04:00
|
|
|
return
|
|
|
|
|
}
|
|
|
|
|
// 20-bit rate | 3-bit channels | 5-bit depth | 36-bit total_samples
|
|
|
|
|
info := binary.BigEndian.Uint64(prefix[18:])
|
|
|
|
|
rate := info >> 44
|
|
|
|
|
if rate == 0 || info&flacMaxTotalSamples != 0 {
|
|
|
|
|
return
|
|
|
|
|
}
|
|
|
|
|
total := math.Round(float64(duration) * float64(rate))
|
|
|
|
|
if total > flacMaxTotalSamples {
|
|
|
|
|
return
|
|
|
|
|
}
|
|
|
|
|
binary.BigEndian.PutUint64(prefix[18:], info|uint64(total))
|
|
|
|
|
}
|