Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Jun Zhao via ffmpeg-devel <ffmpeg-devel@ffmpeg.org>
To: ffmpeg-devel@ffmpeg.org
Cc: Jun Zhao <code@ffmpeg.org>
Subject: [FFmpeg-devel] [PR] lavf/aacdec: improve probe for AAC files with ID3v2 tags (PR #21472)
Date: Thu, 15 Jan 2026 05:30:44 -0000
Message-ID: <176845504547.25.17781000858282546285@4457048688e7> (raw)

PR #21472 opened by Jun Zhao (mypopydev)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/21472
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/21472.patch

The AAC probe was returning low score (1) for valid ADTS AAC files
starting with ID3v2 tags, common in HLS streams, triggering:
"Format aac detected only with low score of 1, misdetection possible!"

Fix by skipping ID3v2 tags at the beginning and between ADTS frames
during probing, similar to mp3_read_probe() in mp3dec.c.

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>


>From 1b222f5de992ccf36f1cc3573180286d3a4cd068 Mon Sep 17 00:00:00 2001
From: Jun Zhao <barryjzhao@tencent.com>
Date: Thu, 15 Jan 2026 12:54:22 +0800
Subject: [PATCH] lavf/aacdec: improve probe for AAC files with ID3v2 tags

The AAC probe was returning low score (1) for valid ADTS AAC files
starting with ID3v2 tags, common in HLS streams, triggering:
"Format aac detected only with low score of 1, misdetection possible!"

Fix by skipping ID3v2 tags at the beginning and between ADTS frames
during probing, similar to mp3_read_probe() in mp3dec.c.

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
---
 libavformat/aacdec.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/libavformat/aacdec.c b/libavformat/aacdec.c
index fef3c69f0b..b87619ec7c 100644
--- a/libavformat/aacdec.c
+++ b/libavformat/aacdec.c
@@ -40,15 +40,39 @@ static int adts_aac_probe(const AVProbeData *p)
     const uint8_t *buf2;
     const uint8_t *buf;
     const uint8_t *end = buf0 + p->buf_size - 7;
+    int id3_skip = 0;
 
     buf = buf0;
 
+    // Skip ID3v2 tags at the beginning to find the actual ADTS data
+    while (buf + ID3v2_HEADER_SIZE < end &&
+           ff_id3v2_match(buf, ID3v2_DEFAULT_MAGIC)) {
+        int taglen = ff_id3v2_tag_len(buf);
+        if (id3_skip + taglen > p->buf_size)
+            break;
+        id3_skip += taglen;
+        buf = p->buf + id3_skip;
+    }
+
+    // Update buf0 to point after ID3v2 tags for first_frames calculation
+    buf0 = buf;
+
     for (; buf < end; buf = buf2 + 1) {
         buf2 = buf;
 
         for (frames = 0; buf2 < end; frames++) {
             uint32_t header = AV_RB16(buf2);
             if ((header & 0xFFF6) != 0xFFF0) {
+                // Skip embedded ID3v2 tags (common in HLS AAC streams)
+                if (buf2 + ID3v2_HEADER_SIZE < end &&
+                    ff_id3v2_match(buf2, ID3v2_DEFAULT_MAGIC)) {
+                    int id3len = ff_id3v2_tag_len(buf2);
+                    if (buf2 + id3len < end) {
+                        buf2 += id3len;
+                        frames--;  // compensate for the frames++ in the loop
+                        continue;
+                    }
+                }
                 if (buf != buf0) {
                     // Found something that isn't an ADTS header, starting
                     // from a position other than the start of the buffer.
@@ -75,6 +99,9 @@ static int adts_aac_probe(const AVProbeData *p)
         return AVPROBE_SCORE_EXTENSION;
     else if (max_frames >= 3)
         return AVPROBE_SCORE_EXTENSION / 2;
+    // If file starts with ID3v2 and we found ADTS frames after it, boost score
+    else if (id3_skip > 0 && max_frames >= 1)
+        return AVPROBE_SCORE_EXTENSION / 2;
     else if (first_frames >= 1)
         return 1;
     else
-- 
2.49.1

_______________________________________________
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.org
To unsubscribe send an email to ffmpeg-devel-leave@ffmpeg.org

                 reply	other threads:[~2026-01-15  5:32 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=176845504547.25.17781000858282546285@4457048688e7 \
    --to=ffmpeg-devel@ffmpeg.org \
    --cc=code@ffmpeg.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
		ffmpegdev@gitmailbox.com
	public-inbox-index ffmpegdev

Example config snippet for mirrors.


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git