Allows user to set maximum number of buffered packets when
probing a codec. It was a hard-coded parameter before this commit.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Despite the doxy stating that it's called when the muxer is destroyed,
this was not true in practice. It's only called by av_write_trailer()
and on init() failure.
An AVFormatContext may be closed without writing the trailer if errors
ocurred while muxing packets, so in order to prevent memory leaks, it
should effectively be called when freeing the muxer.
Signed-off-by: James Almer <jamrial@gmail.com>
add new function duration_estimate_name to dump duration estimate
method, it's will help to debug some duration issue.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
change the log level to warning if can't get duration, it's will help
to debug some duration issue
Signed-off-by: vacingfang <vacingfang@tencent.com>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
Up until now, read_frame_internal always initialized the packet it
received. But since the recent changes to ff_read_packet, this is no
longer needed: If the parsing queue is initially empty upon entering
read_frame_internal, the packet will now either contain content upon
success or be blank upon failure of ff_read_packet. If the parsing
queue is initially not empty, the packet will be overwritten with the
oldest one from the parsing queue.
Similarly, it is unnecessary to initialize ret in read_frame_internal.
In parse_packet, it is easily possible to only initialize the packet
used as temporary storage for the output if said packet is used at all;
furthermore, this packet doesn't need to be zero-initialized, because
av_init_packet will initialize every field except size and data and
those fields will be set by av_parser_parse2.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Up until now, parse_packet() used a stack packet in case the stream is
flushed. But using such a packet is unnecessary as there is an AVPacket
readily available, it just needs to be used. Whether flushing is intended
or not will now be signalled by an explicit parameter rather than by
whether the packet parameter is NULL. This removes a few checks in
parse_packet(), gets rid of the initialization of the stack packet and
also reduces usage of sizeof(AVPacket) in libavformat.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Up until now, read_frame_internal in avformat/utils.c uses a spare
packet on the stack that serves no real purpose: At no point in this
function is there a need for another packet besides the packet destined
for output:
1. If the packet doesn't need a parser, but is output as is, the content
of the spare packet (that at this point contains a freshly read packet)
is simply copied into the output packet (via simple assignment, not
av_packet_move_ref, thereby confusing ownership).
2. If the packet needs parsing, the spare packet will be reset after
parsing and any packets resulting from the packet read will be put into
a packet list; the output packet is not used here at all.
3. If the stream should be discarded, the spare packet will be
unreferenced; the output packet is not used here at all either.
Therefore the spare packet and the copies can be removed in principle.
In practice, one more thing needs to be taken care of: If ff_read_packet
failed, the output packet was not affected, now it is. But given that
ff_read_packet returns a blank (as if reset via av_packet_unref) packet
on failure, there is no problem from this side either.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
When AVFMT_FLAG_GENPTS is set, av_read_frame would put a reference to a
packet in the packet list (via av_packet_ref) and then immediately
thereafter unreference the original packet. This has been changed to
move the reference instead.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Up until now, avformat_find_stream_info had a potential for memleaks:
When everything was fine, it read packets and (depending upon whether
AVFMT_FLAG_NOBUFFER was set) put them in a packet list or unreferenced
them when they were no longer needed. But upon failure, said packets
would leak if they were not already on the packet list. This patch fixes
this.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Since the recent changes to ff_packet_list_put, the source packet will
be automatically reset when the reference is moved to the packet list,
so that it is unnecessary to reinitialize the packet in the loops in
parse_packet and ff_read_packet; initializing once at the beginning is
enough.
This also fixes a potential, but currently unexisting problem: If the
raw packet buffer was initially not empty and probe_codec() failed,
then the packet returned would not be initialized. But given that
probe_codec() currently can't fail (always returns 0) this was not an
acute danger.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
This is done so that its data is really owned by the packet.
This was already true for the current callers.
Signed-off-by: James Almer <jamrial@gmail.com>
Up until now, ff_packet_list_put had a flaw: When it moved a packet to
the list (meaning, when it ought to move the reference to the packet
list instead of creating a new one via av_packet_ref), it did not reset
the original packet, confusing the ownership of the data in the packet.
This has been done because some callers of this function were not
compatible with resetting the packet.
This commit changes these callers and fixes this flaw. In order to
indicate that the ownership of the packet has moved to the packet list,
pointers to constant AVPackets are used whenever the target of the
pointer might already be owned by the packet list.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
ff_read_packet had potential memleaks:
1. If av_packet_make_refcounted fails, it means that the packet is not
refcounted, but it could nevertheless carry side data and therefore
needs to be unreferenced.
2. If putting a packet on a packet list fails, it wasn't unreferenced.
Furthermore, read_frame_internal leaked a packet's (side) data if a
context update was required and failed.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
There is currently an ordinary check for this (which would lead to a
memleak), but given that no demuxer should ever return a packet with an
invalid stream_index it is more appropriate for this to be an assert.
FATE passes with this change.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
Several subtitle demuxers set negative durations
Fixes: signed integer overflow: 9223372036854775807 - -1 cannot be represented in type 'long'
Fixes: 16925/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5766519790764032
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
1. Instead of relying on ff_packet_list_get to get the oldest element in
an AVPacketList, ff_read_packet used its own ad-hoc code. Said code
forgot to set the end of the list to NULL if the last element of the
list has been removed, thereby leaving the list in an inconsistent state.
2. Furthermore, if the list was not empty, the oldest element of the
list would always be copied into another packet structure before it was
known whether the oldest entry of the list would be removed. This makes
the ownership confusing and potentially copies unnecessarily.
Both of these issues have been fixed. ff_packet_list_get is used now.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
avio_feof() returns true both in case of actual EOF and in case of IO errors.
Some demuxers (matroska) have special handling to be able to return the proper
error for this exact reason, e.g.:
if (avio_feof(pb)) {
if (pb->error) {
return pb->error;
} else {
return AVERROR_EOF;
}
}
However, most of the demuxers do not, and they simply return AVERROR_EOF if
avio_feof() is true, so there is a real chance that IO errors are mistaken for
EOF.
We might just say that the API user should always check the IO context error
attribute on EOF to make sure no IO errors happened, but not even ffmpeg.c does
this. It should be more intuitive to the API user if we simply return the IO
error as the return value of av_read_frame() instead of AVERROR_EOF.
Signed-off-by: Marton Balint <cus@passwd.hu>
fix error like "offset 0x1f85: partial file", the root cause is when
read the mp4 file from http, and the moov in the end of the mp4 file,
reconfig the buffer will drop some data.
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
Signed-off-by: Zhongxing Jiang <tomajsjiang@tencent.com>
Fixes: signed integer overflow: 9151595917793558550 + 297519050751678697 cannot be represented in type 'long'
Fixes: 15496/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5722866475073536
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: division by 0
Fixes: 15480/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5746727434321920
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This reverts commit dd06f022b0.
Fixes ticket #7871 and reopens ticket #7816.
The introduced regression caused URL's with @ in them to be parsed incorrectly
which is a bigger issue then not being able to specify the slash character as a
password.
I think there are better ways to fix the original issue, like being able to
specify HTTP username and password as a protocol option, or adding a protocol
option to percent-decode the URL first.
Signed-off-by: Marton Balint <cus@passwd.hu>
This patch improves the logs when the message "cur_dts is invalid" appears.
If helps to identify which stream generates the trouble,
and the status of the stream.
A lot of users suffers with the message, and the origin varies.
The improved message can help to discover the cause.
Signed-off-by: Andreas Hakon <andreas.hakon@protonmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
After this change we always parse the full specifier even if we know the result
in the middle of the parsing. Sligtly slower, but this is needed to
consistently reject incorrect specifiers in both matching and non-matching
cases.
Signed-off-by: Marton Balint <cus@passwd.hu>
This reworks the code to be more strict about accepting stream specifiers. From
now on we strictly enforce the syntax in the documentation up until the
decisive part of the stream specifier. Therefore matching stream specifiers
always need to be correct, non matching specifiers only need to be correct
until the decisive part.
Also recursion is changed to a simple loop.
Signed-off-by: Marton Balint <cus@passwd.hu>
This removes lots of code duplication and also allows more complex specifiers,
for example you can use p:204:aⓂ️language:eng to select the English language
audio stream from program 204.
Signed-off-by: Marton Balint <cus@passwd.hu>
Fixes: signed integer overflow: 7738135736989908991 - -7954308516317364223 cannot be represented in type 'long'
Fixes: find_stream_info_usan
Reported-by: Thomas Guilbert <tguilbert@google.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This avoids surprising developers. Its bad to surprise developers with
such unexpected things.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
We check for the documented explanation of the "Ignore code" in extract_extradata_check() already
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Because it will be used by avformat/segment.c or other module which
need to automatically create sub-directories operation.
Signed-off-by: Steven Liu <lq@chinaffmpeg.org>
This utility function creates 64-bit NTP time format as per the RFC
5905.
A simple explaination of 64-bit NTP time format is here
http://www.beaglesoft.com/Manual/page53.htm
These fields will allow the mpegts demuxer to expose details about
the PMT/program which created the AVProgram and its AVStreams.
In mpegts, a PMT which advertises streams has a version number
which can be incremented at any time. When the version changes,
the pids which correspond to each of it's streams can also change.
Since ffmpeg creates a new AVStream per pid by default, an API user
needs the ability to (a) detect when the PMT changed, and (b) tell
which AVStream were added to replace earlier streams.
This has been a long-standing issue with ffmpeg's handling of mpegts
streams with PMT changes, and I found two related patches in the wild
that attempt to solve the same problem:
The first is in MythTV's ffmpeg fork, where they added a
void (*streams_changed)(void*); to AVFormatContext and call it from
their fork of the mpegts demuxer whenever the PMT changes.
The second was proposed by XBMC in
https://ffmpeg.org/pipermail/ffmpeg-devel/2012-December/135036.html,
where they created a new AVMEDIA_TYPE_DATA stream with id=0 and
attempted to send packets to it whenever the PMT changed.
Signed-off-by: Aman Gupta <aman@tmm1.net>
For seekable mpegts streams, duration is calculated from
pts by seeking to the end of the file for a pts and subtracting
the initial pts to compute a duration.
This can be expensive in terms of added latency during
probe, especially when streaming over a network. This new
option lets you skip the duration calculation, which is useful
when you don't care about the value and want to save some overhead.
This patch is particularly useful when dealing with live mpegts
streams. Normally such streams are not seekable, so durations
are not calculated. However in my case I am dealing with a seekable
live mpegts stream (networked access to a .ts file which is still
being appended to).
Signed-off-by: Aman Gupta <aman@tmm1.net>
Similar to 4c9c4fe8b2, but for durations. This fixes#7151, where
the report duration and bitrate on a mpegts stream is wildly off
due to the dvb_teletext stream's timings.
Signed-off-by: Aman Gupta <aman@tmm1.net>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
If the parser returns full frames, then the output pointer retured by
av_parser_parse2() is guaranteed to point to data contained in the
input packet's buffer.
Create a new reference to said buffer in that case, to avoid
unnecessary data copy when queueing the packet later in the function.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>
