It could be cool to be able to give a Buffer of FileLike Object to probe and convert methods.

For example when you work with a Django FileField, you want to use filesystem abstract layer and just work with File object.

Hi,

I'm trying to probe video info through a web service, I have a webserver based on Flask + uwsgi where I upload a video and would like to return video info.

I use the simple info = c.probe("path/to/filename") which works perfect on command-line, but when I run it through the html call, I get an error I can't make anything of it:

Traceback (most recent call last):
  File "/usr/local/lib/python3.5/dist-packages/flask/app.py", line 1982, in wsgi_app
    response = self.full_dispatch_request()
  File "/usr/local/lib/python3.5/dist-packages/flask/app.py", line 1614, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/usr/local/lib/python3.5/dist-packages/flask/app.py", line 1517, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "/usr/local/lib/python3.5/dist-packages/flask/_compat.py", line 33, in reraise
    raise value
  File "/usr/local/lib/python3.5/dist-packages/flask/app.py", line 1612, in full_dispatch_request
    rv = self.dispatch_request()
  File "/usr/local/lib/python3.5/dist-packages/flask/app.py", line 1598, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "home.py", line 42, in decorated_function
    return f(*args, **kwargs)
  File "home.py", line 93, in uploaded_file
    info = c.probe(os.path.join(os.getcwd(),'uploads',filename))
  File "/usr/local/lib/python3.5/dist-packages/converter/__init__.py", line 224, in probe
    return self.ffmpeg.probe(fname, posters_as_video)
  File "/usr/local/lib/python3.5/dist-packages/converter/ffmpeg.py", line 387, in probe
    '-show_format', '-show_streams', fname])
  File "/usr/local/lib/python3.5/dist-packages/converter/ffmpeg.py", line 354, in _spawn
    close_fds=True)
  File "/usr/lib/python3.5/subprocess.py", line 947, in __init__
    restore_signals, start_new_session)
  File "/usr/lib/python3.5/subprocess.py", line 1484, in _execute_child
    self.pid = _posixsubprocess.fork_exec(
AttributeError: 'NoneType' object has no attribute 'fork_exec'

I already tried looking in current issues, #5 looked promising but solution applies to conversion, not probing.

I'm stuck. Your help is highly appreciated.

I got this error. What am I doing wrong?

File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/converter/__init__.py", line 162, in convert info = self.ffmpeg.probe(infile) File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/converter/ffmpeg.py", line 367, in probe info.parse_ffprobe(raw) File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/converter/ffmpeg.py", line 253, in parse_ffprobe self.format.parse_ffprobe(k, v) File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/converter/ffmpeg.py", line 78, in parse_ffprobe self.duration = float(val) ValueError: could not convert string to float: N/A

Hey,

I am using this code:

from converter import Converter
c = Converter()

conv = c.convert('input.mp4', 'video-transcoded.mkv', {
    'format': 'mkv',
    'audio': {
        'codec': 'ac3',
    },
    'video': {
        'codec': 'h264',
        'width': 1280,
        'height': 720,
        'quality': 25,
        'preset': 'slow'
    }
})

for timecode in conv:
    print "Converting (%f) ...\r" % timecode

And I get this error:

root@nas:/scripts/tools# python encode.test.py 
Traceback (most recent call last):
  File "encode.test.py", line 18, in <module>
    for timecode in conv:
  File "/usr/local/lib/python2.7/dist-packages/converter/__init__.py", line 184, in convert
    timeout=timeout):
  File "/usr/local/lib/python2.7/dist-packages/converter/ffmpeg.py", line 416, in convert
    p = self._spawn(cmds)
  File "/usr/local/lib/python2.7/dist-packages/converter/ffmpeg.py", line 334, in _spawn
    logger.debug('Spawning ffmpeg with command: ' + ' '.join(cmds))
TypeError: sequence item 14: expected string, int found
root@nas:/scripts/tools# 

And without quality option it works..

Please help me.

EDIT:
And I also debugged the _spawn method:
Spawning ffmpeg with command: /usr/bin/ffprobe -show_format -show_streams filename.mp4

How can I get the logger to work?

I have seen the source code, and the error occurs at line 318 of the ffmpeg.py file.

I just do not understand what I did wrong. I followed the installation instructions. Does anyone know how to fix this?

Thanks in advance.

I realized that I had to install ffmpeg separately, I did so from ffmpeg.org , but it still wants me to have an ffprobe, which is within the ffmpeg package I think, so I am trying to get it configured right.

Maybe add 1 line to the readme, stating that you need to install ffmpeg separately (helpful for noobs like me)? 👍

Actually I'm getting confused on how to import a folder with C programs into a python file.

Hope someone can show me some example code to get it setup.

Thanks. :)

I get the following error:

<FFMpegConvertError error="[libx264 @ 0x1fcb7c0] kb/s:4683.39", pid=30866, cmd="/usr/bin/ffmpeg -i /opt/test/media/2015/07/19/99756898-1509-48ba-84db-5b70ff2fd8d2 -acodec aac -ac 2 -ar 48000 -strict experimental -vcodec libx264 -r 30 -s 1920x1080 -aspect 1920:1080 -sn -f mp4 -y /opt/test/media/2015/07/19/99756898-1509-48ba-84db-5b70ff2fd8d2.mp4">

Even though the conversion works fine, I don't get any errors when I run it from the command line either.

The full traceback is...

Traceback (most recent call last):
  File "converter.py", line 2, in <module>
    from converter import Converter
  File "/home/souza/Desktop/converter.py", line 2, in <module>
    from converter import Converter
ImportError: cannot import name 'Converter' from 'converter' (/home/souza/Desktop/converter.py)

I installed using pip and the script is exactly the same as the one on README... Someone had an
idea

I'm trying to use the module on Windows 7 64bit with python 2.7 and get this error when trying to invoke the convert method of the Converter class -

File "C:\python27\lib\subprocess.py", line 637, in init
raise ValueError("close_fds is not supported on Windows "
ValueError: close_fds is not supported on Windows platforms if you redirect stdi
n/stdout/stderr

Is there a work around for this or is the module simply incompatible with windows?

Pasting the quick start example into a python file and running it raises a SyntaxError because the dict isn't properly closed.

$ python convert.py 
  File "convert.py", line 18
    })
     ^
SyntaxError: invalid syntax

While it has always been implicit that python-video-converter is open source and available under a permissive license, it was recently noted that we don’t say it explicitly anywhere, ie. the project doesn’t have an explicit license. As such, it can be argued it’s under all-rights-reserved copyright.

To fix this, I am proposing to (re)license to MIT, a popular open source license. The full text of the license would go into LICENSE.txt file and would read as this:

Copyright (C) 2011.-2017. Python Video Converter contributors

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

So I’m asking all contributors to the package - are you okay with this change? If so, please indicate in the comments / reactions to this GitHub issue.

Pinging @tahajahangir @mutschler @Kami @omidraha @KoleS46 @ouhouhsami @cepreu2github @legrisdev @marvin

I am new to python and came across this repo while searching for a good ffmpeg wrapper in python.

This one is better than all packages I found on pip. So just wondering why this is not distributed via pip?

Is there any technical reason for this?

I'm trying to run the converter using Celery. When executing the command I get this in the logfile:

[2014-05-15 13:01:50,184: INFO/Worker-1] [TRANSCODE] Retrieved transcoding task for file testvideo in 2014/05
[2014-05-15 13:01:50,184: INFO/Worker-1] [DO_TRANSCODE] Converting testvideo to flv 640x480
[2014-05-15 13:01:50,192: INFO/Worker-1] Spawning ffmpeg with command: /usr/local/bin/ffprobe -show_format -show_streams /home/storage/advideo/source/2014/05/testvideo
[2014-05-15 13:01:50,268: INFO/Worker-1] Spawning ffmpeg with command: /usr/local/bin/ffmpeg -probesize 2147483647 -analyzeduration 2147483647 -i /home/storage/advideo/source/2014/05/testvideo -acodec libmp3lame -ac 2 -ab 128k -ar 44100 -vcodec flv -r 18 -vb 1536k -s 640x360 -aspect 640:480 -vf pad=640:480:0:60 -sn -f flv -y /tmp/transcoding/91daeef2b6c245bcb3f23772d040552d.flv
[2014-05-15 13:01:50,301: ERROR/MainProcess] Task tasks.advideo[1cf57028-81f6-4776-bc5f-c02354c096a4] raised unexpected: FFMpegConvertError('Unknown ffmpeg error',)
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/celery/app/trace.py", line 240, in trace_task
R = retval = fun(_args, *_kwargs)
File "/usr/local/lib/python2.7/dist-packages/celery/app/trace.py", line 437, in protected_call
return self.run(_args, *_kwargs)
File "/data/tasks/tasks.py", line 81, in advideo
{ 'codec': 'flv', 'width': 640, 'height': 480, 'fps': 18, 'bitrate': 1536, 'mode': 'pad'}
File "/data/tasks/tasks.py", line 62, in dotranscode
for timecode in conv:
File "/usr/local/lib/python2.7/dist-packages/converter/init.py", line 213, in convert
timeout=timeout):
File "/usr/local/lib/python2.7/dist-packages/converter/ffmpeg.py", line 490, in convert
total_output, line, pid=p.pid)
FFMpegConvertError: Unknown ffmpeg error

If I do the ffmpeg command on the command line, there is no problem at al. I allready set the timeout to None

PS. I'm running Debian Wheezy. I've tried the same setup on my laptop with Linux Mint 16 without any problems. Could be a debian related problem with the process communication?

Please help.... ;-)

Is there any way of specifying custom options? Could not find something looking at the source.
I need to add -movflags +faststart

When video is recorded from iPhone, it has some metadata information of orientation. While creating the thumbnail from ffmpeg it doesn't take that into the consideration and creates thumbnail of different orientation.

It would be great if you read that orientation in thumbnail() and pass the rotating argument to ffmpeg accordingly

May be this snippet should work under thumbnails()

        probe = self.probe(fname)
        rotation = 0
        for stream in probe.streams:
            if stream.type == 'video' and stream.metadata.get('rotate'):
                rotation = int(stream.metadata.get('rotate'))
                break
        if rotation == 180:
            cmds.extend(["-vf", "vflip,hflip"])
        elif rotation == 90:
            cmds.extend(["-vf", "transpose=1"])
        elif rotation == 270 or rotation == -90:
            cmds.extend(["-vf", "transpose=2"])

Hi,

When converting very small videos (for instance 2-3 seconds), the exception "Unknown ffmpeg error" is raised, because the yielded variable did not have the time to be set to True (I think). But the video has correctly been converted!

Line 493 of ffmpeg.py:

if not yielded:
    raise FFMpegConvertError('Unknown ffmpeg error', cmd,
         total_output, line, pid=p.pid)

Running on freebsd. Compiled ffmpeg from source.

root@cringe:/usr/local/share/python-video-converter # python setup.py test
running test
.E........
======================================================================
ERROR: test_converter (__main__.TestFFMpeg)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/local/share/python-video-converter/test/test.py", line 297, in test_converter
    self.assertTrue(verify_progress(conv))
  File "/usr/local/share/python-video-converter/test/test.py", line 22, in verify_progress
    li = list(p)
  File "../converter/__init__.py", line 213, in convert
    timeout=timeout):
  File "../converter/ffmpeg.py", line 495, in convert
    total_output, line, pid=p.pid)
FFMpegConvertError: <FFMpegConvertError error="video:0kB audio:10kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 2.264493%", pid=60621, cmd="/usr/local/bin/ffmpeg -i test.aac -acodec libmp3lame -ac 1 -ab 32k -vn -sn -f mp3 -y /usr/local/share/python-video-converter/test/H6CUK34E50P1LGDCIV13/output.mp3">

----------------------------------------------------------------------
Ran 10 tests in 21.516s

FAILED (errors=1)
Traceback (most recent call last):
  File "setup.py", line 51, in <module>
    packages=['converter'],
  File "/usr/local/lib/python2.7/distutils/core.py", line 151, in setup
    dist.run_commands()
  File "/usr/local/lib/python2.7/distutils/dist.py", line 953, in run_commands
    self.run_command(cmd)
  File "/usr/local/lib/python2.7/distutils/dist.py", line 972, in run_command
    cmd_obj.run()
  File "setup.py", line 20, in run
    raise Exception('tests failed')
Exception: tests failed

Would be good if the MediaStreamInfo object had a language code.

Hi,

are there any plans to support multiple audio streams? that would be awesome.
I have many files with 2 or more audio streams, and curretly im only able to convert one of them (automatic selected?)

This is the following stack trace when I run the script:

Traceback (most recent call last):
  File "convo.py", line 2, in <module>
    c = Converter()
  File "C:\Users\Yash\AppData\Local\Programs\Python\Python38\lib\site-packages\converter\__init__.py", line 26, in __init__
    self.ffmpeg = FFMpeg(ffmpeg_path=ffmpeg_path,
  File "C:\Users\Yash\AppData\Local\Programs\Python\Python38\lib\site-packages\converter\ffmpeg.py", line 345, in __init__
    raise FFMpegError("ffmpeg binary not found: " + self.ffmpeg_path)
converter.ffmpeg.FFMpegError: ffmpeg binary not found: ffmpeg

c.thumbnail not working with latest ffmpeg build, and also raises errors when running the tests.

Changing the following in ffmpeg.py on line 415 seems to fix the issue:

• currently: 'sameq'
• changed to: '-q:v','0'

If you could please review, that would be appreciated.

Thanks

It would be good if this script would allow converting subtitles also. Example ffmpeg usage:

C:\ffmpeg.exe -i "D:\input.mkv" -vcodec copy -acodec copy -scodec mov_text -map 0 D:\output.mp4

MP4 output requires the mov_text codec, but other containers could use copy instead.

The option -map 0 also copies ALL streams instead of just the first of each. Perhaps you want to expand upon that as well as per http://ffmpeg.org/ffmpeg.html#Subtitle-options_003a so users can copy all subtitle streams.

To enable fast seeking we need to specify option -ss before input parameter -i. However in python-video-converter by default will use slow seeking because option -ss placed after input -i, see:

Here is my (dirty) implementation to enable fast seeking, rizkyabdilah@a17054c

Please review and consider implement fast seeking, thanks.

I wanted to convert video to mpeg4, but apparently python-video-converter doesn't have mpeg4.

Here is a STREAM section of poster image for a mp3 file, reporter by ffprobe:

[STREAM]
index=1
codec_name=mjpeg
codec_long_name=MJPEG (Motion JPEG)
profile=unknown
codec_type=video
codec_time_base=1/90000
codec_tag_string=[0][0][0][0]
codec_tag=0x0000
width=0
height=0
has_b_frames=0
sample_aspect_ratio=0:1
display_aspect_ratio=0:0
pix_fmt=unknown
level=-99
timecode=N/A
id=N/A
r_frame_rate=90000/1
avg_frame_rate=0/0
time_base=1/90000
start_pts=N/A
start_time=N/A
duration_ts=26818560
duration=297.984000
bit_rate=N/A
nb_frames=N/A
nb_read_frames=N/A
nb_read_packets=N/A
DISPOSITION:default=0
DISPOSITION:dub=0
DISPOSITION:original=0
DISPOSITION:comment=0
DISPOSITION:lyrics=0
DISPOSITION:karaoke=0
DISPOSITION:forced=0
DISPOSITION:hearing_impaired=0
DISPOSITION:visual_impaired=0
DISPOSITION:clean_effects=0
DISPOSITION:attached_pic=1
TAG:title=
TAG:comment=Cover (front)
[/STREAM]

setup.py doesn't contain a pypi license classifier and there's also no license file in the root of the tree, so I'm wondering what license the project is actually using.

When i was trying to covert a WMV file it shows this error

Traceback (most recent call last):
File "/virtualenvs/celery/local/lib/python2.7/site-packages/celery/app/trace.py", line 240, in trace_task
R = retval = fun(_args, *_kwargs)
File "/virtualenvs/celery/local/lib/python2.7/site-packages/celery/app/trace.py", line 437, in protected_call
return self.run(_args, *_kwargs)
File "/virtualenvs/celery/inviterjobs/workers/finalvideo/tasks.py", line 38, in finalvideo
for timecode in converter_convert:
File "/virtualenvs/celery/local/lib/python2.7/site-packages/converter/ffmpeg.py", line 453, in convert
ret = ret.decode(console_encoding)
File "/virtualenvs/celery/lib/python2.7/encodings/utf_8.py", line 16, in decode
return codecs.utf_8_decode(input, errors, True)
UnicodeDecodeError: 'utf8' codec can't decode byte 0xc2 in position 9: unexpected end of data

This module, as excellent as it is, does not seem to allow you to convert videos if the conversation is kicked off in a child thread.

Having done some digging, it seems to be down to the use of Pythons signals library, whose docs state: "only the main thread can set a new signal handler".

It would be nice if the module supported threads, so that multiple videos could be converted simultaneously.

I found out the timeout in ffmpeg.convert have no meaning, because the function signal.alarm is in the While True loop.

Hi,
I am running the python video converter library without multi threading inside Django. It's working fine. but while Enabling the multi threading It's giving error.

ValueError: signal only works in main thread

OS- ubuntu18.04
Python version - 3.6

Thanks for your Time