Some ffmpeg filters drop audio streams, and care must be taken
+to preserve the audio in the final output. The .audio and
+.video operators can be used to reference the audio/video
+portions of a stream so that they can be processed separately
+and then re-combined later in the pipeline. This dilemma is
+intrinsic to ffmpeg, and ffmpeg-python tries to stay out of the
+way while users may refer to the official ffmpeg documentation
+as to why certain filters drop audio.
+
stream.audio is a shorthand for stream['a'].
+
Example
+
Process the audio and video portions of a stream independently:
Some ffmpeg filters drop audio streams, and care must be taken
+to preserve the audio in the final output. The .audio and
+.video operators can be used to reference the audio/video
+portions of a stream so that they can be processed separately
+and then re-combined later in the pipeline. This dilemma is
+intrinsic to ffmpeg, and ffmpeg-python tries to stay out of the
+way while users may refer to the official ffmpeg documentation
+as to why certain filters drop audio.
+
stream.video is a shorthand for stream['v'].
+
Example
+
Process the audio and video portions of a stream independently:
Any supplied keyword arguments are passed to ffmpeg verbatim (e.g.
+t=20, f='mp4', acodec='pcm', vcodec='rawvideo',
+etc.). Some keyword-arguments are handled specially, as shown below.
+
+
Parameters
+
+
video_bitrate – parameter for -b:v, e.g. video_bitrate=1000.
+
audio_bitrate – parameter for -b:a, e.g. audio_bitrate=200.
+
format – alias for -f parameter, e.g. format='mp4'
+(equivalent to f='mp4').
+
+
+
+
If multiple streams are provided, they are mapped to the same
+output.
+
To tell ffmpeg to write to stdout, use pipe: as the filename.
Run ffprobe on the specified file and return a JSON representation of the output.
+
+
Raises
+
ffmpeg.Error – if ffprobe returns a non-zero exit code,
+ an Error is returned with a generic error message.
+ The stderr output can be retrieved by accessing the
+ stderr property of the exception.
The run() function uses this to build the command line
+arguments and should work in most cases, but calling this function
+directly is useful for debugging or if you need to invoke ffmpeg
+manually for whatever reason.
+
This is the same as calling get_args() except that it also
+includes the ffmpeg command as the first argument.
Concatenate audio and video streams, joining them together one after the other.
The filter works on segments of synchronized video and audio streams. All segments must have the same number of
streams of each type, and that will also be the number of streams at output.
-
-
-
-
-
Parameters:
unsafe – Activate unsafe mode: do not fail if segments have a different format.
-
-
-
+
+
Parameters
+
unsafe – Activate unsafe mode: do not fail if segments have a different format.
+
+
Related streams do not always have exactly the same duration, for various reasons including codec frame size or
sloppy authoring. For that reason, related synchronized streams (e.g. a video and its audio track) should be
concatenated at once. The concat filter will use the duration of the longest stream in each segment (except the
@@ -82,196 +330,182 @@
-ffmpeg.drawbox(stream, x, y, width, height, color, thickness=None, **kwargs)¶
+ffmpeg.drawbox(stream, x, y, width, height, color, thickness=None, **kwargs)¶
Draw a colored box on the input image.
-
-
-
-
-
Parameters:
-
x – The expression which specifies the top left corner x coordinate of the box. It defaults to 0.
-
y – The expression which specifies the top left corner y coordinate of the box. It defaults to 0.
-
width – Specify the width of the box; if 0 interpreted as the input width. It defaults to 0.
-
heigth – Specify the height of the box; if 0 interpreted as the input height. It defaults to 0.
-
color – Specify the color of the box to write. For the general syntax of this option, check the “Color” section
+
+
Parameters
+
+
x – The expression which specifies the top left corner x coordinate of the box. It defaults to 0.
+
y – The expression which specifies the top left corner y coordinate of the box. It defaults to 0.
+
width – Specify the width of the box; if 0 interpreted as the input width. It defaults to 0.
+
height – Specify the height of the box; if 0 interpreted as the input height. It defaults to 0.
+
color – Specify the color of the box to write. For the general syntax of this option, check the “Color” section
in the ffmpeg-utils manual. If the special value invert is used, the box edge color is the same as the
-video with inverted luma.
-
thickness – The expression which sets the thickness of the box edge. Default value is 3.
-
w – Alias for width.
-
h – Alias for height.
-
c – Alias for color.
-
t – Alias for thickness.
+video with inverted luma.
+
thickness – The expression which sets the thickness of the box edge. Default value is 3.
Draw a text string or text from a specified file on top of a video, using the libfreetype library.
To enable compilation of this filter, you need to configure FFmpeg with --enable-libfreetype. To enable default
font fallback and the font option you need to configure FFmpeg with --enable-libfontconfig. To enable the
text_shaping option, you need to configure FFmpeg with --enable-libfribidi.
-
-
-
-
-
Parameters:
-
box – Used to draw a box around text using the background color. The value must be either 1 (enable) or 0
-(disable). The default value of box is 0.
-
boxborderw – Set the width of the border to be drawn around the box using boxcolor. The default value of
-boxborderw is 0.
-
boxcolor – The color to be used for drawing box around text. For the syntax of this option, check the “Color”
-section in the ffmpeg-utils manual. The default value of boxcolor is “white”.
-
line_spacing – Set the line spacing in pixels of the border to be drawn around the box using box. The default
-value of line_spacing is 0.
-
borderw – Set the width of the border to be drawn around the text using bordercolor. The default value of
-borderw is 0.
-
bordercolor – Set the color to be used for drawing border around text. For the syntax of this option, check the
-“Color” section in the ffmpeg-utils manual. The default value of bordercolor is “black”.
-
expansion – Select how the text is expanded. Can be either none, strftime (deprecated) or normal (default). See
-the Text expansion section below for details.
-
basetime – Set a start time for the count. Value is in microseconds. Only applied in the deprecated strftime
+
+
Parameters
+
+
box – Used to draw a box around text using the background color. The value must be either 1 (enable) or 0
+(disable). The default value of box is 0.
+
boxborderw – Set the width of the border to be drawn around the box using boxcolor. The default value of
+boxborderw is 0.
+
boxcolor – The color to be used for drawing box around text. For the syntax of this option, check the “Color”
+section in the ffmpeg-utils manual. The default value of boxcolor is “white”.
+
line_spacing – Set the line spacing in pixels of the border to be drawn around the box using box. The default
+value of line_spacing is 0.
+
borderw – Set the width of the border to be drawn around the text using bordercolor. The default value of
+borderw is 0.
+
bordercolor – Set the color to be used for drawing border around text. For the syntax of this option, check the
+“Color” section in the ffmpeg-utils manual. The default value of bordercolor is “black”.
+
expansion – Select how the text is expanded. Can be either none, strftime (deprecated) or normal (default). See
+the Text expansion section below for details.
+
basetime – Set a start time for the count. Value is in microseconds. Only applied in the deprecated strftime
expansion mode. To emulate in normal expansion mode use the pts function, supplying the start time (in
-seconds) as the second argument.
-
fix_bounds – If true, check and fix text coords to avoid clipping.
-
fontcolor – The color to be used for drawing fonts. For the syntax of this option, check the “Color” section in
-the ffmpeg-utils manual. The default value of fontcolor is “black”.
-
fontcolor_expr – String which is expanded the same way as text to obtain dynamic fontcolor value. By default
-this option has empty value and is not processed. When this option is set, it overrides fontcolor option.
-
font – The font family to be used for drawing text. By default Sans.
-
fontfile – The font file to be used for drawing text. The path must be included. This parameter is mandatory if
-the fontconfig support is disabled.
-
alpha – Draw the text applying alpha blending. The value can be a number between 0.0 and 1.0. The expression
-accepts the same variables x, y as well. The default value is 1. Please see fontcolor_expr.
-
fontsize – The font size to be used for drawing text. The default value of fontsize is 16.
-
text_shaping – If set to 1, attempt to shape the text (for example, reverse the order of right-to-left text and
+seconds) as the second argument.
+
fix_bounds – If true, check and fix text coords to avoid clipping.
+
fontcolor – The color to be used for drawing fonts. For the syntax of this option, check the “Color” section in
+the ffmpeg-utils manual. The default value of fontcolor is “black”.
+
fontcolor_expr – String which is expanded the same way as text to obtain dynamic fontcolor value. By default
+this option has empty value and is not processed. When this option is set, it overrides fontcolor option.
+
font – The font family to be used for drawing text. By default Sans.
+
fontfile – The font file to be used for drawing text. The path must be included. This parameter is mandatory if
+the fontconfig support is disabled.
+
alpha – Draw the text applying alpha blending. The value can be a number between 0.0 and 1.0. The expression
+accepts the same variables x, y as well. The default value is 1. Please see fontcolor_expr.
+
fontsize – The font size to be used for drawing text. The default value of fontsize is 16.
+
text_shaping – If set to 1, attempt to shape the text (for example, reverse the order of right-to-left text and
join Arabic characters) before drawing it. Otherwise, just draw the text exactly as given. By default 1 (if
-supported).
-
ft_load_flags –
The flags to be used for loading the fonts. The flags map the corresponding flags supported by
+supported).
+
ft_load_flags –
The flags to be used for loading the fonts. The flags map the corresponding flags supported by
libfreetype, and are a combination of the following values:
-
default
-
no_scale
-
no_hinting
-
render
-
no_bitmap
-
vertical_layout
-
force_autohint
-
crop_bitmap
-
pedantic
-
ignore_global_advance_width
-
no_recurse
-
ignore_transform
-
monochrome
-
linear_design
-
no_autohint
+
default
+
no_scale
+
no_hinting
+
render
+
no_bitmap
+
vertical_layout
+
force_autohint
+
crop_bitmap
+
pedantic
+
ignore_global_advance_width
+
no_recurse
+
ignore_transform
+
monochrome
+
linear_design
+
no_autohint
Default value is “default”. For more information consult the documentation for the FT_LOAD_* libfreetype
flags.
-
-
shadowcolor – The color to be used for drawing a shadow behind the drawn text. For the syntax of this option,
-check the “Color” section in the ffmpeg-utils manual. The default value of shadowcolor is “black”.
-
shadowx – The x offset for the text shadow position with respect to the position of the text. It can be either
-positive or negative values. The default value is “0”.
-
shadowy – The y offset for the text shadow position with respect to the position of the text. It can be either
-positive or negative values. The default value is “0”.
-
start_number – The starting frame number for the n/frame_num variable. The default value is “0”.
-
tabsize – The size in number of spaces to use for rendering the tab. Default value is 4.
-
timecode – Set the initial timecode representation in “hh:mm:ss[:;.]ff” format. It can be used with or without
-text parameter. timecode_rate option must be specified.
-
rate – Set the timecode frame rate (timecode only).
-
timecode_rate – Alias for rate.
-
r – Alias for rate.
-
tc24hmax – If set to 1, the output of the timecode option will wrap around at 24 hours. Default is 0 (disabled).
-
text – The text string to be drawn. The text must be a sequence of UTF-8 encoded characters. This parameter is
-mandatory if no file is specified with the parameter textfile.
-
textfile – A text file containing text to be drawn. The text must be a sequence of UTF-8 encoded characters.
+
+
shadowcolor – The color to be used for drawing a shadow behind the drawn text. For the syntax of this option,
+check the “Color” section in the ffmpeg-utils manual. The default value of shadowcolor is “black”.
+
shadowx – The x offset for the text shadow position with respect to the position of the text. It can be either
+positive or negative values. The default value is “0”.
+
shadowy – The y offset for the text shadow position with respect to the position of the text. It can be either
+positive or negative values. The default value is “0”.
+
start_number – The starting frame number for the n/frame_num variable. The default value is “0”.
+
tabsize – The size in number of spaces to use for rendering the tab. Default value is 4.
+
timecode – Set the initial timecode representation in “hh:mm:ss[:;.]ff” format. It can be used with or without
+text parameter. timecode_rate option must be specified.
+
rate – Set the timecode frame rate (timecode only).
+
timecode_rate – Alias for rate.
+
r – Alias for rate.
+
tc24hmax – If set to 1, the output of the timecode option will wrap around at 24 hours. Default is 0 (disabled).
+
text – The text string to be drawn. The text must be a sequence of UTF-8 encoded characters. This parameter is
+mandatory if no file is specified with the parameter textfile.
+
textfile – A text file containing text to be drawn. The text must be a sequence of UTF-8 encoded characters.
This parameter is mandatory if no text string is specified with the parameter text. If both text and
-textfile are specified, an error is thrown.
-
reload – If set to 1, the textfile will be reloaded before each frame. Be sure to update it atomically, or it
-may be read partially, or even fail.
-
x – The expression which specifies the offset where text will be drawn within the video frame. It is relative to
-the left border of the output image. The default value is “0”.
-
y – The expression which specifies the offset where text will be drawn within the video frame. It is relative to
+textfile are specified, an error is thrown.
+
reload – If set to 1, the textfile will be reloaded before each frame. Be sure to update it atomically, or it
+may be read partially, or even fail.
+
x – The expression which specifies the offset where text will be drawn within the video frame. It is relative to
+the left border of the output image. The default value is “0”.
+
y – The expression which specifies the offset where text will be drawn within the video frame. It is relative to
the top border of the output image. The default value is “0”. See below for the list of accepted constants
-and functions.
+and functions.
-
-
-
-
-
-
Expression constants:
-
-
The parameters for x and y are expressions containing the following constants and functions:
-
-
dar: input display aspect ratio, it is the same as (w/h)*sar
-
hsub: horizontal chroma subsample values. For example for the pixel format “yuv422p” hsub is 2 and vsub
-is 1.
-
vsub: vertical chroma subsample values. For example for the pixel format “yuv422p” hsub is 2 and vsub
-is 1.
-
line_h: the height of each text line
-
lh: Alias for line_h.
-
main_h: the input height
-
h: Alias for main_h.
-
H: Alias for main_h.
-
main_w: the input width
-
w: Alias for main_w.
-
W: Alias for main_w.
-
ascent: the maximum distance from the baseline to the highest/upper grid coordinate used to place a glyph
+
+
+
+
Expression constants:
+
The parameters for x and y are expressions containing the following constants and functions:
+
dar: input display aspect ratio, it is the same as (w/h)*sar
+
hsub: horizontal chroma subsample values. For example for the pixel format “yuv422p” hsub is 2 and vsub
+is 1.
+
vsub: vertical chroma subsample values. For example for the pixel format “yuv422p” hsub is 2 and vsub
+is 1.
+
line_h: the height of each text line
+
lh: Alias for line_h.
+
main_h: the input height
+
h: Alias for main_h.
+
H: Alias for main_h.
+
main_w: the input width
+
w: Alias for main_w.
+
W: Alias for main_w.
+
ascent: the maximum distance from the baseline to the highest/upper grid coordinate used to place a glyph
outline point, for all the rendered glyphs. It is a positive value, due to the grid’s orientation with the Y
-axis upwards.
-
max_glyph_a: Alias for ascent.
-
descent: the maximum distance from the baseline to the lowest grid coordinate used to place a glyph outline
+axis upwards.
+
max_glyph_a: Alias for ascent.
+
descent: the maximum distance from the baseline to the lowest grid coordinate used to place a glyph outline
point, for all the rendered glyphs. This is a negative value, due to the grid’s orientation, with the Y axis
-upwards.
-
max_glyph_d: Alias for descent.
-
max_glyph_h: maximum glyph height, that is the maximum height for all the glyphs contained in the rendered
-text, it is equivalent to ascent - descent.
-
max_glyph_w: maximum glyph width, that is the maximum width for all the glyphs contained in the rendered
-text.
-
n: the number of input frame, starting from 0
-
rand(min, max): return a random number included between min and max
-
sar: The input sample aspect ratio.
-
t: timestamp expressed in seconds, NAN if the input timestamp is unknown
-
text_h: the height of the rendered text
-
th: Alias for text_h.
-
text_w: the width of the rendered text
-
tw: Alias for text_w.
-
x: the x offset coordinates where the text is drawn.
-
y: the y offset coordinates where the text is drawn.
+upwards.
+
max_glyph_d: Alias for descent.
+
max_glyph_h: maximum glyph height, that is the maximum height for all the glyphs contained in the rendered
+text, it is equivalent to ascent - descent.
+
max_glyph_w: maximum glyph width, that is the maximum width for all the glyphs contained in the rendered
+text.
+
n: the number of input frame, starting from 0
+
rand(min, max): return a random number included between min and max
+
sar: The input sample aspect ratio.
+
t: timestamp expressed in seconds, NAN if the input timestamp is unknown
+
text_h: the height of the rendered text
+
th: Alias for text_h.
+
text_w: the width of the rendered text
+
tw: Alias for text_w.
+
x: the x offset coordinates where the text is drawn.
+
y: the y offset coordinates where the text is drawn.
-
These parameters allow the x and y expressions to refer each other, so you can for example specify
+
These parameters allow the x and y expressions to refer each other, so you can for example specify
y=x/dar.
filter_ is normally used by higher-level filter functions such as hflip, but if a filter implementation
-is missing from fmpeg-python, you can call filter_ directly to have fmpeg-python pass the filter name
+is missing from ffmpeg-python, you can call filter_ directly to have ffmpeg-python pass the filter name
and arguments to ffmpeg verbatim.
-
-
-
-
-
Parameters:
-
stream_spec – a Stream, list of Streams, or label-to-Stream dictionary mapping
-
filter_name – ffmpeg filter name, e.g. colorchannelmixer
-
*args – list of args to pass to ffmpeg verbatim
-
**kwargs – list of keyword-args to pass to ffmpeg verbatim
+
+
Parameters
+
+
stream_spec – a Stream, list of Streams, or label-to-Stream dictionary mapping
+
filter_name – ffmpeg filter name, e.g. colorchannelmixer
+
*args – list of args to pass to ffmpeg verbatim
+
**kwargs – list of keyword-args to pass to ffmpeg verbatim
-
-
-
-
+
+
The function name is suffixed with _ in order avoid confusion with the standard python filter function.
x – Set the expression for the x coordinates of the overlaid video on the main video. Default value is 0. In
+
+
Parameters
+
+
x – Set the expression for the x coordinates of the overlaid video on the main video. Default value is 0. In
case the expression is invalid, it is set to a huge value (meaning that the overlay will not be displayed
-within the output visible area).
-
y – Set the expression for the y coordinates of the overlaid video on the main video. Default value is 0. In
+within the output visible area).
+
y – Set the expression for the y coordinates of the overlaid video on the main video. Default value is 0. In
case the expression is invalid, it is set to a huge value (meaning that the overlay will not be displayed
-within the output visible area).
-
eof_action –
The action to take when EOF is encountered on the secondary input; it accepts one of the following
+within the output visible area).
+
eof_action –
The action to take when EOF is encountered on the secondary input; it accepts one of the following
values:
-
repeat: Repeat the last frame (the default).
-
endall: End both streams.
-
pass: Pass the main input through.
+
repeat: Repeat the last frame (the default).
+
endall: End both streams.
+
pass: Pass the main input through.
-
-
eval –
Set when the expressions for x, and y are evaluated.
+
+
eval –
Set when the expressions for x, and y are evaluated.
It accepts the following values:
-
-
init: only evaluate expressions once during the filter initialization or when a command is
-
processed
+
+
init: only evaluate expressions once during the filter initialization or when a command is
processed
+
-
frame: evaluate expressions for each incoming frame
+
frame: evaluate expressions for each incoming frame
Default value is frame.
-
-
shortest – If set to 1, force the output to terminate when the shortest input terminates. Default value is 0.
-
format –
Set the format for the output video.
+
+
shortest – If set to 1, force the output to terminate when the shortest input terminates. Default value is 0.
+
format –
Set the format for the output video.
It accepts the following values:
-
yuv420: force YUV420 output
-
yuv422: force YUV422 output
-
yuv444: force YUV444 output
-
rgb: force packed RGB output
-
gbrp: force planar RGB output
+
yuv420: force YUV420 output
+
yuv422: force YUV422 output
+
yuv444: force YUV444 output
+
rgb: force packed RGB output
+
gbrp: force planar RGB output
Default value is yuv420.
-
-
rgb (deprecated) – If set to 1, force the filter to accept inputs in the RGB color space. Default value is 0.
-This option is deprecated, use format instead.
-
repeatlast – If set to 1, force the filter to draw the last overlay frame over the main input until the end of
-the stream. A value of 0 disables this behavior. Default value is 1.
+
+
rgb (deprecated) – If set to 1, force the filter to accept inputs in the RGB color space. Default value is 0.
+This option is deprecated, use format instead.
+
repeatlast – If set to 1, force the filter to draw the last overlay frame over the main input until the end of
+the stream. A value of 0 disables this behavior. Default value is 1.
Trim the input so that the output contains one continuous subpart of the input.
-
-
-
-
-
Parameters:
-
start – Specify the time of the start of the kept section, i.e. the frame with the timestamp start will be the
-first frame in the output.
-
end – Specify the time of the first frame that will be dropped, i.e. the frame immediately preceding the one
-with the timestamp end will be the last frame in the output.
-
start_pts – This is the same as start, except this option sets the start timestamp in timebase units instead of
-seconds.
-
end_pts – This is the same as end, except this option sets the end timestamp in timebase units instead of
-seconds.
-
duration – The maximum duration of the output in seconds.
-
start_frame – The number of the first frame that should be passed to the output.
-
end_frame – The number of the first frame that should be dropped.
+
+
Parameters
+
+
start – Specify the time of the start of the kept section, i.e. the frame with the timestamp start will be the
+first frame in the output.
+
end – Specify the time of the first frame that will be dropped, i.e. the frame immediately preceding the one
+with the timestamp end will be the last frame in the output.
+
start_pts – This is the same as start, except this option sets the start timestamp in timebase units instead of
+seconds.
+
end_pts – This is the same as end, except this option sets the end timestamp in timebase units instead of
+seconds.
+
duration – The maximum duration of the output in seconds.
+
start_frame – The number of the first frame that should be passed to the output.
+
end_frame – The number of the first frame that should be dropped.
Any supplied keyword arguments are passed to ffmpeg verbatim (e.g.
-t=20, f='mp4', acodec='pcm', vcodec='rawvideo',
-etc.). Some keyword-arguments are handled specially, as shown below.
-
-
-
-
-
Parameters:
-
video_bitrate – parameter for -b:v, e.g. video_bitrate=1000.
-
audio_bitrate – parameter for -b:a, e.g. audio_bitrate=200.
-
format – alias for -f parameter, e.g. format='mp4'
-(equivalent to f='mp4').
-
-
-
-
-
-
If multiple streams are provided, they are mapped to the same
-output.
-
To tell ffmpeg to write to stdout, use pipe: as the filename.
The run() function uses this to build the commnad line
-arguments and should work in most cases, but calling this function
-directly is useful for debugging or if you need to invoke ffmpeg
-manually for whatever reason.
-
This is the same as calling get_args() except that it also
-includes the ffmpeg command as the first argument.
Run ffprobe on the specified file and return a JSON representation of the output.
-
-
-
-
-
Raises:
ffmpeg.Error – if ffprobe returns a non-zero exit code,
-an Error is returned with a generic error message.
-The stderr output can be retrieved by accessing the
-stderr property of the exception.
\ No newline at end of file
diff --git a/doc/html/objects.inv b/doc/html/objects.inv
index e5afeacf..053a659f 100644
Binary files a/doc/html/objects.inv and b/doc/html/objects.inv differ
diff --git a/doc/html/py-modindex.html b/doc/html/py-modindex.html
index ac323611..00af8197 100644
--- a/doc/html/py-modindex.html
+++ b/doc/html/py-modindex.html
@@ -1,18 +1,17 @@
-
+
-
-
+
Python Module Index — ffmpeg-python documentation
-
+
+
@@ -66,13 +65,11 @@
Python Module Index
-
Quick search
+
Quick search
@@ -95,7 +92,7 @@
Navigation
\ No newline at end of file
diff --git a/doc/html/search.html b/doc/html/search.html
index a5ad01f3..cdc885db 100644
--- a/doc/html/search.html
+++ b/doc/html/search.html
@@ -1,26 +1,22 @@
-
+
-
-
+
Search — ffmpeg-python documentation
-
+
+
+
-
-
-
+
@@ -89,7 +85,7 @@
Navigation
\ No newline at end of file
diff --git a/doc/html/searchindex.js b/doc/html/searchindex.js
index 06ef010e..a8825d29 100644
--- a/doc/html/searchindex.js
+++ b/doc/html/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["index"],envversion:53,filenames:["index.rst"],objects:{"":{ffmpeg:[0,0,0,"-"]},ffmpeg:{Error:[0,1,1,""],colorchannelmixer:[0,2,1,""],compile:[0,2,1,""],concat:[0,2,1,""],crop:[0,2,1,""],drawbox:[0,2,1,""],drawtext:[0,2,1,""],filter:[0,2,1,""],filter_:[0,2,1,""],filter_multi_output:[0,2,1,""],get_args:[0,2,1,""],hflip:[0,2,1,""],hue:[0,2,1,""],input:[0,2,1,""],merge_outputs:[0,2,1,""],output:[0,2,1,""],overlay:[0,2,1,""],overwrite_output:[0,2,1,""],probe:[0,2,1,""],run:[0,2,1,""],setpts:[0,2,1,""],trim:[0,2,1,""],vflip:[0,2,1,""],view:[0,2,1,""],zoompan:[0,2,1,""]}},objnames:{"0":["py","module","Python module"],"1":["py","exception","Python exception"],"2":["py","function","Python function"]},objtypes:{"0":"py:module","1":"py:exception","2":"py:function"},terms:{"case":0,"default":0,"function":0,"return":0,"true":0,For:0,PTS:0,The:0,These:0,Used:0,accept:0,access:0,acodec:0,action:0,activ:0,adjust:0,after:0,alia:0,all:0,allow:0,alpha:0,also:0,altern:0,alwai:0,angl:0,ani:0,anoth:0,appli:0,arab:0,area:0,arg:0,argument:0,around:0,ascent:0,asetpt:0,ask:0,aspect:0,atom:0,attempt:0,audio:0,audio_bitr:0,author:0,automat:0,avoid:0,axi:0,background:0,base:0,baselin:0,basetim:0,befor:0,behavior:0,behind:0,below:0,between:0,black:0,blend:0,border:0,bordercolor:0,borderw:0,both:0,box:0,boxborderw:0,boxcolor:0,bracket:0,bright:0,build:0,built:0,call:0,can:0,captur:0,capture_stderr:0,capture_stdout:0,chang:0,channel:0,charact:0,check:0,chroma:0,clip:0,cmd:0,code:0,codec:0,collid:0,color:0,colorchannelmix:0,com:0,combin:0,command:0,commnad:0,common:0,compil:0,concat:0,concaten:0,configur:0,confus:0,constant:0,construct:0,consult:0,contain:0,continu:0,convert:0,coord:0,coordin:0,corner:0,correctli:0,correspond:0,count:0,crop:0,crop_bitmap:0,custom:0,dar:0,data:0,debug:0,degre:0,deprec:0,descent:0,detail:0,dictionari:0,differ:0,directli:0,disabl:0,displai:0,distanc:0,document:0,draw:0,drawbox:0,drawn:0,drawtext:0,drop:0,due:0,durat:0,dure:0,dynam:0,each:0,edg:0,effect:0,either:0,empti:0,emul:0,enabl:0,encod:0,encount:0,end:0,end_fram:0,end_pt:0,endal:0,eof:0,eof_act:0,equival:0,err:0,error:0,escape_text:0,etc:0,eval:0,evalu:0,even:0,exactli:0,exampl:0,except:0,exit:0,expand:0,expans:0,explicitli:0,expr:0,express:0,fail:0,fallback:0,fals:0,famili:0,ffmpeg_arg:0,ffprobe:0,file:0,filenam:0,filter:0,filter_:0,filter_multi_output:0,filter_nam:0,first:0,fix:0,fix_bound:0,flag:0,flip:0,fmpeg:0,follow:0,font:0,fontcolor:0,fontcolor_expr:0,fontconfig:0,fontfil:0,fontsiz:0,forc:0,force_autohint:0,format:0,fps:0,frame:0,frame_num:0,from:0,ft_load_:0,ft_load_flag:0,gbrp:0,gener:0,get_arg:0,github:0,given:0,glyph:0,graph:0,greater:0,grid:0,handl:0,has:0,have:0,hd720:0,height:0,heigth:0,hflip:0,higher:0,highest:0,horizont:0,hour:0,how:0,hsub:0,http:0,hue:0,huge:0,ignore_global_advance_width:0,ignore_transform:0,imag:0,immedi:0,implement:0,includ:0,incom:0,index:0,inform:0,init:0,initi:0,input:0,instead:0,interpret:0,invalid:0,invert:0,invok:0,its:0,ivok:0,join:0,json:0,just:0,kept:0,keyword:0,kkroen:0,kwarg:0,label:0,last:0,layout:0,left:0,level:0,libfontconfig:0,libfreetyp:0,libfribidi:0,librari:0,line:0,line_h:0,line_spac:0,linear_design:0,list:0,load:0,longest:0,lowest:0,luma:0,mai:0,main:0,main_h:0,main_parent_nod:0,main_w:0,mandatori:0,mani:0,manual:0,map:0,max:0,max_glyph_a:0,max_glyph_d:0,max_glyph_h:0,max_glyph_w:0,maximum:0,mean:0,merge_output:0,messag:0,microsecond:0,min:0,miss:0,mix:0,mode:0,modifi:0,modul:0,monochrom:0,more:0,most:0,mp4:0,multipl:0,must:0,name:0,nan:0,necessari:0,need:0,neg:0,no_autohint:0,no_bitmap:0,no_hint:0,no_recurs:0,no_scal:0,node:0,non:0,none:0,normal:0,number:0,obtain:0,offici:0,offset:0,onc:0,one:0,onli:0,oper:0,option:0,order:0,orient:0,other:0,otherwis:0,out:0,outlin:0,output:0,over:0,overlai:0,overlaid:0,overlay_parent_nod:0,overrid:0,overwrit:0,overwrite_output:0,pack:0,pad:0,page:0,pan:0,paramet:0,partial:0,pass:0,path:0,pcm:0,pedant:0,pipe:0,pixel:0,place:0,planar:0,pleas:0,point:0,posit:0,preced:0,present:0,probe:0,process:0,produc:0,properti:0,provid:0,pts:0,quiet:0,radian:0,rais:0,rand:0,random:0,rang:0,rate:0,ratio:0,rawvideo:0,read:0,reason:0,refer:0,rel:0,relat:0,reload:0,render:0,repeat:0,repeatlast:0,represent:0,resolut:0,respect:0,result:0,retriev:0,revers:0,rgb:0,right:0,run:0,same:0,sampl:0,san:0,sar:0,satur:0,search:0,second:0,secondari:0,section:0,see:0,segment:0,select:0,sent:0,sequenc:0,set:0,setpt:0,shadow:0,shadowcolor:0,shadowi:0,shadowx:0,shape:0,shorter:0,shortest:0,shorthand:0,should:0,shown:0,silenc:0,singl:0,size:0,sloppi:0,some:0,space:0,special:0,specifi:0,split0:0,split1:0,split:0,standard:0,start:0,start_fram:0,start_numb:0,start_pt:0,stderr:0,stdin:0,stdout:0,stream1:0,stream2:0,stream3:0,stream:0,stream_spec:0,streams_and_filenam:0,strftime:0,string:0,subpart:0,subsampl:0,suffix:0,suppli:0,support:0,sure:0,synchron:0,synopsi:0,syntax:0,system:0,tab:0,tabsiz:0,take:0,tc24hmax:0,tell:0,termin:0,text:0,text_h:0,text_shap:0,text_w:0,textfil:0,than:0,thei:0,them:0,thi:0,thick:0,through:0,thrown:0,time:0,timebas:0,timecod:0,timecode_r:0,timestamp:0,togeth:0,top:0,track:0,trim:0,tupl:0,type:0,unit:0,unknown:0,unsaf:0,until:0,updat:0,upper:0,upward:0,url:0,use:0,used:0,useful:0,user:0,uses:0,using:0,utf:0,util:0,valu:0,variabl:0,variou:0,vcodec:0,verbatim:0,vertic:0,vertical_layout:0,vflip:0,video:0,video_bitr:0,view:0,visibl:0,vsub:0,wai:0,well:0,whatev:0,when:0,where:0,which:0,white:0,width:0,within:0,without:0,work:0,wrap:0,write:0,you:0,yuv420:0,yuv422:0,yuv422p:0,yuv444:0,zero:0,zoom:0,zoompan:0},titles:["ffmpeg-python: Python bindings for FFmpeg"],titleterms:{bind:0,ffmpeg:0,indic:0,python:0,tabl:0}})
\ No newline at end of file
+Search.setIndex({docnames:["index"],envversion:{"sphinx.domains.c":1,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":1,"sphinx.domains.javascript":1,"sphinx.domains.math":2,"sphinx.domains.python":1,"sphinx.domains.rst":1,"sphinx.domains.std":1,sphinx:56},filenames:["index.rst"],objects:{"":{ffmpeg:[0,0,0,"-"]},"ffmpeg.Stream":{audio:[0,3,1,""],video:[0,3,1,""],view:[0,3,1,""]},ffmpeg:{Error:[0,1,1,""],Stream:[0,2,1,""],colorchannelmixer:[0,4,1,""],compile:[0,4,1,""],concat:[0,4,1,""],crop:[0,4,1,""],drawbox:[0,4,1,""],drawtext:[0,4,1,""],filter:[0,4,1,""],filter_:[0,4,1,""],filter_multi_output:[0,4,1,""],get_args:[0,4,1,""],hflip:[0,4,1,""],hue:[0,4,1,""],input:[0,4,1,""],merge_outputs:[0,4,1,""],output:[0,4,1,""],overlay:[0,4,1,""],overwrite_output:[0,4,1,""],probe:[0,4,1,""],run:[0,4,1,""],run_async:[0,4,1,""],setpts:[0,4,1,""],trim:[0,4,1,""],vflip:[0,4,1,""],view:[0,4,1,""],zoompan:[0,4,1,""]}},objnames:{"0":["py","module","Python module"],"1":["py","exception","Python exception"],"2":["py","class","Python class"],"3":["py","method","Python method"],"4":["py","function","Python function"]},objtypes:{"0":"py:module","1":"py:exception","2":"py:class","3":"py:method","4":"py:function"},terms:{"break":0,"case":0,"class":0,"default":0,"final":0,"function":0,"return":0,"true":0,"while":0,For:0,PTS:0,The:0,These:0,Used:0,accept:0,access:0,acodec:0,action:0,activ:0,adjust:0,aecho:0,after:0,alia:0,all:0,allow:0,alpha:0,also:0,altern:0,alwai:0,angl:0,ani:0,anoth:0,appli:0,arab:0,area:0,arg:0,argument:0,around:0,ascent:0,asetpt:0,ask:0,aspect:0,astyp:0,asynchron:0,atom:0,attempt:0,audio:0,audio_bitr:0,author:0,automat:0,avoid:0,axi:0,background:0,base:0,baselin:0,basetim:0,befor:0,behavior:0,behind:0,below:0,between:0,black:0,blend:0,border:0,bordercolor:0,borderw:0,both:0,box:0,boxborderw:0,boxcolor:0,bracket:0,bright:0,build:0,built:0,call:0,can:0,captur:0,capture_stderr:0,capture_stdout:0,care:0,certain:0,chang:0,channel:0,charact:0,check:0,child:0,chroma:0,clip:0,close:0,cmd:0,code:0,codec:0,collid:0,color:0,colorchannelmix:0,com:0,combin:0,command:0,commnad:0,common:0,commun:0,compil:0,concat:0,concaten:0,configur:0,confus:0,connect:0,constant:0,construct:0,consult:0,contain:0,continu:0,convert:0,coord:0,coordin:0,corner:0,correctli:0,correspond:0,count:0,creat:0,crop:0,crop_bitmap:0,custom:0,dar:0,data:0,debug:0,degre:0,deprec:0,descent:0,detail:0,dictionari:0,differ:0,dilemma:0,directli:0,disabl:0,displai:0,distanc:0,document:0,downstream:0,draw:0,drawbox:0,drawn:0,drawtext:0,drop:0,due:0,durat:0,dure:0,dynam:0,each:0,edg:0,effect:0,either:0,empti:0,emul:0,enabl:0,encod:0,encount:0,end:0,end_fram:0,end_pt:0,endal:0,eof:0,eof_act:0,equival:0,err:0,error:0,escape_text:0,etc:0,eval:0,evalu:0,even:0,exactli:0,exampl:0,except:0,exit:0,expand:0,expans:0,explicitli:0,expr:0,express:0,fail:0,fallback:0,fals:0,famili:0,ffmpeg_arg:0,ffprobe:0,file:0,filenam:0,filter:0,filter_:0,filter_multi_output:0,filter_nam:0,first:0,fix:0,fix_bound:0,flag:0,flip:0,follow:0,font:0,fontcolor:0,fontcolor_expr:0,fontconfig:0,fontfil:0,fontsiz:0,forc:0,force_autohint:0,format:0,fps:0,frame:0,frame_num:0,from:0,frombuff:0,ft_load_:0,ft_load_flag:0,gbrp:0,gener:0,get_arg:0,github:0,given:0,glyph:0,graph:0,greater:0,grid:0,handl:0,has:0,have:0,hd720:0,height:0,heigth:0,hflip:0,higher:0,highest:0,horizont:0,hour:0,how:0,hsub:0,http:0,hue:0,huge:0,ignore_global_advance_width:0,ignore_transform:0,imag:0,immedi:0,implement:0,in_byt:0,in_filenam:0,in_fram:0,includ:0,incom:0,independ:0,index:0,inform:0,init:0,initi:0,input:0,input_data:0,instead:0,interpret:0,intrins:0,invalid:0,invert:0,invok:0,its:0,join:0,json:0,just:0,kept:0,keyword:0,kkroen:0,kwarg:0,label:0,last:0,later:0,layout:0,left:0,level:0,libfontconfig:0,libfreetyp:0,libfribidi:0,librari:0,line:0,line_h:0,line_spac:0,linear_design:0,list:0,load:0,longest:0,lowest:0,luma:0,mai:0,main:0,main_h:0,main_parent_nod:0,main_w:0,mandatori:0,mani:0,manual:0,map:0,max:0,max_glyph_a:0,max_glyph_d:0,max_glyph_h:0,max_glyph_w:0,maximum:0,mean:0,merge_output:0,messag:0,microsecond:0,min:0,miss:0,mix:0,mode:0,modifi:0,modul:0,monochrom:0,more:0,most:0,mp4:0,multipl:0,must:0,name:0,nan:0,necessari:0,need:0,neg:0,no_autohint:0,no_bitmap:0,no_hint:0,no_recurs:0,no_scal:0,node:0,node_typ:0,non:0,none:0,normal:0,number:0,numpi:0,object:0,obtain:0,offici:0,offset:0,onc:0,one:0,onli:0,oper:0,option:0,order:0,orient:0,other:0,otherwis:0,out:0,out_filenam:0,out_fram:0,outgo:0,outlin:0,output:0,over:0,overlai:0,overlaid:0,overlay_parent_nod:0,overrid:0,overwrit:0,overwrite_output:0,pack:0,pad:0,page:0,pan:0,paramet:0,partial:0,pass:0,path:0,pcm:0,pedant:0,pipe:0,pipe_stderr:0,pipe_stdin:0,pipe_stdout:0,pipelin:0,pix_fmt:0,pixel:0,place:0,planar:0,pleas:0,point:0,popen:0,portion:0,posit:0,preced:0,present:0,preserv:0,probe:0,process1:0,process2:0,process:0,produc:0,properti:0,provid:0,pts:0,quiet:0,radian:0,rais:0,rand:0,random:0,rang:0,rate:0,ratio:0,rawvideo:0,read:0,reason:0,refer:0,rel:0,relat:0,reload:0,render:0,repeat:0,repeatlast:0,repres:0,represent:0,reshap:0,resolut:0,respect:0,result:0,retriev:0,revers:0,rgb24:0,rgb:0,right:0,run:0,run_async:0,same:0,sampl:0,san:0,sar:0,satur:0,search:0,second:0,secondari:0,section:0,see:0,segment:0,select:0,sent:0,separ:0,sequenc:0,set:0,setpt:0,shadow:0,shadowcolor:0,shadowi:0,shadowx:0,shape:0,shorter:0,shortest:0,shorthand:0,should:0,shown:0,silenc:0,singl:0,size:0,sloppi:0,some:0,space:0,special:0,specifi:0,split0:0,split1:0,split:0,stai:0,standard:0,start:0,start_fram:0,start_numb:0,start_pt:0,stderr:0,stdin:0,stdout:0,stream1:0,stream2:0,stream3:0,stream:0,stream_spec:0,streams_and_filenam:0,strftime:0,string:0,subpart:0,subprocess:0,subsampl:0,suffix:0,suppli:0,support:0,sure:0,synchron:0,synopsi:0,syntax:0,system:0,tab:0,tabsiz:0,take:0,taken:0,tc24hmax:0,tell:0,termin:0,text:0,text_h:0,text_shap:0,text_w:0,textfil:0,than:0,thei:0,them:0,thi:0,thick:0,through:0,thrown:0,time:0,timebas:0,timecod:0,timecode_r:0,timestamp:0,tobyt:0,togeth:0,top:0,track:0,tri:0,trim:0,tupl:0,type:0,uint8:0,unit:0,unknown:0,unsaf:0,until:0,updat:0,upper:0,upstream:0,upstream_label:0,upstream_nod:0,upstream_selector:0,upward:0,url:0,use:0,used:0,useful:0,user:0,uses:0,using:0,utf:0,util:0,valu:0,variabl:0,variou:0,vcodec:0,verbatim:0,vertic:0,vertical_layout:0,vflip:0,video:0,video_bitr:0,view:0,visibl:0,vsub:0,wai:0,wait:0,well:0,whatev:0,when:0,where:0,which:0,white:0,why:0,width:0,within:0,without:0,work:0,wrap:0,write:0,you:0,yuv420:0,yuv420p:0,yuv422:0,yuv422p:0,yuv444:0,zero:0,zoom:0,zoompan:0},titles:["ffmpeg-python: Python bindings for FFmpeg"],titleterms:{bind:0,ffmpeg:0,indic:0,python:0,tabl:0}})
\ No newline at end of file
diff --git a/examples/README.md b/examples/README.md
index 92feaece..dabc7398 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -51,7 +51,7 @@ out, _ = (
.input(in_filename)
.filter('select', 'gte(n,{})'.format(frame_num))
.output('pipe:', vframes=1, format='image2', vcodec='mjpeg')
- .run(capture_output=True)
+ .run(capture_stdout=True)
)
```
@@ -104,10 +104,10 @@ With additional filtering:
```python
in1 = ffmpeg.input('in1.mp4')
in2 = ffmpeg.input('in2.mp4')
-v1 = in1['v'].hflip()
-a1 = in1['a']
-v2 = in2['v'].filter('reverse').filter('hue', s=0)
-a2 = in2['a'].filter('areverse').filter('aphaser')
+v1 = in1.video.hflip()
+a1 = in1.audio
+v2 = in2.video.filter('reverse').filter('hue', s=0)
+a2 = in2.audio.filter('areverse').filter('aphaser')
joined = ffmpeg.concat(v1, a1, v2, a2, v=1, a=1).node
v3 = joined[0]
a3 = joined[1].filter('volume', 0.8)
@@ -115,6 +115,36 @@ out = ffmpeg.output(v3, a3, 'out.mp4')
out.run()
```
+## Mono to stereo with offsets and video
+
+
+
+```python
+audio_left = (
+ ffmpeg
+ .input('audio-left.wav')
+ .filter('atrim', start=5)
+ .filter('asetpts', 'PTS-STARTPTS')
+)
+
+audio_right = (
+ ffmpeg
+ .input('audio-right.wav')
+ .filter('atrim', start=10)
+ .filter('asetpts', 'PTS-STARTPTS')
+)
+
+input_video = ffmpeg.input('input-video.mp4')
+
+(
+ ffmpeg
+ .filter((audio_left, audio_right), 'join', inputs=2, channel_layout='stereo')
+ .output(input_video.video, 'output-video.mp4', shortest=None, vcodec='copy')
+ .overwrite_output()
+ .run()
+)
+```
+
## [Jupyter Frame Viewer](https://github.com/kkroening/ffmpeg-python/blob/master/examples/ffmpeg-numpy.ipynb)
@@ -123,3 +153,111 @@ out.run()
+## [Tensorflow Streaming](https://github.com/kkroening/ffmpeg-python/blob/master/examples/tensorflow_stream.py)
+
+
+
+- Decode input video with ffmpeg
+- Process video with tensorflow using "deep dream" example
+- Encode output video with ffmpeg
+
+```python
+process1 = (
+ ffmpeg
+ .input(in_filename)
+ .output('pipe:', format='rawvideo', pix_fmt='rgb24', vframes=8)
+ .run_async(pipe_stdout=True)
+)
+
+process2 = (
+ ffmpeg
+ .input('pipe:', format='rawvideo', pix_fmt='rgb24', s='{}x{}'.format(width, height))
+ .output(out_filename, pix_fmt='yuv420p')
+ .overwrite_output()
+ .run_async(pipe_stdin=True)
+)
+
+while True:
+ in_bytes = process1.stdout.read(width * height * 3)
+ if not in_bytes:
+ break
+ in_frame = (
+ np
+ .frombuffer(in_bytes, np.uint8)
+ .reshape([height, width, 3])
+ )
+
+ # See examples/tensorflow_stream.py:
+ out_frame = deep_dream.process_frame(in_frame)
+
+ process2.stdin.write(
+ out_frame
+ .astype(np.uint8)
+ .tobytes()
+ )
+
+process2.stdin.close()
+process1.wait()
+process2.wait()
+```
+
+
+
+## [FaceTime webcam input (OS X)](https://github.com/kkroening/ffmpeg-python/blob/master/examples/facetime.py)
+
+```python
+(
+ ffmpeg
+ .input('FaceTime', format='avfoundation', pix_fmt='uyvy422', framerate=30)
+ .output('out.mp4', pix_fmt='yuv420p', vframes=100)
+ .run()
+)
+```
+
+## Stream from a local video to HTTP server
+
+```python
+video_format = "flv"
+server_url = "http://127.0.0.1:8080"
+
+process = (
+ ffmpeg
+ .input("input.mp4")
+ .output(
+ server_url,
+ codec = "copy", # use same codecs of the original video
+ listen=1, # enables HTTP server
+ f=video_format)
+ .global_args("-re") # argument to act as a live stream
+ .run()
+)
+
+```
+
+to receive the video you can use ffplay in the terminal:
+
+```
+$ ffplay -f flv http://localhost:8080
+```
+
+## Stream from RTSP server to TCP socket
+
+```python
+packet_size = 4096
+
+process = (
+ ffmpeg
+ .input('rtsp://%s:8554/default')
+ .output('-', format='h264')
+ .run_async(pipe_stdout=True)
+)
+
+while process.poll() is None:
+ packet = process.stdout.read(packet_size)
+ try:
+ tcp_socket.send(packet)
+ except socket.error:
+ process.stdout.close()
+ process.wait()
+ break
+```
diff --git a/examples/facetime.py b/examples/facetime.py
new file mode 100644
index 00000000..58d083ec
--- /dev/null
+++ b/examples/facetime.py
@@ -0,0 +1,8 @@
+import ffmpeg
+
+(
+ ffmpeg
+ .input('FaceTime', format='avfoundation', pix_fmt='uyvy422', framerate=30)
+ .output('out.mp4', pix_fmt='yuv420p', vframes=100)
+ .run()
+)
diff --git a/examples/graphs/dream.png b/examples/graphs/dream.png
new file mode 100644
index 00000000..090b530b
Binary files /dev/null and b/examples/graphs/dream.png differ
diff --git a/examples/graphs/mono-to-stereo.png b/examples/graphs/mono-to-stereo.png
new file mode 100644
index 00000000..92ce0e5b
Binary files /dev/null and b/examples/graphs/mono-to-stereo.png differ
diff --git a/examples/graphs/tensorflow-stream.png b/examples/graphs/tensorflow-stream.png
new file mode 100644
index 00000000..e5047a16
Binary files /dev/null and b/examples/graphs/tensorflow-stream.png differ
diff --git a/examples/split_silence.py b/examples/split_silence.py
index a889db10..90b46d95 100755
--- a/examples/split_silence.py
+++ b/examples/split_silence.py
@@ -27,10 +27,10 @@
parser.add_argument('--end-time', type=float, help='End time (seconds)')
parser.add_argument('-v', dest='verbose', action='store_true', help='Verbose mode')
-silence_start_re = re.compile(' silence_start: (?P[0-9]+(\.?[0-9]*))$')
-silence_end_re = re.compile(' silence_end: (?P[0-9]+(\.?[0-9]*)) ')
+silence_start_re = re.compile(r' silence_start: (?P[0-9]+(\.?[0-9]*))$')
+silence_end_re = re.compile(r' silence_end: (?P[0-9]+(\.?[0-9]*)) ')
total_duration_re = re.compile(
- 'size=[^ ]+ time=(?P[0-9]{2}):(?P[0-9]{2}):(?P[0-9\.]{5}) bitrate=')
+ r'size=[^ ]+ time=(?P[0-9]{2}):(?P[0-9]{2}):(?P[0-9\.]{5}) bitrate=')
def _logged_popen(cmd_line, *args, **kwargs):
diff --git a/examples/tensorflow_stream.py b/examples/tensorflow_stream.py
new file mode 100644
index 00000000..6c9c9c9d
--- /dev/null
+++ b/examples/tensorflow_stream.py
@@ -0,0 +1,248 @@
+'''Example streaming ffmpeg numpy processing.
+
+Demonstrates using ffmpeg to decode video input, process the frames in
+python, and then encode video output using ffmpeg.
+
+This example uses two ffmpeg processes - one to decode the input video
+and one to encode an output video - while the raw frame processing is
+done in python with numpy.
+
+At a high level, the signal graph looks like this:
+
+ (input video) -> [ffmpeg process 1] -> [python] -> [ffmpeg process 2] -> (output video)
+
+This example reads/writes video files on the local filesystem, but the
+same pattern can be used for other kinds of input/output (e.g. webcam,
+rtmp, etc.).
+
+The simplest processing example simply darkens each frame by
+multiplying the frame's numpy array by a constant value; see
+``process_frame_simple``.
+
+A more sophisticated example processes each frame with tensorflow using
+the "deep dream" tensorflow tutorial; activate this mode by calling
+the script with the optional `--dream` argument. (Make sure tensorflow
+is installed before running)
+'''
+from __future__ import print_function
+import argparse
+import ffmpeg
+import logging
+import numpy as np
+import os
+import subprocess
+import zipfile
+
+
+parser = argparse.ArgumentParser(description='Example streaming ffmpeg numpy processing')
+parser.add_argument('in_filename', help='Input filename')
+parser.add_argument('out_filename', help='Output filename')
+parser.add_argument(
+ '--dream', action='store_true', help='Use DeepDream frame processing (requires tensorflow)')
+
+logger = logging.getLogger(__name__)
+logging.basicConfig(level=logging.INFO)
+
+
+def get_video_size(filename):
+ logger.info('Getting video size for {!r}'.format(filename))
+ probe = ffmpeg.probe(filename)
+ video_info = next(s for s in probe['streams'] if s['codec_type'] == 'video')
+ width = int(video_info['width'])
+ height = int(video_info['height'])
+ return width, height
+
+
+def start_ffmpeg_process1(in_filename):
+ logger.info('Starting ffmpeg process1')
+ args = (
+ ffmpeg
+ .input(in_filename)
+ .output('pipe:', format='rawvideo', pix_fmt='rgb24')
+ .compile()
+ )
+ return subprocess.Popen(args, stdout=subprocess.PIPE)
+
+
+def start_ffmpeg_process2(out_filename, width, height):
+ logger.info('Starting ffmpeg process2')
+ args = (
+ ffmpeg
+ .input('pipe:', format='rawvideo', pix_fmt='rgb24', s='{}x{}'.format(width, height))
+ .output(out_filename, pix_fmt='yuv420p')
+ .overwrite_output()
+ .compile()
+ )
+ return subprocess.Popen(args, stdin=subprocess.PIPE)
+
+
+def read_frame(process1, width, height):
+ logger.debug('Reading frame')
+
+ # Note: RGB24 == 3 bytes per pixel.
+ frame_size = width * height * 3
+ in_bytes = process1.stdout.read(frame_size)
+ if len(in_bytes) == 0:
+ frame = None
+ else:
+ assert len(in_bytes) == frame_size
+ frame = (
+ np
+ .frombuffer(in_bytes, np.uint8)
+ .reshape([height, width, 3])
+ )
+ return frame
+
+
+def process_frame_simple(frame):
+ '''Simple processing example: darken frame.'''
+ return frame * 0.3
+
+
+def write_frame(process2, frame):
+ logger.debug('Writing frame')
+ process2.stdin.write(
+ frame
+ .astype(np.uint8)
+ .tobytes()
+ )
+
+
+def run(in_filename, out_filename, process_frame):
+ width, height = get_video_size(in_filename)
+ process1 = start_ffmpeg_process1(in_filename)
+ process2 = start_ffmpeg_process2(out_filename, width, height)
+ while True:
+ in_frame = read_frame(process1, width, height)
+ if in_frame is None:
+ logger.info('End of input stream')
+ break
+
+ logger.debug('Processing frame')
+ out_frame = process_frame(in_frame)
+ write_frame(process2, out_frame)
+
+ logger.info('Waiting for ffmpeg process1')
+ process1.wait()
+
+ logger.info('Waiting for ffmpeg process2')
+ process2.stdin.close()
+ process2.wait()
+
+ logger.info('Done')
+
+
+class DeepDream(object):
+ '''DeepDream implementation, adapted from official tensorflow deepdream tutorial:
+ https://github.com/tensorflow/tensorflow/tree/master/tensorflow/examples/tutorials/deepdream
+
+ Credit: Alexander Mordvintsev
+ '''
+
+ _DOWNLOAD_URL = 'https://storage.googleapis.com/download.tensorflow.org/models/inception5h.zip'
+ _ZIP_FILENAME = 'deepdream_model.zip'
+ _MODEL_FILENAME = 'tensorflow_inception_graph.pb'
+
+ @staticmethod
+ def _download_model():
+ logger.info('Downloading deepdream model...')
+ try:
+ from urllib.request import urlretrieve # python 3
+ except ImportError:
+ from urllib import urlretrieve # python 2
+ urlretrieve(DeepDream._DOWNLOAD_URL, DeepDream._ZIP_FILENAME)
+
+ logger.info('Extracting deepdream model...')
+ zipfile.ZipFile(DeepDream._ZIP_FILENAME, 'r').extractall('.')
+
+ @staticmethod
+ def _tffunc(*argtypes):
+ '''Helper that transforms TF-graph generating function into a regular one.
+ See `_resize` function below.
+ '''
+ placeholders = list(map(tf.placeholder, argtypes))
+ def wrap(f):
+ out = f(*placeholders)
+ def wrapper(*args, **kw):
+ return out.eval(dict(zip(placeholders, args)), session=kw.get('session'))
+ return wrapper
+ return wrap
+
+ @staticmethod
+ def _base_resize(img, size):
+ '''Helper function that uses TF to resize an image'''
+ img = tf.expand_dims(img, 0)
+ return tf.image.resize_bilinear(img, size)[0,:,:,:]
+
+ def __init__(self):
+ if not os.path.exists(DeepDream._MODEL_FILENAME):
+ self._download_model()
+
+ self._graph = tf.Graph()
+ self._session = tf.InteractiveSession(graph=self._graph)
+ self._resize = self._tffunc(np.float32, np.int32)(self._base_resize)
+ with tf.gfile.FastGFile(DeepDream._MODEL_FILENAME, 'rb') as f:
+ graph_def = tf.GraphDef()
+ graph_def.ParseFromString(f.read())
+ self._t_input = tf.placeholder(np.float32, name='input') # define the input tensor
+ imagenet_mean = 117.0
+ t_preprocessed = tf.expand_dims(self._t_input-imagenet_mean, 0)
+ tf.import_graph_def(graph_def, {'input':t_preprocessed})
+
+ self.t_obj = self.T('mixed4d_3x3_bottleneck_pre_relu')[:,:,:,139]
+ #self.t_obj = tf.square(self.T('mixed4c'))
+
+ def T(self, layer_name):
+ '''Helper for getting layer output tensor'''
+ return self._graph.get_tensor_by_name('import/%s:0'%layer_name)
+
+ def _calc_grad_tiled(self, img, t_grad, tile_size=512):
+ '''Compute the value of tensor t_grad over the image in a tiled way.
+ Random shifts are applied to the image to blur tile boundaries over
+ multiple iterations.'''
+ sz = tile_size
+ h, w = img.shape[:2]
+ sx, sy = np.random.randint(sz, size=2)
+ img_shift = np.roll(np.roll(img, sx, 1), sy, 0)
+ grad = np.zeros_like(img)
+ for y in range(0, max(h-sz//2, sz),sz):
+ for x in range(0, max(w-sz//2, sz),sz):
+ sub = img_shift[y:y+sz,x:x+sz]
+ g = self._session.run(t_grad, {self._t_input:sub})
+ grad[y:y+sz,x:x+sz] = g
+ return np.roll(np.roll(grad, -sx, 1), -sy, 0)
+
+ def process_frame(self, frame, iter_n=10, step=1.5, octave_n=4, octave_scale=1.4):
+ t_score = tf.reduce_mean(self.t_obj) # defining the optimization objective
+ t_grad = tf.gradients(t_score, self._t_input)[0] # behold the power of automatic differentiation!
+
+ # split the image into a number of octaves
+ img = frame
+ octaves = []
+ for i in range(octave_n-1):
+ hw = img.shape[:2]
+ lo = self._resize(img, np.int32(np.float32(hw)/octave_scale))
+ hi = img-self._resize(lo, hw)
+ img = lo
+ octaves.append(hi)
+
+ # generate details octave by octave
+ for octave in range(octave_n):
+ if octave>0:
+ hi = octaves[-octave]
+ img = self._resize(img, hi.shape[:2])+hi
+ for i in range(iter_n):
+ g = self._calc_grad_tiled(img, t_grad)
+ img += g*(step / (np.abs(g).mean()+1e-7))
+ #print('.',end = ' ')
+ return img
+
+
+if __name__ == '__main__':
+ args = parser.parse_args()
+ if args.dream:
+ import tensorflow as tf
+ process_frame = DeepDream().process_frame
+ else:
+ process_frame = process_frame_simple
+ run(args.in_filename, args.out_filename, process_frame)
diff --git a/ffmpeg/__init__.py b/ffmpeg/__init__.py
index 7e913d37..a88d344d 100644
--- a/ffmpeg/__init__.py
+++ b/ffmpeg/__init__.py
@@ -1,10 +1,22 @@
from __future__ import unicode_literals
-
-from . import _filters, _ffmpeg, _run, _probe
-from ._filters import *
+from . import nodes
+from . import _ffmpeg
+from . import _filters
+from . import _probe
+from . import _run
+from . import _view
+from .nodes import *
from ._ffmpeg import *
+from ._filters import *
+from ._probe import *
from ._run import *
from ._view import *
-from ._probe import *
-__all__ = _filters.__all__ + _ffmpeg.__all__ + _run.__all__ + _view.__all__ + _probe.__all__
+__all__ = (
+ nodes.__all__
+ + _ffmpeg.__all__
+ + _probe.__all__
+ + _run.__all__
+ + _view.__all__
+ + _filters.__all__
+)
diff --git a/ffmpeg/_ffmpeg.py b/ffmpeg/_ffmpeg.py
index 30e8e712..007624bb 100644
--- a/ffmpeg/_ffmpeg.py
+++ b/ffmpeg/_ffmpeg.py
@@ -34,8 +34,7 @@ def input(filename, **kwargs):
@output_operator()
def global_args(stream, *args):
- """Add extra global command-line argument(s), e.g. ``-progress``.
- """
+ """Add extra global command-line argument(s), e.g. ``-progress``."""
return GlobalNode(stream, global_args.__name__, args).stream()
@@ -50,8 +49,7 @@ def overwrite_output(stream):
@output_operator()
def merge_outputs(*streams):
- """Include all given outputs in one ffmpeg command line
- """
+ """Include all given outputs in one ffmpeg command line"""
return MergeOutputsNode(streams, merge_outputs.__name__).stream()
@@ -94,9 +92,4 @@ def output(*streams_and_filename, **kwargs):
return OutputNode(streams, output.__name__, kwargs=kwargs).stream()
-__all__ = [
- 'input',
- 'merge_outputs',
- 'output',
- 'overwrite_output',
-]
+__all__ = ['input', 'merge_outputs', 'output', 'overwrite_output']
diff --git a/ffmpeg/_filters.py b/ffmpeg/_filters.py
index 6ba30d34..5bca23d8 100644
--- a/ffmpeg/_filters.py
+++ b/ffmpeg/_filters.py
@@ -8,9 +8,11 @@
def filter_multi_output(stream_spec, filter_name, *args, **kwargs):
"""Apply custom filter with one or more outputs.
- This is the same as ``filter_`` except that the filter can produce more than one output.
+ This is the same as ``filter`` except that the filter can produce more than one
+ output.
- To reference an output stream, use either the ``.stream`` operator or bracket shorthand:
+ To reference an output stream, use either the ``.stream`` operator or bracket
+ shorthand:
Example:
@@ -21,16 +23,19 @@ def filter_multi_output(stream_spec, filter_name, *args, **kwargs):
ffmpeg.concat(split0, split1).output('out.mp4').run()
```
"""
- return FilterNode(stream_spec, filter_name, args=args, kwargs=kwargs, max_inputs=None)
+ return FilterNode(
+ stream_spec, filter_name, args=args, kwargs=kwargs, max_inputs=None
+ )
@filter_operator()
def filter(stream_spec, filter_name, *args, **kwargs):
"""Apply custom filter.
- ``filter_`` is normally used by higher-level filter functions such as ``hflip``, but if a filter implementation
- is missing from ``fmpeg-python``, you can call ``filter_`` directly to have ``fmpeg-python`` pass the filter name
- and arguments to ffmpeg verbatim.
+ ``filter_`` is normally used by higher-level filter functions such as ``hflip``,
+ but if a filter implementation is missing from ``ffmpeg-python``, you can call
+ ``filter_`` directly to have ``ffmpeg-python`` pass the filter name and arguments
+ to ffmpeg verbatim.
Args:
stream_spec: a Stream, list of Streams, or label-to-Stream dictionary mapping
@@ -38,7 +43,8 @@ def filter(stream_spec, filter_name, *args, **kwargs):
*args: list of args to pass to ffmpeg verbatim
**kwargs: list of keyword-args to pass to ffmpeg verbatim
- The function name is suffixed with ``_`` in order avoid confusion with the standard python ``filter`` function.
+ The function name is suffixed with ``_`` in order avoid confusion with the standard
+ python ``filter`` function.
Example:
@@ -70,7 +76,8 @@ def setpts(stream, expr):
"""Change the PTS (presentation timestamp) of the input frames.
Args:
- expr: The expression which is evaluated for each frame to construct its timestamp.
+ expr: The expression which is evaluated for each frame to construct its
+ timestamp.
Official documentation: `setpts, asetpts `__
"""
@@ -82,14 +89,15 @@ def trim(stream, **kwargs):
"""Trim the input so that the output contains one continuous subpart of the input.
Args:
- start: Specify the time of the start of the kept section, i.e. the frame with the timestamp start will be the
- first frame in the output.
- end: Specify the time of the first frame that will be dropped, i.e. the frame immediately preceding the one
- with the timestamp end will be the last frame in the output.
- start_pts: This is the same as start, except this option sets the start timestamp in timebase units instead of
- seconds.
- end_pts: This is the same as end, except this option sets the end timestamp in timebase units instead of
- seconds.
+ start: Specify the time of the start of the kept section, i.e. the frame with
+ the timestamp start will be the first frame in the output.
+ end: Specify the time of the first frame that will be dropped, i.e. the frame
+ immediately preceding the one with the timestamp end will be the last frame
+ in the output.
+ start_pts: This is the same as start, except this option sets the start
+ timestamp in timebase units instead of seconds.
+ end_pts: This is the same as end, except this option sets the end timestamp in
+ timebase units instead of seconds.
duration: The maximum duration of the output in seconds.
start_frame: The number of the first frame that should be passed to the output.
end_frame: The number of the first frame that should be dropped.
@@ -104,14 +112,16 @@ def overlay(main_parent_node, overlay_parent_node, eof_action='repeat', **kwargs
"""Overlay one video on top of another.
Args:
- x: Set the expression for the x coordinates of the overlaid video on the main video. Default value is 0. In
- case the expression is invalid, it is set to a huge value (meaning that the overlay will not be displayed
- within the output visible area).
- y: Set the expression for the y coordinates of the overlaid video on the main video. Default value is 0. In
- case the expression is invalid, it is set to a huge value (meaning that the overlay will not be displayed
- within the output visible area).
- eof_action: The action to take when EOF is encountered on the secondary input; it accepts one of the following
- values:
+ x: Set the expression for the x coordinates of the overlaid video on the main
+ video. Default value is 0. In case the expression is invalid, it is set to
+ a huge value (meaning that the overlay will not be displayed within the
+ output visible area).
+ y: Set the expression for the y coordinates of the overlaid video on the main
+ video. Default value is 0. In case the expression is invalid, it is set to
+ a huge value (meaning that the overlay will not be displayed within the
+ output visible area).
+ eof_action: The action to take when EOF is encountered on the secondary input;
+ it accepts one of the following values:
* ``repeat``: Repeat the last frame (the default).
* ``endall``: End both streams.
@@ -120,12 +130,13 @@ def overlay(main_parent_node, overlay_parent_node, eof_action='repeat', **kwargs
eval: Set when the expressions for x, and y are evaluated.
It accepts the following values:
- * ``init``: only evaluate expressions once during the filter initialization or when a command is
- processed
+ * ``init``: only evaluate expressions once during the filter initialization
+ or when a command is processed
* ``frame``: evaluate expressions for each incoming frame
Default value is ``frame``.
- shortest: If set to 1, force the output to terminate when the shortest input terminates. Default value is 0.
+ shortest: If set to 1, force the output to terminate when the shortest input
+ terminates. Default value is 0.
format: Set the format for the output video.
It accepts the following values:
@@ -136,15 +147,22 @@ def overlay(main_parent_node, overlay_parent_node, eof_action='repeat', **kwargs
* ``gbrp``: force planar RGB output
Default value is ``yuv420``.
- rgb (deprecated): If set to 1, force the filter to accept inputs in the RGB color space. Default value is 0.
- This option is deprecated, use format instead.
- repeatlast: If set to 1, force the filter to draw the last overlay frame over the main input until the end of
- the stream. A value of 0 disables this behavior. Default value is 1.
+ rgb (deprecated): If set to 1, force the filter to accept inputs in the RGB
+ color space. Default value is 0. This option is deprecated, use format
+ instead.
+ repeatlast: If set to 1, force the filter to draw the last overlay frame over
+ the main input until the end of the stream. A value of 0 disables this
+ behavior. Default value is 1.
Official documentation: `overlay `__
"""
kwargs['eof_action'] = eof_action
- return FilterNode([main_parent_node, overlay_parent_node], overlay.__name__, kwargs=kwargs, max_inputs=2).stream()
+ return FilterNode(
+ [main_parent_node, overlay_parent_node],
+ overlay.__name__,
+ kwargs=kwargs,
+ max_inputs=2,
+ ).stream()
@filter_operator()
@@ -175,15 +193,12 @@ def crop(stream, x, y, width, height, **kwargs):
y: The vertical position, in the input video, of the top edge of the
output video.
width: The width of the output video. Must be greater than 0.
- heigth: The height of the output video. Must be greater than 0.
+ height: The height of the output video. Must be greater than 0.
Official documentation: `crop `__
"""
return FilterNode(
- stream,
- crop.__name__,
- args=[width, height, x, y],
- kwargs=kwargs
+ stream, crop.__name__, args=[width, height, x, y], kwargs=kwargs
).stream()
@@ -192,14 +207,20 @@ def drawbox(stream, x, y, width, height, color, thickness=None, **kwargs):
"""Draw a colored box on the input image.
Args:
- x: The expression which specifies the top left corner x coordinate of the box. It defaults to 0.
- y: The expression which specifies the top left corner y coordinate of the box. It defaults to 0.
- width: Specify the width of the box; if 0 interpreted as the input width. It defaults to 0.
- heigth: Specify the height of the box; if 0 interpreted as the input height. It defaults to 0.
- color: Specify the color of the box to write. For the general syntax of this option, check the "Color" section
- in the ffmpeg-utils manual. If the special value invert is used, the box edge color is the same as the
- video with inverted luma.
- thickness: The expression which sets the thickness of the box edge. Default value is 3.
+ x: The expression which specifies the top left corner x coordinate of the box.
+ It defaults to 0.
+ y: The expression which specifies the top left corner y coordinate of the box.
+ It defaults to 0.
+ width: Specify the width of the box; if 0 interpreted as the input width. It
+ defaults to 0.
+ height: Specify the height of the box; if 0 interpreted as the input height. It
+ defaults to 0.
+ color: Specify the color of the box to write. For the general syntax of this
+ option, check the "Color" section in the ffmpeg-utils manual. If the
+ special value invert is used, the box edge color is the same as the video
+ with inverted luma.
+ thickness: The expression which sets the thickness of the box edge. Default
+ value is 3.
w: Alias for ``width``.
h: Alias for ``height``.
c: Alias for ``color``.
@@ -209,51 +230,64 @@ def drawbox(stream, x, y, width, height, color, thickness=None, **kwargs):
"""
if thickness:
kwargs['t'] = thickness
- return FilterNode(stream, drawbox.__name__, args=[x, y, width, height, color], kwargs=kwargs).stream()
+ return FilterNode(
+ stream, drawbox.__name__, args=[x, y, width, height, color], kwargs=kwargs
+ ).stream()
@filter_operator()
def drawtext(stream, text=None, x=0, y=0, escape_text=True, **kwargs):
- """Draw a text string or text from a specified file on top of a video, using the libfreetype library.
+ """Draw a text string or text from a specified file on top of a video, using the
+ libfreetype library.
- To enable compilation of this filter, you need to configure FFmpeg with ``--enable-libfreetype``. To enable default
- font fallback and the font option you need to configure FFmpeg with ``--enable-libfontconfig``. To enable the
+ To enable compilation of this filter, you need to configure FFmpeg with
+ ``--enable-libfreetype``. To enable default font fallback and the font option you
+ need to configure FFmpeg with ``--enable-libfontconfig``. To enable the
text_shaping option, you need to configure FFmpeg with ``--enable-libfribidi``.
Args:
- box: Used to draw a box around text using the background color. The value must be either 1 (enable) or 0
- (disable). The default value of box is 0.
- boxborderw: Set the width of the border to be drawn around the box using boxcolor. The default value of
- boxborderw is 0.
- boxcolor: The color to be used for drawing box around text. For the syntax of this option, check the "Color"
- section in the ffmpeg-utils manual. The default value of boxcolor is "white".
- line_spacing: Set the line spacing in pixels of the border to be drawn around the box using box. The default
- value of line_spacing is 0.
- borderw: Set the width of the border to be drawn around the text using bordercolor. The default value of
- borderw is 0.
- bordercolor: Set the color to be used for drawing border around text. For the syntax of this option, check the
- "Color" section in the ffmpeg-utils manual. The default value of bordercolor is "black".
- expansion: Select how the text is expanded. Can be either none, strftime (deprecated) or normal (default). See
- the Text expansion section below for details.
- basetime: Set a start time for the count. Value is in microseconds. Only applied in the deprecated strftime
- expansion mode. To emulate in normal expansion mode use the pts function, supplying the start time (in
- seconds) as the second argument.
+ box: Used to draw a box around text using the background color. The value must
+ be either 1 (enable) or 0 (disable). The default value of box is 0.
+ boxborderw: Set the width of the border to be drawn around the box using
+ boxcolor. The default value of boxborderw is 0.
+ boxcolor: The color to be used for drawing box around text. For the syntax of
+ this option, check the "Color" section in the ffmpeg-utils manual. The
+ default value of boxcolor is "white".
+ line_spacing: Set the line spacing in pixels of the border to be drawn around
+ the box using box. The default value of line_spacing is 0.
+ borderw: Set the width of the border to be drawn around the text using
+ bordercolor. The default value of borderw is 0.
+ bordercolor: Set the color to be used for drawing border around text. For the
+ syntax of this option, check the "Color" section in the ffmpeg-utils
+ manual. The default value of bordercolor is "black".
+ expansion: Select how the text is expanded. Can be either none, strftime
+ (deprecated) or normal (default). See the Text expansion section below for
+ details.
+ basetime: Set a start time for the count. Value is in microseconds. Only
+ applied in the deprecated strftime expansion mode. To emulate in normal
+ expansion mode use the pts function, supplying the start time (in seconds)
+ as the second argument.
fix_bounds: If true, check and fix text coords to avoid clipping.
- fontcolor: The color to be used for drawing fonts. For the syntax of this option, check the "Color" section in
- the ffmpeg-utils manual. The default value of fontcolor is "black".
- fontcolor_expr: String which is expanded the same way as text to obtain dynamic fontcolor value. By default
- this option has empty value and is not processed. When this option is set, it overrides fontcolor option.
+ fontcolor: The color to be used for drawing fonts. For the syntax of this
+ option, check the "Color" section in the ffmpeg-utils manual. The default
+ value of fontcolor is "black".
+ fontcolor_expr: String which is expanded the same way as text to obtain dynamic
+ fontcolor value. By default this option has empty value and is not
+ processed. When this option is set, it overrides fontcolor option.
font: The font family to be used for drawing text. By default Sans.
- fontfile: The font file to be used for drawing text. The path must be included. This parameter is mandatory if
- the fontconfig support is disabled.
- alpha: Draw the text applying alpha blending. The value can be a number between 0.0 and 1.0. The expression
- accepts the same variables x, y as well. The default value is 1. Please see fontcolor_expr.
- fontsize: The font size to be used for drawing text. The default value of fontsize is 16.
- text_shaping: If set to 1, attempt to shape the text (for example, reverse the order of right-to-left text and
- join Arabic characters) before drawing it. Otherwise, just draw the text exactly as given. By default 1 (if
- supported).
- ft_load_flags: The flags to be used for loading the fonts. The flags map the corresponding flags supported by
- libfreetype, and are a combination of the following values:
+ fontfile: The font file to be used for drawing text. The path must be included.
+ This parameter is mandatory if the fontconfig support is disabled.
+ alpha: Draw the text applying alpha blending. The value can be a number between
+ 0.0 and 1.0. The expression accepts the same variables x, y as well. The
+ default value is 1. Please see fontcolor_expr.
+ fontsize: The font size to be used for drawing text. The default value of
+ fontsize is 16.
+ text_shaping: If set to 1, attempt to shape the text (for example, reverse the
+ order of right-to-left text and join Arabic characters) before drawing it.
+ Otherwise, just draw the text exactly as given. By default 1 (if supported).
+ ft_load_flags: The flags to be used for loading the fonts. The flags map the
+ corresponding flags supported by libfreetype, and are a combination of the
+ following values:
* ``default``
* ``no_scale``
@@ -271,75 +305,89 @@ def drawtext(stream, text=None, x=0, y=0, escape_text=True, **kwargs):
* ``linear_design``
* ``no_autohint``
- Default value is "default". For more information consult the documentation for the FT_LOAD_* libfreetype
- flags.
- shadowcolor: The color to be used for drawing a shadow behind the drawn text. For the syntax of this option,
- check the "Color" section in the ffmpeg-utils manual. The default value of shadowcolor is "black".
- shadowx: The x offset for the text shadow position with respect to the position of the text. It can be either
- positive or negative values. The default value is "0".
- shadowy: The y offset for the text shadow position with respect to the position of the text. It can be either
- positive or negative values. The default value is "0".
- start_number: The starting frame number for the n/frame_num variable. The default value is "0".
- tabsize: The size in number of spaces to use for rendering the tab. Default value is 4.
- timecode: Set the initial timecode representation in "hh:mm:ss[:;.]ff" format. It can be used with or without
- text parameter. timecode_rate option must be specified.
+ Default value is "default". For more information consult the documentation
+ for the FT_LOAD_* libfreetype flags.
+ shadowcolor: The color to be used for drawing a shadow behind the drawn text.
+ For the syntax of this option, check the "Color" section in the ffmpeg-utils
+ manual. The default value of shadowcolor is "black".
+ shadowx: The x offset for the text shadow position with respect to the position
+ of the text. It can be either positive or negative values. The default value
+ is "0".
+ shadowy: The y offset for the text shadow position with respect to the position
+ of the text. It can be either positive or negative values. The default value
+ is "0".
+ start_number: The starting frame number for the n/frame_num variable. The
+ default value is "0".
+ tabsize: The size in number of spaces to use for rendering the tab. Default
+ value is 4.
+ timecode: Set the initial timecode representation in "hh:mm:ss[:;.]ff" format.
+ It can be used with or without text parameter. timecode_rate option must be
+ specified.
rate: Set the timecode frame rate (timecode only).
timecode_rate: Alias for ``rate``.
r: Alias for ``rate``.
- tc24hmax: If set to 1, the output of the timecode option will wrap around at 24 hours. Default is 0 (disabled).
- text: The text string to be drawn. The text must be a sequence of UTF-8 encoded characters. This parameter is
- mandatory if no file is specified with the parameter textfile.
- textfile: A text file containing text to be drawn. The text must be a sequence of UTF-8 encoded characters.
- This parameter is mandatory if no text string is specified with the parameter text. If both text and
- textfile are specified, an error is thrown.
- reload: If set to 1, the textfile will be reloaded before each frame. Be sure to update it atomically, or it
- may be read partially, or even fail.
- x: The expression which specifies the offset where text will be drawn within the video frame. It is relative to
- the left border of the output image. The default value is "0".
- y: The expression which specifies the offset where text will be drawn within the video frame. It is relative to
- the top border of the output image. The default value is "0". See below for the list of accepted constants
- and functions.
+ tc24hmax: If set to 1, the output of the timecode option will wrap around at 24
+ hours. Default is 0 (disabled).
+ text: The text string to be drawn. The text must be a sequence of UTF-8 encoded
+ characters. This parameter is mandatory if no file is specified with the
+ parameter textfile.
+ textfile: A text file containing text to be drawn. The text must be a sequence
+ of UTF-8 encoded characters. This parameter is mandatory if no text string
+ is specified with the parameter text. If both text and textfile are
+ specified, an error is thrown.
+ reload: If set to 1, the textfile will be reloaded before each frame. Be sure
+ to update it atomically, or it may be read partially, or even fail.
+ x: The expression which specifies the offset where text will be drawn within
+ the video frame. It is relative to the left border of the output image. The
+ default value is "0".
+ y: The expression which specifies the offset where text will be drawn within
+ the video frame. It is relative to the top border of the output image. The
+ default value is "0". See below for the list of accepted constants and
+ functions.
Expression constants:
- The parameters for x and y are expressions containing the following constants and functions:
- - dar: input display aspect ratio, it is the same as ``(w / h) * sar``
- - hsub: horizontal chroma subsample values. For example for the pixel format "yuv422p" hsub is 2 and vsub
- is 1.
- - vsub: vertical chroma subsample values. For example for the pixel format "yuv422p" hsub is 2 and vsub
- is 1.
- - line_h: the height of each text line
- - lh: Alias for ``line_h``.
- - main_h: the input height
- - h: Alias for ``main_h``.
- - H: Alias for ``main_h``.
- - main_w: the input width
- - w: Alias for ``main_w``.
- - W: Alias for ``main_w``.
- - ascent: the maximum distance from the baseline to the highest/upper grid coordinate used to place a glyph
- outline point, for all the rendered glyphs. It is a positive value, due to the grid's orientation with the Y
- axis upwards.
- - max_glyph_a: Alias for ``ascent``.
- - descent: the maximum distance from the baseline to the lowest grid coordinate used to place a glyph outline
- point, for all the rendered glyphs. This is a negative value, due to the grid's orientation, with the Y axis
- upwards.
- - max_glyph_d: Alias for ``descent``.
- - max_glyph_h: maximum glyph height, that is the maximum height for all the glyphs contained in the rendered
- text, it is equivalent to ascent - descent.
- - max_glyph_w: maximum glyph width, that is the maximum width for all the glyphs contained in the rendered
- text.
- - n: the number of input frame, starting from 0
- - rand(min, max): return a random number included between min and max
- - sar: The input sample aspect ratio.
- - t: timestamp expressed in seconds, NAN if the input timestamp is unknown
- - text_h: the height of the rendered text
- - th: Alias for ``text_h``.
- - text_w: the width of the rendered text
- - tw: Alias for ``text_w``.
- - x: the x offset coordinates where the text is drawn.
- - y: the y offset coordinates where the text is drawn.
-
- These parameters allow the x and y expressions to refer each other, so you can for example specify
- ``y=x/dar``.
+ The parameters for x and y are expressions containing the following constants
+ and functions:
+ - dar: input display aspect ratio, it is the same as ``(w / h) * sar``
+ - hsub: horizontal chroma subsample values. For example for the pixel format
+ "yuv422p" hsub is 2 and vsub is 1.
+ - vsub: vertical chroma subsample values. For example for the pixel format
+ "yuv422p" hsub is 2 and vsub is 1.
+ - line_h: the height of each text line
+ - lh: Alias for ``line_h``.
+ - main_h: the input height
+ - h: Alias for ``main_h``.
+ - H: Alias for ``main_h``.
+ - main_w: the input width
+ - w: Alias for ``main_w``.
+ - W: Alias for ``main_w``.
+ - ascent: the maximum distance from the baseline to the highest/upper grid
+ coordinate used to place a glyph outline point, for all the rendered glyphs.
+ It is a positive value, due to the grid's orientation with the Y axis
+ upwards.
+ - max_glyph_a: Alias for ``ascent``.
+ - descent: the maximum distance from the baseline to the lowest grid
+ coordinate used to place a glyph outline
+ point, for all the rendered glyphs. This is a negative value, due to the
+ grid's orientation, with the Y axis upwards.
+ - max_glyph_d: Alias for ``descent``.
+ - max_glyph_h: maximum glyph height, that is the maximum height for all the
+ glyphs contained in the rendered text, it is equivalent to ascent - descent.
+ - max_glyph_w: maximum glyph width, that is the maximum width for all the
+ glyphs contained in the rendered text.
+ - n: the number of input frame, starting from 0
+ - rand(min, max): return a random number included between min and max
+ - sar: The input sample aspect ratio.
+ - t: timestamp expressed in seconds, NAN if the input timestamp is unknown
+ - text_h: the height of the rendered text
+ - th: Alias for ``text_h``.
+ - text_w: the width of the rendered text
+ - tw: Alias for ``text_w``.
+ - x: the x offset coordinates where the text is drawn.
+ - y: the y offset coordinates where the text is drawn.
+
+ These parameters allow the x and y expressions to refer each other, so you can
+ for example specify ``y=x/dar``.
Official documentation: `drawtext `__
"""
@@ -358,25 +406,28 @@ def drawtext(stream, text=None, x=0, y=0, escape_text=True, **kwargs):
def concat(*streams, **kwargs):
"""Concatenate audio and video streams, joining them together one after the other.
- The filter works on segments of synchronized video and audio streams. All segments must have the same number of
- streams of each type, and that will also be the number of streams at output.
+ The filter works on segments of synchronized video and audio streams. All segments
+ must have the same number of streams of each type, and that will also be the number
+ of streams at output.
Args:
unsafe: Activate unsafe mode: do not fail if segments have a different format.
- Related streams do not always have exactly the same duration, for various reasons including codec frame size or
- sloppy authoring. For that reason, related synchronized streams (e.g. a video and its audio track) should be
- concatenated at once. The concat filter will use the duration of the longest stream in each segment (except the
- last one), and if necessary pad shorter audio streams with silence.
+ Related streams do not always have exactly the same duration, for various reasons
+ including codec frame size or sloppy authoring. For that reason, related
+ synchronized streams (e.g. a video and its audio track) should be concatenated at
+ once. The concat filter will use the duration of the longest stream in each segment
+ (except the last one), and if necessary pad shorter audio streams with silence.
For this filter to work correctly, all segments must start at timestamp 0.
- All corresponding streams must have the same parameters in all segments; the filtering system will automatically
- select a common pixel format for video streams, and a common sample format, sample rate and channel layout for
- audio streams, but other settings, such as resolution, must be converted explicitly by the user.
+ All corresponding streams must have the same parameters in all segments; the
+ filtering system will automatically select a common pixel format for video streams,
+ and a common sample format, sample rate and channel layout for audio streams, but
+ other settings, such as resolution, must be converted explicitly by the user.
- Different frame rates are acceptable but will result in variable frame rate at output; be sure to configure the
- output file to handle it.
+ Different frame rates are acceptable but will result in variable frame rate at
+ output; be sure to configure the output file to handle it.
Official documentation: `concat `__
"""
@@ -385,8 +436,10 @@ def concat(*streams, **kwargs):
stream_count = video_stream_count + audio_stream_count
if len(streams) % stream_count != 0:
raise ValueError(
- 'Expected concat input streams to have length multiple of {} (v={}, a={}); got {}'
- .format(stream_count, video_stream_count, audio_stream_count, len(streams)))
+ 'Expected concat input streams to have length multiple of {} (v={}, a={}); got {}'.format(
+ stream_count, video_stream_count, audio_stream_count, len(streams)
+ )
+ )
kwargs['n'] = int(len(streams) / stream_count)
return FilterNode(streams, concat.__name__, kwargs=kwargs, max_inputs=None).stream()
@@ -399,8 +452,8 @@ def zoompan(stream, **kwargs):
zoom: Set the zoom expression. Default is 1.
x: Set the x expression. Default is 0.
y: Set the y expression. Default is 0.
- d: Set the duration expression in number of frames. This sets for how many number of frames effect will last
- for single input image.
+ d: Set the duration expression in number of frames. This sets for how many
+ number of frames effect will last for single input image.
s: Set the output image size, default is ``hd720``.
fps: Set the output frame rate, default is 25.
z: Alias for ``zoom``.
@@ -415,10 +468,14 @@ def hue(stream, **kwargs):
"""Modify the hue and/or the saturation of the input.
Args:
- h: Specify the hue angle as a number of degrees. It accepts an expression, and defaults to "0".
- s: Specify the saturation in the [-10,10] range. It accepts an expression and defaults to "1".
- H: Specify the hue angle as a number of radians. It accepts an expression, and defaults to "0".
- b: Specify the brightness in the [-10,10] range. It accepts an expression and defaults to "0".
+ h: Specify the hue angle as a number of degrees. It accepts an expression, and
+ defaults to "0".
+ s: Specify the saturation in the [-10,10] range. It accepts an expression and
+ defaults to "1".
+ H: Specify the hue angle as a number of radians. It accepts an expression, and
+ defaults to "0".
+ b: Specify the brightness in the [-10,10] range. It accepts an expression and
+ defaults to "0".
Official documentation: `hue `__
"""
diff --git a/ffmpeg/_probe.py b/ffmpeg/_probe.py
index 2bcf6368..090d7abf 100644
--- a/ffmpeg/_probe.py
+++ b/ffmpeg/_probe.py
@@ -1,9 +1,10 @@
import json
import subprocess
from ._run import Error
+from ._utils import convert_kwargs_to_cmd_line_args
-def probe(filename, cmd='ffprobe'):
+def probe(filename, cmd='ffprobe', timeout=None, **kwargs):
"""Run ffprobe on the specified file and return a JSON representation of the output.
Raises:
@@ -12,14 +13,18 @@ def probe(filename, cmd='ffprobe'):
The stderr output can be retrieved by accessing the
``stderr`` property of the exception.
"""
- args = [cmd, '-show_format', '-show_streams', '-of', 'json', filename]
+ args = [cmd, '-show_format', '-show_streams', '-of', 'json']
+ args += convert_kwargs_to_cmd_line_args(kwargs)
+ args += [filename]
+
p = subprocess.Popen(args, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
- out, err = p.communicate()
+ communicate_kwargs = {}
+ if timeout is not None:
+ communicate_kwargs['timeout'] = timeout
+ out, err = p.communicate(**communicate_kwargs)
if p.returncode != 0:
raise Error('ffprobe', out, err)
return json.loads(out.decode('utf-8'))
-__all__ = [
- 'probe',
-]
+__all__ = ['probe']
diff --git a/ffmpeg/_run.py b/ffmpeg/_run.py
index 87e44ff6..f42d1d73 100644
--- a/ffmpeg/_run.py
+++ b/ffmpeg/_run.py
@@ -1,17 +1,13 @@
from __future__ import unicode_literals
from .dag import get_outgoing_edges, topo_sort
-from ._utils import basestring
+from ._utils import basestring, convert_kwargs_to_cmd_line_args
from builtins import str
from functools import reduce
-import collections
import copy
import operator
import subprocess
-from ._ffmpeg import (
- input,
- output,
-)
+from ._ffmpeg import input, output
from .nodes import (
get_stream_spec_nodes,
FilterNode,
@@ -21,24 +17,21 @@
output_operator,
)
+try:
+ from collections.abc import Iterable
+except ImportError:
+ from collections import Iterable
+
class Error(Exception):
def __init__(self, cmd, stdout, stderr):
- super(Error, self).__init__('{} error (see stderr output for detail)'.format(cmd))
+ super(Error, self).__init__(
+ '{} error (see stderr output for detail)'.format(cmd)
+ )
self.stdout = stdout
self.stderr = stderr
-def _convert_kwargs_to_cmd_line_args(kwargs):
- args = []
- for k in sorted(kwargs.keys()):
- v = kwargs[k]
- args.append('-{}'.format(k))
- if v is not None:
- args.append('{}'.format(v))
- return args
-
-
def _get_input_args(input_node):
if input_node.name == input.__name__:
kwargs = copy.copy(input_node.kwargs)
@@ -50,7 +43,7 @@ def _get_input_args(input_node):
args += ['-f', fmt]
if video_size:
args += ['-video_size', '{}x{}'.format(video_size[0], video_size[1])]
- args += _convert_kwargs_to_cmd_line_args(kwargs)
+ args += convert_kwargs_to_cmd_line_args(kwargs)
args += ['-i', filename]
else:
raise ValueError('Unsupported input node: {}'.format(input_node))
@@ -79,9 +72,15 @@ def _format_output_stream_name(stream_name_map, edge):
def _get_filter_spec(node, outgoing_edge_map, stream_name_map):
incoming_edges = node.incoming_edges
outgoing_edges = get_outgoing_edges(node, outgoing_edge_map)
- inputs = [_format_input_stream_name(stream_name_map, edge) for edge in incoming_edges]
- outputs = [_format_output_stream_name(stream_name_map, edge) for edge in outgoing_edges]
- filter_spec = '{}{}{}'.format(''.join(inputs), node._get_filter(outgoing_edges), ''.join(outputs))
+ inputs = [
+ _format_input_stream_name(stream_name_map, edge) for edge in incoming_edges
+ ]
+ outputs = [
+ _format_output_stream_name(stream_name_map, edge) for edge in outgoing_edges
+ ]
+ filter_spec = '{}{}{}'.format(
+ ''.join(inputs), node._get_filter(outgoing_edges), ''.join(outputs)
+ )
return filter_spec
@@ -89,19 +88,25 @@ def _allocate_filter_stream_names(filter_nodes, outgoing_edge_maps, stream_name_
stream_count = 0
for upstream_node in filter_nodes:
outgoing_edge_map = outgoing_edge_maps[upstream_node]
- for upstream_label, downstreams in list(outgoing_edge_map.items()):
+ for upstream_label, downstreams in sorted(outgoing_edge_map.items()):
if len(downstreams) > 1:
# TODO: automatically insert `splits` ahead of time via graph transformation.
raise ValueError(
- 'Encountered {} with multiple outgoing edges with same upstream label {!r}; a '
- '`split` filter is probably required'.format(upstream_node, upstream_label))
+ 'Encountered {} with multiple outgoing edges with same upstream '
+ 'label {!r}; a `split` filter is probably required'.format(
+ upstream_node, upstream_label
+ )
+ )
stream_name_map[upstream_node, upstream_label] = 's{}'.format(stream_count)
stream_count += 1
def _get_filter_arg(filter_nodes, outgoing_edge_maps, stream_name_map):
_allocate_filter_stream_names(filter_nodes, outgoing_edge_maps, stream_name_map)
- filter_specs = [_get_filter_spec(node, outgoing_edge_maps[node], stream_name_map) for node in filter_nodes]
+ filter_specs = [
+ _get_filter_spec(node, outgoing_edge_maps[node], stream_name_map)
+ for node in filter_nodes
+ ]
return ';'.join(filter_specs)
@@ -119,7 +124,9 @@ def _get_output_args(node, stream_name_map):
for edge in node.incoming_edges:
# edge = node.incoming_edges[0]
- stream_name = _format_input_stream_name(stream_name_map, edge, is_final_arg=True)
+ stream_name = _format_input_stream_name(
+ stream_name_map, edge, is_final_arg=True
+ )
if stream_name != '0' or len(node.incoming_edges) > 1:
args += ['-map', stream_name]
@@ -133,10 +140,10 @@ def _get_output_args(node, stream_name_map):
args += ['-b:a', str(kwargs.pop('audio_bitrate'))]
if 'video_size' in kwargs:
video_size = kwargs.pop('video_size')
- if not isinstance(video_size, basestring) and isinstance(video_size, collections.Iterable):
+ if not isinstance(video_size, basestring) and isinstance(video_size, Iterable):
video_size = '{}x{}'.format(video_size[0], video_size[1])
args += ['-video_size', video_size]
- args += _convert_kwargs_to_cmd_line_args(kwargs)
+ args += convert_kwargs_to_cmd_line_args(kwargs)
args += [filename]
return args
@@ -157,7 +164,9 @@ def get_args(stream_spec, overwrite_output=False):
args += reduce(operator.add, [_get_input_args(node) for node in input_nodes])
if filter_arg:
args += ['-filter_complex', filter_arg]
- args += reduce(operator.add, [_get_output_args(node, stream_name_map) for node in output_nodes])
+ args += reduce(
+ operator.add, [_get_output_args(node, stream_name_map) for node in output_nodes]
+ )
args += reduce(operator.add, [_get_global_args(node) for node in global_nodes], [])
if overwrite_output:
args += ['-y']
@@ -168,7 +177,7 @@ def get_args(stream_spec, overwrite_output=False):
def compile(stream_spec, cmd='ffmpeg', overwrite_output=False):
"""Build command-line for invoking ffmpeg.
- The :meth:`run` function uses this to build the commnad line
+ The :meth:`run` function uses this to build the command line
arguments and should work in most cases, but calling this function
directly is useful for debugging or if you need to invoke ffmpeg
manually for whatever reason.
@@ -183,11 +192,122 @@ def compile(stream_spec, cmd='ffmpeg', overwrite_output=False):
return cmd + get_args(stream_spec, overwrite_output=overwrite_output)
+@output_operator()
+def run_async(
+ stream_spec,
+ cmd='ffmpeg',
+ pipe_stdin=False,
+ pipe_stdout=False,
+ pipe_stderr=False,
+ quiet=False,
+ overwrite_output=False,
+ cwd=None,
+):
+ """Asynchronously invoke ffmpeg for the supplied node graph.
+
+ Args:
+ pipe_stdin: if True, connect pipe to subprocess stdin (to be
+ used with ``pipe:`` ffmpeg inputs).
+ pipe_stdout: if True, connect pipe to subprocess stdout (to be
+ used with ``pipe:`` ffmpeg outputs).
+ pipe_stderr: if True, connect pipe to subprocess stderr.
+ quiet: shorthand for setting ``capture_stdout`` and
+ ``capture_stderr``.
+ **kwargs: keyword-arguments passed to ``get_args()`` (e.g.
+ ``overwrite_output=True``).
+
+ Returns:
+ A `subprocess Popen`_ object representing the child process.
+
+ Examples:
+ Run and stream input::
+
+ process = (
+ ffmpeg
+ .input('pipe:', format='rawvideo', pix_fmt='rgb24', s='{}x{}'.format(width, height))
+ .output(out_filename, pix_fmt='yuv420p')
+ .overwrite_output()
+ .run_async(pipe_stdin=True)
+ )
+ process.communicate(input=input_data)
+
+ Run and capture output::
+
+ process = (
+ ffmpeg
+ .input(in_filename)
+ .output('pipe:', format='rawvideo', pix_fmt='rgb24')
+ .run_async(pipe_stdout=True, pipe_stderr=True)
+ )
+ out, err = process.communicate()
+
+ Process video frame-by-frame using numpy::
+
+ process1 = (
+ ffmpeg
+ .input(in_filename)
+ .output('pipe:', format='rawvideo', pix_fmt='rgb24')
+ .run_async(pipe_stdout=True)
+ )
+
+ process2 = (
+ ffmpeg
+ .input('pipe:', format='rawvideo', pix_fmt='rgb24', s='{}x{}'.format(width, height))
+ .output(out_filename, pix_fmt='yuv420p')
+ .overwrite_output()
+ .run_async(pipe_stdin=True)
+ )
+
+ while True:
+ in_bytes = process1.stdout.read(width * height * 3)
+ if not in_bytes:
+ break
+ in_frame = (
+ np
+ .frombuffer(in_bytes, np.uint8)
+ .reshape([height, width, 3])
+ )
+ out_frame = in_frame * 0.3
+ process2.stdin.write(
+ frame
+ .astype(np.uint8)
+ .tobytes()
+ )
+
+ process2.stdin.close()
+ process1.wait()
+ process2.wait()
+
+ .. _subprocess Popen: https://docs.python.org/3/library/subprocess.html#popen-objects
+ """
+ args = compile(stream_spec, cmd, overwrite_output=overwrite_output)
+ stdin_stream = subprocess.PIPE if pipe_stdin else None
+ stdout_stream = subprocess.PIPE if pipe_stdout else None
+ stderr_stream = subprocess.PIPE if pipe_stderr else None
+ if quiet:
+ stderr_stream = subprocess.STDOUT
+ stdout_stream = subprocess.DEVNULL
+ return subprocess.Popen(
+ args,
+ stdin=stdin_stream,
+ stdout=stdout_stream,
+ stderr=stderr_stream,
+ cwd=cwd,
+ )
+
+
@output_operator()
def run(
- stream_spec, cmd='ffmpeg', capture_stdout=False, capture_stderr=False, input=None,
- quiet=False, overwrite_output=False):
- """Ivoke ffmpeg for the supplied node graph.
+ stream_spec,
+ cmd='ffmpeg',
+ capture_stdout=False,
+ capture_stderr=False,
+ input=None,
+ quiet=False,
+ overwrite_output=False,
+ cwd=None,
+):
+ """Invoke ffmpeg for the supplied node graph.
Args:
capture_stdout: if True, capture stdout (to be used with
@@ -201,13 +321,18 @@ def run(
Returns: (out, err) tuple containing captured stdout and stderr data.
"""
- args = compile(stream_spec, cmd, overwrite_output=overwrite_output)
- stdin_stream = subprocess.PIPE if input else None
- stdout_stream = subprocess.PIPE if capture_stdout or quiet else None
- stderr_stream = subprocess.PIPE if capture_stderr or quiet else None
- p = subprocess.Popen(args, stdin=stdin_stream, stdout=stdout_stream, stderr=stderr_stream)
- out, err = p.communicate(input)
- retcode = p.poll()
+ process = run_async(
+ stream_spec,
+ cmd,
+ pipe_stdin=input is not None,
+ pipe_stdout=capture_stdout,
+ pipe_stderr=capture_stderr,
+ quiet=quiet,
+ overwrite_output=overwrite_output,
+ cwd=cwd,
+ )
+ out, err = process.communicate(input)
+ retcode = process.poll()
if retcode:
raise Error('ffmpeg', out, err)
return out, err
@@ -218,4 +343,5 @@ def run(
'Error',
'get_args',
'run',
+ 'run_async',
]
diff --git a/ffmpeg/_utils.py b/ffmpeg/_utils.py
index 3514409d..9baa2c78 100644
--- a/ffmpeg/_utils.py
+++ b/ffmpeg/_utils.py
@@ -4,10 +4,16 @@
import hashlib
import sys
+
if sys.version_info.major == 2:
# noinspection PyUnresolvedReferences,PyShadowingBuiltins
str = str
+try:
+ from collections.abc import Iterable
+except ImportError:
+ from collections import Iterable
+
# `past.builtins.basestring` module can't be imported on Python3 in some environments (Ubuntu).
# This code is copy-pasted from it to avoid crashes.
@@ -34,8 +40,10 @@ def __new__(cls, name, this_bases, d):
if sys.version_info.major >= 3:
+
class basestring(with_metaclass(BaseBaseString)):
pass
+
else:
# noinspection PyUnresolvedReferences,PyCompatibility
from builtins import basestring
@@ -44,15 +52,18 @@ class basestring(with_metaclass(BaseBaseString)):
def _recursive_repr(item):
"""Hack around python `repr` to deterministically represent dictionaries.
- This is able to represent more things than json.dumps, since it does not require things to be JSON serializable
- (e.g. datetimes).
+ This is able to represent more things than json.dumps, since it does not require
+ things to be JSON serializable (e.g. datetimes).
"""
if isinstance(item, basestring):
result = str(item)
elif isinstance(item, list):
result = '[{}]'.format(', '.join([_recursive_repr(x) for x in item]))
elif isinstance(item, dict):
- kv_pairs = ['{}: {}'.format(_recursive_repr(k), _recursive_repr(item[k])) for k in sorted(item)]
+ kv_pairs = [
+ '{}: {}'.format(_recursive_repr(k), _recursive_repr(item[k]))
+ for k in sorted(item)
+ ]
result = '{' + ', '.join(kv_pairs) + '}'
else:
result = repr(item)
@@ -78,3 +89,20 @@ def escape_chars(text, chars):
for ch in chars:
text = text.replace(ch, '\\' + ch)
return text
+
+
+def convert_kwargs_to_cmd_line_args(kwargs):
+ """Helper function to build command line arguments out of dict."""
+ args = []
+ for k in sorted(kwargs.keys()):
+ v = kwargs[k]
+ if isinstance(v, Iterable) and not isinstance(v, str):
+ for value in v:
+ args.append('-{}'.format(k))
+ if value is not None:
+ args.append('{}'.format(value))
+ continue
+ args.append('-{}'.format(k))
+ if v is not None:
+ args.append('{}'.format(v))
+ return args
diff --git a/ffmpeg/_view.py b/ffmpeg/_view.py
index 06e6ff4c..31955afd 100644
--- a/ffmpeg/_view.py
+++ b/ffmpeg/_view.py
@@ -34,8 +34,10 @@ def view(stream_spec, detail=False, filename=None, pipe=False, **kwargs):
try:
import graphviz
except ImportError:
- raise ImportError('failed to import graphviz; please make sure graphviz is installed (e.g. `pip install '
- 'graphviz`)')
+ raise ImportError(
+ 'failed to import graphviz; please make sure graphviz is installed (e.g. '
+ '`pip install graphviz`)'
+ )
show_labels = kwargs.pop('show_labels', True)
if pipe and filename is not None:
@@ -49,7 +51,9 @@ def view(stream_spec, detail=False, filename=None, pipe=False, **kwargs):
graph = graphviz.Digraph(format='png')
graph.attr(rankdir='LR')
if len(list(kwargs.keys())) != 0:
- raise ValueError('Invalid kwargs key(s): {}'.format(', '.join(list(kwargs.keys()))))
+ raise ValueError(
+ 'Invalid kwargs key(s): {}'.format(', '.join(list(kwargs.keys())))
+ )
for node in sorted_nodes:
color = _get_node_color(node)
@@ -57,11 +61,15 @@ def view(stream_spec, detail=False, filename=None, pipe=False, **kwargs):
if detail:
lines = [node.short_repr]
lines += ['{!r}'.format(arg) for arg in node.args]
- lines += ['{}={!r}'.format(key, node.kwargs[key]) for key in sorted(node.kwargs)]
+ lines += [
+ '{}={!r}'.format(key, node.kwargs[key]) for key in sorted(node.kwargs)
+ ]
node_text = '\n'.join(lines)
else:
node_text = node.short_repr
- graph.node(str(hash(node)), node_text, shape='box', style='filled', fillcolor=color)
+ graph.node(
+ str(hash(node)), node_text, shape='box', style='filled', fillcolor=color
+ )
outgoing_edge_map = outgoing_edge_maps.get(node, {})
for edge in get_outgoing_edges(node, outgoing_edge_map):
@@ -70,7 +78,11 @@ def view(stream_spec, detail=False, filename=None, pipe=False, **kwargs):
down_label = edge.downstream_label
up_selector = edge.upstream_selector
- if show_labels and (up_label is not None or down_label is not None or up_selector is not None):
+ if show_labels and (
+ up_label is not None
+ or down_label is not None
+ or up_selector is not None
+ ):
if up_label is None:
up_label = ''
if up_selector is not None:
@@ -93,6 +105,4 @@ def view(stream_spec, detail=False, filename=None, pipe=False, **kwargs):
return stream_spec
-__all__ = [
- 'view',
-]
+__all__ = ['view']
diff --git a/ffmpeg/dag.py b/ffmpeg/dag.py
index fb51c6a1..4bdac44d 100644
--- a/ffmpeg/dag.py
+++ b/ffmpeg/dag.py
@@ -9,38 +9,45 @@ class DagNode(object):
"""Node in a directed-acyclic graph (DAG).
Edges:
- DagNodes are connected by edges. An edge connects two nodes with a label for each side:
+ DagNodes are connected by edges. An edge connects two nodes with a label for
+ each side:
- ``upstream_node``: upstream/parent node
- ``upstream_label``: label on the outgoing side of the upstream node
- ``downstream_node``: downstream/child node
- ``downstream_label``: label on the incoming side of the downstream node
- For example, DagNode A may be connected to DagNode B with an edge labelled "foo" on A's side, and "bar" on B's
- side:
+ For example, DagNode A may be connected to DagNode B with an edge labelled
+ "foo" on A's side, and "bar" on B's side:
_____ _____
| | | |
| A >[foo]---[bar]> B |
|_____| |_____|
- Edge labels may be integers or strings, and nodes cannot have more than one incoming edge with the same label.
+ Edge labels may be integers or strings, and nodes cannot have more than one
+ incoming edge with the same label.
- DagNodes may have any number of incoming edges and any number of outgoing edges. DagNodes keep track only of
- their incoming edges, but the entire graph structure can be inferred by looking at the furthest downstream
- nodes and working backwards.
+ DagNodes may have any number of incoming edges and any number of outgoing
+ edges. DagNodes keep track only of their incoming edges, but the entire graph
+ structure can be inferred by looking at the furthest downstream nodes and
+ working backwards.
Hashing:
- DagNodes must be hashable, and two nodes are considered to be equivalent if they have the same hash value.
+ DagNodes must be hashable, and two nodes are considered to be equivalent if
+ they have the same hash value.
- Nodes are immutable, and the hash should remain constant as a result. If a node with new contents is required,
- create a new node and throw the old one away.
+ Nodes are immutable, and the hash should remain constant as a result. If a
+ node with new contents is required, create a new node and throw the old one
+ away.
String representation:
- In order for graph visualization tools to show useful information, nodes must be representable as strings. The
- ``repr`` operator should provide a more or less "full" representation of the node, and the ``short_repr``
- property should be a shortened, concise representation.
+ In order for graph visualization tools to show useful information, nodes must
+ be representable as strings. The ``repr`` operator should provide a more or
+ less "full" representation of the node, and the ``short_repr`` property should
+ be a shortened, concise representation.
- Again, because nodes are immutable, the string representations should remain constant.
+ Again, because nodes are immutable, the string representations should remain
+ constant.
"""
def __hash__(self):
@@ -48,7 +55,9 @@ def __hash__(self):
raise NotImplementedError()
def __eq__(self, other):
- """Compare two nodes; implementations should return True if (and only if) hashes match."""
+ """Compare two nodes; implementations should return True if (and only if)
+ hashes match.
+ """
raise NotImplementedError()
def __repr__(self, other):
@@ -64,42 +73,75 @@ def short_repr(self):
def incoming_edge_map(self):
"""Provides information about all incoming edges that connect to this node.
- The edge map is a dictionary that maps an ``incoming_label`` to ``(outgoing_node, outgoing_label)``. Note that
- implicity, ``incoming_node`` is ``self``. See "Edges" section above.
+ The edge map is a dictionary that maps an ``incoming_label`` to
+ ``(outgoing_node, outgoing_label)``. Note that implicitly, ``incoming_node`` is
+ ``self``. See "Edges" section above.
"""
raise NotImplementedError()
-DagEdge = namedtuple('DagEdge', ['downstream_node', 'downstream_label', 'upstream_node', 'upstream_label', 'upstream_selector'])
+DagEdge = namedtuple(
+ 'DagEdge',
+ [
+ 'downstream_node',
+ 'downstream_label',
+ 'upstream_node',
+ 'upstream_label',
+ 'upstream_selector',
+ ],
+)
def get_incoming_edges(downstream_node, incoming_edge_map):
edges = []
for downstream_label, upstream_info in list(incoming_edge_map.items()):
upstream_node, upstream_label, upstream_selector = upstream_info
- edges += [DagEdge(downstream_node, downstream_label, upstream_node, upstream_label, upstream_selector)]
+ edges += [
+ DagEdge(
+ downstream_node,
+ downstream_label,
+ upstream_node,
+ upstream_label,
+ upstream_selector,
+ )
+ ]
return edges
def get_outgoing_edges(upstream_node, outgoing_edge_map):
edges = []
- for upstream_label, downstream_infos in list(outgoing_edge_map.items()):
+ for upstream_label, downstream_infos in sorted(outgoing_edge_map.items()):
for downstream_info in downstream_infos:
downstream_node, downstream_label, downstream_selector = downstream_info
- edges += [DagEdge(downstream_node, downstream_label, upstream_node, upstream_label, downstream_selector)]
+ edges += [
+ DagEdge(
+ downstream_node,
+ downstream_label,
+ upstream_node,
+ upstream_label,
+ downstream_selector,
+ )
+ ]
return edges
class KwargReprNode(DagNode):
- """A DagNode that can be represented as a set of args+kwargs.
- """
+ """A DagNode that can be represented as a set of args+kwargs."""
@property
def __upstream_hashes(self):
hashes = []
for downstream_label, upstream_info in list(self.incoming_edge_map.items()):
upstream_node, upstream_label, upstream_selector = upstream_info
- hashes += [hash(x) for x in [downstream_label, upstream_node, upstream_label, upstream_selector]]
+ hashes += [
+ hash(x)
+ for x in [
+ downstream_label,
+ upstream_node,
+ upstream_label,
+ upstream_selector,
+ ]
+ ]
return hashes
@property
@@ -130,7 +172,9 @@ def short_hash(self):
def long_repr(self, include_hash=True):
formatted_props = ['{!r}'.format(arg) for arg in self.args]
- formatted_props += ['{}={!r}'.format(key, self.kwargs[key]) for key in sorted(self.kwargs)]
+ formatted_props += [
+ '{}={!r}'.format(key, self.kwargs[key]) for key in sorted(self.kwargs)
+ ]
out = '{}({})'.format(self.name, ', '.join(formatted_props))
if include_hash:
out += ' <{}>'.format(self.short_hash)
@@ -157,21 +201,35 @@ def topo_sort(downstream_nodes):
sorted_nodes = []
outgoing_edge_maps = {}
- def visit(upstream_node, upstream_label, downstream_node, downstream_label, downstream_selector=None):
+ def visit(
+ upstream_node,
+ upstream_label,
+ downstream_node,
+ downstream_label,
+ downstream_selector=None,
+ ):
if upstream_node in marked_nodes:
raise RuntimeError('Graph is not a DAG')
if downstream_node is not None:
outgoing_edge_map = outgoing_edge_maps.get(upstream_node, {})
outgoing_edge_infos = outgoing_edge_map.get(upstream_label, [])
- outgoing_edge_infos += [(downstream_node, downstream_label, downstream_selector)]
+ outgoing_edge_infos += [
+ (downstream_node, downstream_label, downstream_selector)
+ ]
outgoing_edge_map[upstream_label] = outgoing_edge_infos
outgoing_edge_maps[upstream_node] = outgoing_edge_map
if upstream_node not in sorted_nodes:
marked_nodes.append(upstream_node)
for edge in upstream_node.incoming_edges:
- visit(edge.upstream_node, edge.upstream_label, edge.downstream_node, edge.downstream_label, edge.upstream_selector)
+ visit(
+ edge.upstream_node,
+ edge.upstream_label,
+ edge.downstream_node,
+ edge.downstream_label,
+ edge.upstream_selector,
+ )
marked_nodes.remove(upstream_node)
sorted_nodes.append(upstream_node)
diff --git a/ffmpeg/nodes.py b/ffmpeg/nodes.py
index ae507633..e8b28385 100644
--- a/ffmpeg/nodes.py
+++ b/ffmpeg/nodes.py
@@ -21,12 +21,19 @@ def _get_types_str(types):
class Stream(object):
- """Represents the outgoing edge of an upstream node; may be used to create more downstream nodes."""
+ """Represents the outgoing edge of an upstream node; may be used to create more
+ downstream nodes.
+ """
- def __init__(self, upstream_node, upstream_label, node_types, upstream_selector=None):
+ def __init__(
+ self, upstream_node, upstream_label, node_types, upstream_selector=None
+ ):
if not _is_of_types(upstream_node, node_types):
- raise TypeError('Expected upstream node to be of one of the following type(s): {}; got {}'.format(
- _get_types_str(node_types), type(upstream_node)))
+ raise TypeError(
+ 'Expected upstream node to be of one of the following type(s): {}; got {}'.format(
+ _get_types_str(node_types), type(upstream_node)
+ )
+ )
self.node = upstream_node
self.label = upstream_label
self.selector = upstream_selector
@@ -42,7 +49,9 @@ def __repr__(self):
selector = ''
if self.selector:
selector = ':{}'.format(self.selector)
- out = '{}[{!r}{}] <{}>'.format(node_repr, self.label, selector, self.node.short_hash)
+ out = '{}[{!r}{}] <{}>'.format(
+ node_repr, self.label, selector, self.node.short_hash
+ )
return out
def __getitem__(self, index):
@@ -53,8 +62,8 @@ def __getitem__(self, index):
Process the audio and video portions of a stream independently::
input = ffmpeg.input('in.mp4')
- audio = input[:'a'].filter("aecho", 0.8, 0.9, 1000, 0.3)
- video = input[:'v'].hflip()
+ audio = input['a'].filter("aecho", 0.8, 0.9, 1000, 0.3)
+ video = input['v'].hflip()
out = ffmpeg.output(audio, video, 'out.mp4')
"""
if self.selector is not None:
@@ -63,6 +72,56 @@ def __getitem__(self, index):
raise TypeError("Expected string index (e.g. 'a'); got {!r}".format(index))
return self.node.stream(label=self.label, selector=index)
+ @property
+ def audio(self):
+ """Select the audio-portion of a stream.
+
+ Some ffmpeg filters drop audio streams, and care must be taken
+ to preserve the audio in the final output. The ``.audio`` and
+ ``.video`` operators can be used to reference the audio/video
+ portions of a stream so that they can be processed separately
+ and then re-combined later in the pipeline. This dilemma is
+ intrinsic to ffmpeg, and ffmpeg-python tries to stay out of the
+ way while users may refer to the official ffmpeg documentation
+ as to why certain filters drop audio.
+
+ ``stream.audio`` is a shorthand for ``stream['a']``.
+
+ Example:
+ Process the audio and video portions of a stream independently::
+
+ input = ffmpeg.input('in.mp4')
+ audio = input.audio.filter("aecho", 0.8, 0.9, 1000, 0.3)
+ video = input.video.hflip()
+ out = ffmpeg.output(audio, video, 'out.mp4')
+ """
+ return self['a']
+
+ @property
+ def video(self):
+ """Select the video-portion of a stream.
+
+ Some ffmpeg filters drop audio streams, and care must be taken
+ to preserve the audio in the final output. The ``.audio`` and
+ ``.video`` operators can be used to reference the audio/video
+ portions of a stream so that they can be processed separately
+ and then re-combined later in the pipeline. This dilemma is
+ intrinsic to ffmpeg, and ffmpeg-python tries to stay out of the
+ way while users may refer to the official ffmpeg documentation
+ as to why certain filters drop audio.
+
+ ``stream.video`` is a shorthand for ``stream['v']``.
+
+ Example:
+ Process the audio and video portions of a stream independently::
+
+ input = ffmpeg.input('in.mp4')
+ audio = input.audio.filter("aecho", 0.8, 0.9, 1000, 0.3)
+ video = input.video.hflip()
+ out = ffmpeg.output(audio, video, 'out.mp4')
+ """
+ return self['v']
+
def get_stream_map(stream_spec):
if stream_spec is None:
@@ -96,26 +155,50 @@ class Node(KwargReprNode):
@classmethod
def __check_input_len(cls, stream_map, min_inputs, max_inputs):
if min_inputs is not None and len(stream_map) < min_inputs:
- raise ValueError('Expected at least {} input stream(s); got {}'.format(min_inputs, len(stream_map)))
+ raise ValueError(
+ 'Expected at least {} input stream(s); got {}'.format(
+ min_inputs, len(stream_map)
+ )
+ )
elif max_inputs is not None and len(stream_map) > max_inputs:
- raise ValueError('Expected at most {} input stream(s); got {}'.format(max_inputs, len(stream_map)))
+ raise ValueError(
+ 'Expected at most {} input stream(s); got {}'.format(
+ max_inputs, len(stream_map)
+ )
+ )
@classmethod
def __check_input_types(cls, stream_map, incoming_stream_types):
for stream in list(stream_map.values()):
if not _is_of_types(stream, incoming_stream_types):
- raise TypeError('Expected incoming stream(s) to be of one of the following types: {}; got {}'
- .format(_get_types_str(incoming_stream_types), type(stream)))
+ raise TypeError(
+ 'Expected incoming stream(s) to be of one of the following types: {}; got {}'.format(
+ _get_types_str(incoming_stream_types), type(stream)
+ )
+ )
@classmethod
def __get_incoming_edge_map(cls, stream_map):
incoming_edge_map = {}
for downstream_label, upstream in list(stream_map.items()):
- incoming_edge_map[downstream_label] = (upstream.node, upstream.label, upstream.selector)
+ incoming_edge_map[downstream_label] = (
+ upstream.node,
+ upstream.label,
+ upstream.selector,
+ )
return incoming_edge_map
- def __init__(self, stream_spec, name, incoming_stream_types, outgoing_stream_type, min_inputs,
- max_inputs, args=[], kwargs={}):
+ def __init__(
+ self,
+ stream_spec,
+ name,
+ incoming_stream_types,
+ outgoing_stream_type,
+ min_inputs,
+ max_inputs,
+ args=[],
+ kwargs={},
+ ):
stream_map = get_stream_map(stream_spec)
self.__check_input_len(stream_map, min_inputs, max_inputs)
self.__check_input_types(stream_map, incoming_stream_types)
@@ -133,9 +216,10 @@ def stream(self, label=None, selector=None):
return self.__outgoing_stream_type(self, label, upstream_selector=selector)
def __getitem__(self, item):
- """Create an outgoing stream originating from this node; syntactic sugar for ``self.stream(label)``.
- It can also be used to apply a selector: e.g. ``node[0:'a']`` returns a stream with label 0 and
- selector ``'a'``, which is the same as ``node.stream(label=0, selector='a')``.
+ """Create an outgoing stream originating from this node; syntactic sugar for
+ ``self.stream(label)``. It can also be used to apply a selector: e.g.
+ ``node[0:'a']`` returns a stream with label 0 and selector ``'a'``, which is
+ the same as ``node.stream(label=0, selector='a')``.
Example:
Process the audio and video portions of a stream independently::
@@ -153,8 +237,9 @@ def __getitem__(self, item):
class FilterableStream(Stream):
def __init__(self, upstream_node, upstream_label, upstream_selector=None):
- super(FilterableStream, self).__init__(upstream_node, upstream_label, {InputNode, FilterNode},
- upstream_selector)
+ super(FilterableStream, self).__init__(
+ upstream_node, upstream_label, {InputNode, FilterNode}, upstream_selector
+ )
# noinspection PyMethodOverriding
@@ -170,7 +255,7 @@ def __init__(self, name, args=[], kwargs={}):
min_inputs=0,
max_inputs=0,
args=args,
- kwargs=kwargs
+ kwargs=kwargs,
)
@property
@@ -189,7 +274,7 @@ def __init__(self, stream_spec, name, max_inputs=1, args=[], kwargs={}):
min_inputs=1,
max_inputs=max_inputs,
args=args,
- kwargs=kwargs
+ kwargs=kwargs,
)
"""FilterNode"""
@@ -229,7 +314,7 @@ def __init__(self, stream, name, args=[], kwargs={}):
min_inputs=1,
max_inputs=None,
args=args,
- kwargs=kwargs
+ kwargs=kwargs,
)
@property
@@ -239,8 +324,12 @@ def short_repr(self):
class OutputStream(Stream):
def __init__(self, upstream_node, upstream_label, upstream_selector=None):
- super(OutputStream, self).__init__(upstream_node, upstream_label, {OutputNode, GlobalNode, MergeOutputsNode},
- upstream_selector=upstream_selector)
+ super(OutputStream, self).__init__(
+ upstream_node,
+ upstream_label,
+ {OutputNode, GlobalNode, MergeOutputsNode},
+ upstream_selector=upstream_selector,
+ )
# noinspection PyMethodOverriding
@@ -252,7 +341,7 @@ def __init__(self, streams, name):
incoming_stream_types={OutputStream},
outgoing_stream_type=OutputStream,
min_inputs=1,
- max_inputs=None
+ max_inputs=None,
)
@@ -267,7 +356,7 @@ def __init__(self, stream, name, args=[], kwargs={}):
min_inputs=1,
max_inputs=1,
args=args,
- kwargs=kwargs
+ kwargs=kwargs,
)
@@ -286,3 +375,6 @@ def filter_operator(name=None):
def output_operator(name=None):
return stream_operator(stream_classes={OutputStream}, name=name)
+
+
+__all__ = ['Stream']
diff --git a/ffmpeg/tests/test_ffmpeg.py b/ffmpeg/tests/test_ffmpeg.py
index b62f2efc..8dbc271a 100644
--- a/ffmpeg/tests/test_ffmpeg.py
+++ b/ffmpeg/tests/test_ffmpeg.py
@@ -1,14 +1,20 @@
from __future__ import unicode_literals
-
-from builtins import str
from builtins import bytes
from builtins import range
+from builtins import str
import ffmpeg
import os
import pytest
import random
import re
import subprocess
+import sys
+
+
+try:
+ import mock # python 2
+except ImportError:
+ from unittest import mock # python 3
TEST_DIR = os.path.dirname(__file__)
@@ -24,9 +30,12 @@
def test_escape_chars():
- assert ffmpeg._utils.escape_chars('a:b', ':') == 'a\:b'
+ assert ffmpeg._utils.escape_chars('a:b', ':') == r'a\:b'
assert ffmpeg._utils.escape_chars('a\\:b', ':\\') == 'a\\\\\\:b'
- assert ffmpeg._utils.escape_chars('a:b,c[d]e%{}f\'g\'h\\i', '\\\':,[]%') == 'a\\:b\\,c\\[d\\]e\\%{}f\\\'g\\\'h\\\\i'
+ assert (
+ ffmpeg._utils.escape_chars('a:b,c[d]e%{}f\'g\'h\\i', '\\\':,[]%')
+ == 'a\\:b\\,c\\[d\\]e\\%{}f\\\'g\\\'h\\\\i'
+ )
assert ffmpeg._utils.escape_chars(123, ':\\') == '123'
@@ -58,23 +67,16 @@ def test_fluent_concat():
def test_fluent_output():
- (ffmpeg
- .input('dummy.mp4')
- .trim(start_frame=10, end_frame=20)
- .output('dummy2.mp4')
- )
+ ffmpeg.input('dummy.mp4').trim(start_frame=10, end_frame=20).output('dummy2.mp4')
def test_fluent_complex_filter():
in_file = ffmpeg.input('dummy.mp4')
- return (ffmpeg
- .concat(
- in_file.trim(start_frame=10, end_frame=20),
- in_file.trim(start_frame=30, end_frame=40),
- in_file.trim(start_frame=50, end_frame=60)
- )
- .output('dummy2.mp4')
- )
+ return ffmpeg.concat(
+ in_file.trim(start_frame=10, end_frame=20),
+ in_file.trim(start_frame=30, end_frame=40),
+ in_file.trim(start_frame=50, end_frame=60),
+ ).output('dummy2.mp4')
def test_node_repr():
@@ -84,21 +86,50 @@ def test_node_repr():
trim3 = ffmpeg.trim(in_file, start_frame=50, end_frame=60)
concatted = ffmpeg.concat(trim1, trim2, trim3)
output = ffmpeg.output(concatted, 'dummy2.mp4')
- assert repr(in_file.node) == 'input(filename={!r}) <{}>'.format('dummy.mp4', in_file.node.short_hash)
- assert repr(trim1.node) == 'trim(end_frame=20, start_frame=10) <{}>'.format(trim1.node.short_hash)
- assert repr(trim2.node) == 'trim(end_frame=40, start_frame=30) <{}>'.format(trim2.node.short_hash)
- assert repr(trim3.node) == 'trim(end_frame=60, start_frame=50) <{}>'.format(trim3.node.short_hash)
+ assert repr(in_file.node) == 'input(filename={!r}) <{}>'.format(
+ 'dummy.mp4', in_file.node.short_hash
+ )
+ assert repr(trim1.node) == 'trim(end_frame=20, start_frame=10) <{}>'.format(
+ trim1.node.short_hash
+ )
+ assert repr(trim2.node) == 'trim(end_frame=40, start_frame=30) <{}>'.format(
+ trim2.node.short_hash
+ )
+ assert repr(trim3.node) == 'trim(end_frame=60, start_frame=50) <{}>'.format(
+ trim3.node.short_hash
+ )
assert repr(concatted.node) == 'concat(n=3) <{}>'.format(concatted.node.short_hash)
- assert repr(output.node) == 'output(filename={!r}) <{}>'.format('dummy2.mp4', output.node.short_hash)
+ assert repr(output.node) == 'output(filename={!r}) <{}>'.format(
+ 'dummy2.mp4', output.node.short_hash
+ )
def test_stream_repr():
in_file = ffmpeg.input('dummy.mp4')
- assert repr(in_file) == 'input(filename={!r})[None] <{}>'.format('dummy.mp4', in_file.node.short_hash)
+ assert repr(in_file) == 'input(filename={!r})[None] <{}>'.format(
+ 'dummy.mp4', in_file.node.short_hash
+ )
split0 = in_file.filter_multi_output('split')[0]
assert repr(split0) == 'split()[0] <{}>'.format(split0.node.short_hash)
dummy_out = in_file.filter_multi_output('dummy')['out']
- assert repr(dummy_out) == 'dummy()[{!r}] <{}>'.format(dummy_out.label, dummy_out.node.short_hash)
+ assert repr(dummy_out) == 'dummy()[{!r}] <{}>'.format(
+ dummy_out.label, dummy_out.node.short_hash
+ )
+
+
+def test_repeated_args():
+ out_file = ffmpeg.input('dummy.mp4').output(
+ 'dummy2.mp4', streamid=['0:0x101', '1:0x102']
+ )
+ assert out_file.get_args() == [
+ '-i',
+ 'dummy.mp4',
+ '-streamid',
+ '0:0x101',
+ '-streamid',
+ '1:0x102',
+ 'dummy2.mp4',
+ ]
def test__get_args__simple():
@@ -107,8 +138,18 @@ def test__get_args__simple():
def test_global_args():
- out_file = ffmpeg.input('dummy.mp4').output('dummy2.mp4').global_args('-progress', 'someurl')
- assert out_file.get_args() == ['-i', 'dummy.mp4', 'dummy2.mp4', '-progress', 'someurl']
+ out_file = (
+ ffmpeg.input('dummy.mp4')
+ .output('dummy2.mp4')
+ .global_args('-progress', 'someurl')
+ )
+ assert out_file.get_args() == [
+ '-i',
+ 'dummy.mp4',
+ 'dummy2.mp4',
+ '-progress',
+ 'someurl',
+ ]
def _get_simple_example():
@@ -116,18 +157,14 @@ def _get_simple_example():
def _get_complex_filter_example():
- split = (ffmpeg
- .input(TEST_INPUT_FILE1)
- .vflip()
- .split()
- )
+ split = ffmpeg.input(TEST_INPUT_FILE1).vflip().split()
split0 = split[0]
split1 = split[1]
overlay_file = ffmpeg.input(TEST_OVERLAY_FILE)
overlay_file = ffmpeg.crop(overlay_file, 10, 10, 158, 112)
- return (ffmpeg
- .concat(
+ return (
+ ffmpeg.concat(
split0.trim(start_frame=10, end_frame=20),
split1.trim(start_frame=30, end_frame=40),
)
@@ -141,20 +178,25 @@ def _get_complex_filter_example():
def test__get_args__complex_filter():
out = _get_complex_filter_example()
args = ffmpeg.get_args(out)
- assert args == ['-i', TEST_INPUT_FILE1,
- '-i', TEST_OVERLAY_FILE,
+ assert args == [
+ '-i',
+ TEST_INPUT_FILE1,
+ '-i',
+ TEST_OVERLAY_FILE,
'-filter_complex',
- '[0]vflip[s0];' \
- '[s0]split=2[s1][s2];' \
- '[s1]trim=end_frame=20:start_frame=10[s3];' \
- '[s2]trim=end_frame=40:start_frame=30[s4];' \
- '[s3][s4]concat=n=2[s5];' \
- '[1]crop=158:112:10:10[s6];' \
- '[s6]hflip[s7];' \
- '[s5][s7]overlay=eof_action=repeat[s8];' \
- '[s8]drawbox=50:50:120:120:red:t=5[s9]',
- '-map', '[s9]', TEST_OUTPUT_FILE1,
- '-y'
+ '[0]vflip[s0];'
+ '[s0]split=2[s1][s2];'
+ '[s1]trim=end_frame=20:start_frame=10[s3];'
+ '[s2]trim=end_frame=40:start_frame=30[s4];'
+ '[s3][s4]concat=n=2[s5];'
+ '[1]crop=158:112:10:10[s6];'
+ '[s6]hflip[s7];'
+ '[s5][s7]overlay=eof_action=repeat[s8];'
+ '[s8]drawbox=50:50:120:120:red:t=5[s9]',
+ '-map',
+ '[s9]',
+ TEST_OUTPUT_FILE1,
+ '-y',
]
@@ -163,30 +205,41 @@ def test_combined_output():
i2 = ffmpeg.input(TEST_OVERLAY_FILE)
out = ffmpeg.output(i1, i2, TEST_OUTPUT_FILE1)
assert out.get_args() == [
- '-i', TEST_INPUT_FILE1,
- '-i', TEST_OVERLAY_FILE,
- '-map', '0',
- '-map', '1',
- TEST_OUTPUT_FILE1
+ '-i',
+ TEST_INPUT_FILE1,
+ '-i',
+ TEST_OVERLAY_FILE,
+ '-map',
+ '0',
+ '-map',
+ '1',
+ TEST_OUTPUT_FILE1,
]
-def test_filter_with_selector():
+@pytest.mark.parametrize('use_shorthand', [True, False])
+def test_filter_with_selector(use_shorthand):
i = ffmpeg.input(TEST_INPUT_FILE1)
- v1 = i['v'].hflip()
- a1 = i['a'].filter('aecho', 0.8, 0.9, 1000, 0.3)
+ if use_shorthand:
+ v1 = i.video.hflip()
+ a1 = i.audio.filter('aecho', 0.8, 0.9, 1000, 0.3)
+ else:
+ v1 = i['v'].hflip()
+ a1 = i['a'].filter('aecho', 0.8, 0.9, 1000, 0.3)
out = ffmpeg.output(a1, v1, TEST_OUTPUT_FILE1)
assert out.get_args() == [
- '-i', TEST_INPUT_FILE1,
+ '-i',
+ TEST_INPUT_FILE1,
'-filter_complex',
- '[0:a]aecho=0.8:0.9:1000:0.3[s0];' \
- '[0:v]hflip[s1]',
- '-map', '[s0]', '-map', '[s1]',
- TEST_OUTPUT_FILE1
+ '[0:a]aecho=0.8:0.9:1000:0.3[s0];' '[0:v]hflip[s1]',
+ '-map',
+ '[s0]',
+ '-map',
+ '[s1]',
+ TEST_OUTPUT_FILE1,
]
-
def test_get_item_with_bad_selectors():
input = ffmpeg.input(TEST_INPUT_FILE1)
@@ -204,16 +257,12 @@ def test_get_item_with_bad_selectors():
def _get_complex_filter_asplit_example():
- split = (ffmpeg
- .input(TEST_INPUT_FILE1)
- .vflip()
- .asplit()
- )
+ split = ffmpeg.input(TEST_INPUT_FILE1).vflip().asplit()
split0 = split[0]
split1 = split[1]
- return (ffmpeg
- .concat(
+ return (
+ ffmpeg.concat(
split0.filter('atrim', start=10, end=20),
split1.filter('atrim', start=30, end=40),
)
@@ -225,12 +274,7 @@ def _get_complex_filter_asplit_example():
def test_filter_concat__video_only():
in1 = ffmpeg.input('in1.mp4')
in2 = ffmpeg.input('in2.mp4')
- args = (
- ffmpeg
- .concat(in1, in2)
- .output('out.mp4')
- .get_args()
- )
+ args = ffmpeg.concat(in1, in2).output('out.mp4').get_args()
assert args == [
'-i',
'in1.mp4',
@@ -247,12 +291,7 @@ def test_filter_concat__video_only():
def test_filter_concat__audio_only():
in1 = ffmpeg.input('in1.mp4')
in2 = ffmpeg.input('in2.mp4')
- args = (
- ffmpeg
- .concat(in1, in2, v=0, a=1)
- .output('out.mp4')
- .get_args()
- )
+ args = ffmpeg.concat(in1, in2, v=0, a=1).output('out.mp4').get_args()
assert args == [
'-i',
'in1.mp4',
@@ -262,19 +301,15 @@ def test_filter_concat__audio_only():
'[0][1]concat=a=1:n=2:v=0[s0]',
'-map',
'[s0]',
- 'out.mp4'
+ 'out.mp4',
]
def test_filter_concat__audio_video():
in1 = ffmpeg.input('in1.mp4')
in2 = ffmpeg.input('in2.mp4')
- joined = ffmpeg.concat(in1['v'], in1['a'], in2.hflip(), in2['a'], v=1, a=1).node
- args = (
- ffmpeg
- .output(joined[0], joined[1], 'out.mp4')
- .get_args()
- )
+ joined = ffmpeg.concat(in1.video, in1.audio, in2.hflip(), in2['a'], v=1, a=1).node
+ args = ffmpeg.output(joined[0], joined[1], 'out.mp4').get_args()
assert args == [
'-i',
'in1.mp4',
@@ -294,9 +329,11 @@ def test_filter_concat__wrong_stream_count():
in1 = ffmpeg.input('in1.mp4')
in2 = ffmpeg.input('in2.mp4')
with pytest.raises(ValueError) as excinfo:
- ffmpeg.concat(in1['v'], in1['a'], in2.hflip(), v=1, a=1).node
- assert str(excinfo.value) == \
- 'Expected concat input streams to have length multiple of 2 (v=1, a=1); got 3'
+ ffmpeg.concat(in1.video, in1.audio, in2.hflip(), v=1, a=1).node
+ assert (
+ str(excinfo.value)
+ == 'Expected concat input streams to have length multiple of 2 (v=1, a=1); got 3'
+ )
def test_filter_asplit():
@@ -306,19 +343,23 @@ def test_filter_asplit():
'-i',
TEST_INPUT_FILE1,
'-filter_complex',
- '[0]vflip[s0];[s0]asplit=2[s1][s2];[s1]atrim=end=20:start=10[s3];[s2]atrim=end=40:start=30[s4];[s3]'
- '[s4]concat=n=2[s5]',
+ (
+ '[0]vflip[s0];'
+ '[s0]asplit=2[s1][s2];'
+ '[s1]atrim=end=20:start=10[s3];'
+ '[s2]atrim=end=40:start=30[s4];'
+ '[s3][s4]concat=n=2[s5]'
+ ),
'-map',
'[s5]',
TEST_OUTPUT_FILE1,
- '-y'
+ '-y',
]
def test__output__bitrate():
args = (
- ffmpeg
- .input('in')
+ ffmpeg.input('in')
.output('out', video_bitrate=1000, audio_bitrate=200)
.get_args()
)
@@ -327,28 +368,32 @@ def test__output__bitrate():
@pytest.mark.parametrize('video_size', [(320, 240), '320x240'])
def test__output__video_size(video_size):
- args = (
- ffmpeg
- .input('in')
- .output('out', video_size=video_size)
- .get_args()
- )
+ args = ffmpeg.input('in').output('out', video_size=video_size).get_args()
assert args == ['-i', 'in', '-video_size', '320x240', 'out']
def test_filter_normal_arg_escape():
- """Test string escaping of normal filter args (e.g. ``font`` param of ``drawtext`` filter)."""
+ """Test string escaping of normal filter args (e.g. ``font`` param of ``drawtext``
+ filter).
+ """
+
def _get_drawtext_font_repr(font):
- """Build a command-line arg using drawtext ``font`` param and extract the ``-filter_complex`` arg."""
- args = (ffmpeg
- .input('in')
+ """Build a command-line arg using drawtext ``font`` param and extract the
+ ``-filter_complex`` arg.
+ """
+ args = (
+ ffmpeg.input('in')
.drawtext('test', font='a{}b'.format(font))
.output('out')
.get_args()
)
assert args[:3] == ['-i', 'in', '-filter_complex']
assert args[4:] == ['-map', '[s0]', 'out']
- match = re.match(r'\[0\]drawtext=font=a((.|\n)*)b:text=test\[s0\]', args[3], re.MULTILINE)
+ match = re.match(
+ r'\[0\]drawtext=font=a((.|\n)*)b:text=test\[s0\]',
+ args[3],
+ re.MULTILINE,
+ )
assert match is not None, 'Invalid -filter_complex arg: {!r}'.format(args[3])
return match.group(1)
@@ -371,15 +416,15 @@ def _get_drawtext_font_repr(font):
def test_filter_text_arg_str_escape():
- """Test string escaping of normal filter args (e.g. ``text`` param of ``drawtext`` filter)."""
+ """Test string escaping of normal filter args (e.g. ``text`` param of ``drawtext``
+ filter).
+ """
+
def _get_drawtext_text_repr(text):
- """Build a command-line arg using drawtext ``text`` param and extract the ``-filter_complex`` arg."""
- args = (ffmpeg
- .input('in')
- .drawtext('a{}b'.format(text))
- .output('out')
- .get_args()
- )
+ """Build a command-line arg using drawtext ``text`` param and extract the
+ ``-filter_complex`` arg.
+ """
+ args = ffmpeg.input('in').drawtext('a{}b'.format(text)).output('out').get_args()
assert args[:3] == ['-i', 'in', '-filter_complex']
assert args[4:] == ['-map', '[s0]', 'out']
match = re.match(r'\[0\]drawtext=text=a((.|\n)*)b\[s0\]', args[3], re.MULTILINE)
@@ -404,14 +449,49 @@ def _get_drawtext_text_repr(text):
assert expected == actual
-#def test_version():
+# def test_version():
# subprocess.check_call(['ffmpeg', '-version'])
def test__compile():
out_file = ffmpeg.input('dummy.mp4').output('dummy2.mp4')
assert out_file.compile() == ['ffmpeg', '-i', 'dummy.mp4', 'dummy2.mp4']
- assert out_file.compile(cmd='ffmpeg.old') == ['ffmpeg.old', '-i', 'dummy.mp4', 'dummy2.mp4']
+ assert out_file.compile(cmd='ffmpeg.old') == [
+ 'ffmpeg.old',
+ '-i',
+ 'dummy.mp4',
+ 'dummy2.mp4',
+ ]
+
+
+@pytest.mark.parametrize('pipe_stdin', [True, False])
+@pytest.mark.parametrize('pipe_stdout', [True, False])
+@pytest.mark.parametrize('pipe_stderr', [True, False])
+@pytest.mark.parametrize('cwd', [None, '/tmp'])
+def test__run_async(mocker, pipe_stdin, pipe_stdout, pipe_stderr, cwd):
+ process__mock = mock.Mock()
+ popen__mock = mocker.patch.object(subprocess, 'Popen', return_value=process__mock)
+ stream = _get_simple_example()
+ process = ffmpeg.run_async(
+ stream,
+ pipe_stdin=pipe_stdin,
+ pipe_stdout=pipe_stdout,
+ pipe_stderr=pipe_stderr,
+ cwd=cwd,
+ )
+ assert process is process__mock
+
+ expected_stdin = subprocess.PIPE if pipe_stdin else None
+ expected_stdout = subprocess.PIPE if pipe_stdout else None
+ expected_stderr = subprocess.PIPE if pipe_stderr else None
+ (args,), kwargs = popen__mock.call_args
+ assert args == ffmpeg.compile(stream)
+ assert kwargs == dict(
+ stdin=expected_stdin,
+ stdout=expected_stdout,
+ stderr=expected_stderr,
+ cwd=cwd,
+ )
def test__run():
@@ -426,7 +506,9 @@ def test__run():
def test__run__capture_out(mocker, capture_stdout, capture_stderr):
mocker.patch.object(ffmpeg._run, 'compile', return_value=['echo', 'test'])
stream = _get_simple_example()
- out, err = ffmpeg.run(stream, capture_stdout=capture_stdout, capture_stderr=capture_stderr)
+ out, err = ffmpeg.run(
+ stream, capture_stdout=capture_stdout, capture_stderr=capture_stderr
+ )
if capture_stdout:
assert out == 'test\n'.encode()
else:
@@ -451,7 +533,9 @@ def test__run__error(mocker, capture_stdout, capture_stderr):
mocker.patch.object(ffmpeg._run, 'compile', return_value=['ffmpeg'])
stream = _get_complex_filter_example()
with pytest.raises(ffmpeg.Error) as excinfo:
- out, err = ffmpeg.run(stream, capture_stdout=capture_stdout, capture_stderr=capture_stderr)
+ out, err = ffmpeg.run(
+ stream, capture_stdout=capture_stdout, capture_stderr=capture_stderr
+ )
assert str(excinfo.value) == 'ffmpeg error (see stderr output for detail)'
out = excinfo.value.stdout
err = excinfo.value.stderr
@@ -487,24 +571,30 @@ def test__filter__custom():
stream = ffmpeg.filter(stream, 'custom_filter', 'a', 'b', kwarg1='c')
stream = ffmpeg.output(stream, 'dummy2.mp4')
assert stream.get_args() == [
- '-i', 'dummy.mp4',
- '-filter_complex', '[0]custom_filter=a:b:kwarg1=c[s0]',
- '-map', '[s0]',
- 'dummy2.mp4'
+ '-i',
+ 'dummy.mp4',
+ '-filter_complex',
+ '[0]custom_filter=a:b:kwarg1=c[s0]',
+ '-map',
+ '[s0]',
+ 'dummy2.mp4',
]
def test__filter__custom_fluent():
- stream = (ffmpeg
- .input('dummy.mp4')
+ stream = (
+ ffmpeg.input('dummy.mp4')
.filter('custom_filter', 'a', 'b', kwarg1='c')
.output('dummy2.mp4')
)
assert stream.get_args() == [
- '-i', 'dummy.mp4',
- '-filter_complex', '[0]custom_filter=a:b:kwarg1=c[s0]',
- '-map', '[s0]',
- 'dummy2.mp4'
+ '-i',
+ 'dummy.mp4',
+ '-filter_complex',
+ '[0]custom_filter=a:b:kwarg1=c[s0]',
+ '-map',
+ '[s0]',
+ 'dummy2.mp4',
]
@@ -513,16 +603,29 @@ def test__merge_outputs():
out1 = in_.output('out1.mp4')
out2 = in_.output('out2.mp4')
assert ffmpeg.merge_outputs(out1, out2).get_args() == [
- '-i', 'in.mp4', 'out1.mp4', 'out2.mp4'
- ]
- assert ffmpeg.get_args([out1, out2]) == [
- '-i', 'in.mp4', 'out2.mp4', 'out1.mp4'
+ '-i',
+ 'in.mp4',
+ 'out1.mp4',
+ 'out2.mp4',
]
+ assert ffmpeg.get_args([out1, out2]) == ['-i', 'in.mp4', 'out2.mp4', 'out1.mp4']
def test__input__start_time():
- assert ffmpeg.input('in', ss=10.5).output('out').get_args() == ['-ss', '10.5', '-i', 'in', 'out']
- assert ffmpeg.input('in', ss=0.0).output('out').get_args() == ['-ss', '0.0', '-i', 'in', 'out']
+ assert ffmpeg.input('in', ss=10.5).output('out').get_args() == [
+ '-ss',
+ '10.5',
+ '-i',
+ 'in',
+ 'out',
+ ]
+ assert ffmpeg.input('in', ss=0.0).output('out').get_args() == [
+ '-ss',
+ '0.0',
+ '-i',
+ 'in',
+ 'out',
+ ]
def test_multi_passthrough():
@@ -530,49 +633,53 @@ def test_multi_passthrough():
out2 = ffmpeg.input('in2.mp4').output('out2.mp4')
out = ffmpeg.merge_outputs(out1, out2)
assert ffmpeg.get_args(out) == [
- '-i', 'in1.mp4',
- '-i', 'in2.mp4',
+ '-i',
+ 'in1.mp4',
+ '-i',
+ 'in2.mp4',
'out1.mp4',
- '-map', '1',
- 'out2.mp4'
+ '-map',
+ '1',
+ 'out2.mp4',
]
assert ffmpeg.get_args([out1, out2]) == [
- '-i', 'in2.mp4',
- '-i', 'in1.mp4',
+ '-i',
+ 'in2.mp4',
+ '-i',
+ 'in1.mp4',
'out2.mp4',
- '-map', '1',
- 'out1.mp4'
+ '-map',
+ '1',
+ 'out1.mp4',
]
def test_passthrough_selectors():
i1 = ffmpeg.input(TEST_INPUT_FILE1)
- args = (
- ffmpeg
- .output(i1['1'], i1['2'], TEST_OUTPUT_FILE1)
- .get_args()
- )
+ args = ffmpeg.output(i1['1'], i1['2'], TEST_OUTPUT_FILE1).get_args()
assert args == [
- '-i', TEST_INPUT_FILE1,
- '-map', '0:1',
- '-map', '0:2',
+ '-i',
+ TEST_INPUT_FILE1,
+ '-map',
+ '0:1',
+ '-map',
+ '0:2',
TEST_OUTPUT_FILE1,
]
def test_mixed_passthrough_selectors():
i1 = ffmpeg.input(TEST_INPUT_FILE1)
- args = (
- ffmpeg
- .output(i1['1'].hflip(), i1['2'], TEST_OUTPUT_FILE1)
- .get_args()
- )
+ args = ffmpeg.output(i1['1'].hflip(), i1['2'], TEST_OUTPUT_FILE1).get_args()
assert args == [
- '-i', TEST_INPUT_FILE1,
+ '-i',
+ TEST_INPUT_FILE1,
'-filter_complex',
'[0:1]hflip[s0]',
- '-map', '[s0]',
- '-map', '0:2',
+ '-map',
+ '[s0]',
+ '-map',
+ '0:2',
TEST_OUTPUT_FILE1,
]
@@ -584,36 +691,56 @@ def test_pipe():
frame_count = 10
start_frame = 2
- out = (ffmpeg
- .input('pipe:0', format='rawvideo', pixel_format='rgb24', video_size=(width, height), framerate=10)
+ out = (
+ ffmpeg.input(
+ 'pipe:0',
+ format='rawvideo',
+ pixel_format='rgb24',
+ video_size=(width, height),
+ framerate=10,
+ )
.trim(start_frame=start_frame)
.output('pipe:1', format='rawvideo')
)
args = out.get_args()
assert args == [
- '-f', 'rawvideo',
- '-video_size', '{}x{}'.format(width, height),
- '-framerate', '10',
- '-pixel_format', 'rgb24',
- '-i', 'pipe:0',
+ '-f',
+ 'rawvideo',
+ '-video_size',
+ '{}x{}'.format(width, height),
+ '-framerate',
+ '10',
+ '-pixel_format',
+ 'rgb24',
+ '-i',
+ 'pipe:0',
'-filter_complex',
- '[0]trim=start_frame=2[s0]',
- '-map', '[s0]',
- '-f', 'rawvideo',
- 'pipe:1'
+ '[0]trim=start_frame=2[s0]',
+ '-map',
+ '[s0]',
+ '-f',
+ 'rawvideo',
+ 'pipe:1',
]
cmd = ['ffmpeg'] + args
- p = subprocess.Popen(cmd, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+ p = subprocess.Popen(
+ cmd,
+ stdin=subprocess.PIPE,
+ stdout=subprocess.PIPE,
+ stderr=subprocess.PIPE,
+ )
- in_data = bytes(bytearray([random.randint(0,255) for _ in range(frame_size * frame_count)]))
+ in_data = bytes(
+ bytearray([random.randint(0, 255) for _ in range(frame_size * frame_count)])
+ )
p.stdin.write(in_data) # note: this could block, in which case need to use threads
p.stdin.close()
out_data = p.stdout.read()
assert len(out_data) == frame_size * (frame_count - start_frame)
- assert out_data == in_data[start_frame*frame_size:]
+ assert out_data == in_data[start_frame * frame_size :]
def test__probe():
@@ -622,8 +749,73 @@ def test__probe():
assert data['format']['duration'] == '7.036000'
+@pytest.mark.skipif(sys.version_info < (3, 3), reason='requires python3.3 or higher')
+def test__probe_timeout():
+ with pytest.raises(subprocess.TimeoutExpired) as excinfo:
+ ffmpeg.probe(TEST_INPUT_FILE1, timeout=0)
+ assert 'timed out after 0 seconds' in str(excinfo.value)
+
+
def test__probe__exception():
with pytest.raises(ffmpeg.Error) as excinfo:
ffmpeg.probe(BOGUS_INPUT_FILE)
assert str(excinfo.value) == 'ffprobe error (see stderr output for detail)'
assert 'No such file or directory'.encode() in excinfo.value.stderr
+
+
+def test__probe__extra_args():
+ data = ffmpeg.probe(TEST_INPUT_FILE1, show_frames=None)
+ assert set(data.keys()) == {'format', 'streams', 'frames'}
+
+
+def get_filter_complex_input(flt, name):
+ m = re.search(r'\[([^]]+)\]{}(?=[[;]|$)'.format(name), flt)
+ if m:
+ return m.group(1)
+ else:
+ return None
+
+
+def get_filter_complex_outputs(flt, name):
+ m = re.search(r'(^|[];]){}((\[[^]]+\])+)(?=;|$)'.format(name), flt)
+ if m:
+ return m.group(2)[1:-1].split('][')
+ else:
+ return None
+
+
+def test__get_filter_complex_input():
+ assert get_filter_complex_input('', 'scale') is None
+ assert get_filter_complex_input('scale', 'scale') is None
+ assert get_filter_complex_input('scale[s3][s4];etc', 'scale') is None
+ assert get_filter_complex_input('[s2]scale', 'scale') == 's2'
+ assert get_filter_complex_input('[s2]scale;etc', 'scale') == 's2'
+ assert get_filter_complex_input('[s2]scale[s3][s4];etc', 'scale') == 's2'
+
+
+def test__get_filter_complex_outputs():
+ assert get_filter_complex_outputs('', 'scale') is None
+ assert get_filter_complex_outputs('scale', 'scale') is None
+ assert get_filter_complex_outputs('scalex[s0][s1]', 'scale') is None
+ assert get_filter_complex_outputs('scale[s0][s1]', 'scale') == ['s0', 's1']
+ assert get_filter_complex_outputs('[s5]scale[s0][s1]', 'scale') == ['s0', 's1']
+ assert get_filter_complex_outputs('[s5]scale[s1][s0]', 'scale') == ['s1', 's0']
+ assert get_filter_complex_outputs('[s5]scale[s1]', 'scale') == ['s1']
+ assert get_filter_complex_outputs('[s5]scale[s1];x', 'scale') == ['s1']
+ assert get_filter_complex_outputs('y;[s5]scale[s1];x', 'scale') == ['s1']
+
+
+def test__multi_output_edge_label_order():
+ scale2ref = ffmpeg.filter_multi_output(
+ [ffmpeg.input('x'), ffmpeg.input('y')], 'scale2ref'
+ )
+ out = ffmpeg.merge_outputs(
+ scale2ref[1].filter('scale').output('a'),
+ scale2ref[10000].filter('hflip').output('b'),
+ )
+
+ args = out.get_args()
+ flt_cmpl = args[args.index('-filter_complex') + 1]
+ out1, out2 = get_filter_complex_outputs(flt_cmpl, 'scale2ref')
+ assert out1 == get_filter_complex_input(flt_cmpl, 'scale')
+ assert out2 == get_filter_complex_input(flt_cmpl, 'hflip')
diff --git a/pyproject.toml b/pyproject.toml
new file mode 100644
index 00000000..de71e58d
--- /dev/null
+++ b/pyproject.toml
@@ -0,0 +1,15 @@
+[tool.black]
+skip-string-normalization = true
+target_version = ['py27'] # TODO: drop Python 2 support (... "Soon").
+include = '\.pyi?$'
+exclude = '''
+(
+ /(
+ \.eggs
+ | \.git
+ | \.tox
+ | \venv
+ | dist
+ )/
+)
+'''
diff --git a/requirements-base.txt b/requirements-base.txt
deleted file mode 100644
index 6afd1528..00000000
--- a/requirements-base.txt
+++ /dev/null
@@ -1,6 +0,0 @@
-future
-pytest
-pytest-mock
-pytest-runner
-sphinx
-tox
diff --git a/requirements.txt b/requirements.txt
index dfec4a10..f8b347e9 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,33 +1,40 @@
-alabaster==0.7.10
-apipkg==1.4
-Babel==2.5.1
-certifi==2017.7.27.1
+alabaster==0.7.12
+atomicwrites==1.3.0
+attrs==19.1.0
+Babel==2.7.0
+certifi==2019.3.9
chardet==3.0.4
docutils==0.14
-execnet==1.5.0
-funcsigs==1.0.2
-future==0.16.0
-idna==2.6
-imagesize==0.7.1
-Jinja2==2.9.6
-MarkupSafe==1.0
-mock==2.0.0
-pbr==4.0.3
-pluggy==0.5.2
-py==1.4.34
-Pygments==2.2.0
-pytest==3.2.3
-pytest-forked==0.2
-pytest-mock==1.10.0
-pytest-runner==3.0
-pytest-xdist==1.22.2
-pytz==2017.3
-requests==2.18.4
-six==1.11.0
+filelock==3.0.12
+future==0.17.1
+idna==2.8
+imagesize==1.1.0
+importlib-metadata==0.17
+Jinja2==2.10.1
+MarkupSafe==1.1.1
+more-itertools==7.0.0
+numpy==1.16.4
+packaging==19.0
+pluggy==0.12.0
+py==1.8.0
+Pygments==2.4.2
+pyparsing==2.4.0
+pytest==4.6.1
+pytest-mock==1.10.4
+pytz==2019.1
+requests==2.22.0
+six==1.12.0
snowballstemmer==1.2.1
-Sphinx==1.6.5
-sphinxcontrib-websupport==1.0.1
-tox==2.9.1
-typing==3.6.2
-urllib3==1.22
-virtualenv==15.1.0
+Sphinx==2.1.0
+sphinxcontrib-applehelp==1.0.1
+sphinxcontrib-devhelp==1.0.1
+sphinxcontrib-htmlhelp==1.0.2
+sphinxcontrib-jsmath==1.0.1
+sphinxcontrib-qthelp==1.0.2
+sphinxcontrib-serializinghtml==1.1.3
+toml==0.10.0
+tox==3.12.1
+urllib3==1.25.3
+virtualenv==16.6.0
+wcwidth==0.1.7
+zipp==0.5.1
diff --git a/setup.py b/setup.py
index d551e987..72f381cb 100644
--- a/setup.py
+++ b/setup.py
@@ -1,23 +1,27 @@
from setuptools import setup
from textwrap import dedent
-version = '0.1.16'
-download_url = 'https://github.com/kkroening/ffmpeg-python/archive/v{}.zip'.format(version)
+version = '0.2.0'
+download_url = 'https://github.com/kkroening/ffmpeg-python/archive/v{}.zip'.format(
+ version
+)
-long_description = dedent("""\
+long_description = dedent(
+ '''\
ffmpeg-python: Python bindings for FFmpeg
=========================================
:Github: https://github.com/kkroening/ffmpeg-python
:API Reference: https://kkroening.github.io/ffmpeg-python/
-""")
+'''
+)
file_formats = [
'aac',
'ac3',
'avi',
- 'bmp'
+ 'bmp',
'flac',
'gif',
'mov',
@@ -56,8 +60,6 @@
setup(
name='ffmpeg-python',
packages=['ffmpeg'],
- setup_requires=['pytest-runner'],
- tests_require=['pytest', 'pytest-mock'],
version=version,
description='Python bindings for FFmpeg - with complex filtering support',
author='Karl Kroening',
@@ -67,6 +69,16 @@
keywords=keywords,
long_description=long_description,
install_requires=['future'],
+ extras_require={
+ 'dev': [
+ 'future==0.17.1',
+ 'numpy==1.16.4',
+ 'pytest-mock==1.10.4',
+ 'pytest==4.6.1',
+ 'Sphinx==2.1.0',
+ 'tox==3.12.1',
+ ]
+ },
classifiers=[
'Intended Audience :: Developers',
'License :: OSI Approved :: Apache Software License',
@@ -80,5 +92,9 @@
'Programming Language :: Python :: 3.4',
'Programming Language :: Python :: 3.5',
'Programming Language :: Python :: 3.6',
+ 'Programming Language :: Python :: 3.7',
+ 'Programming Language :: Python :: 3.8',
+ 'Programming Language :: Python :: 3.9',
+ 'Programming Language :: Python :: 3.10',
],
)
diff --git a/tox.ini b/tox.ini
index f86ec4bb..98814078 100644
--- a/tox.ini
+++ b/tox.ini
@@ -4,7 +4,17 @@
# and then run "tox" from this directory.
[tox]
-envlist = py27, py34, py35, py36, pypy
+envlist = py27, py35, py36, py37, py38, py39, py310
+
+[gh-actions]
+python =
+ 2.7: py27
+ 3.5: py35
+ 3.6: py36
+ 3.7: py37
+ 3.8: py38
+ 3.9: py39
+ 3.10: py310
[testenv]
commands = py.test -vv
@@ -32,8 +35,10 @@ import ffmpeg
)
```
+## [API reference](https://kkroening.github.io/ffmpeg-python/)
+
## Complex filter graphs
-FFmpeg is extremely powerful, but its command-line interface gets really complicated really quickly - especially when working with signal graphs and doing anything more than trivial.
+FFmpeg is extremely powerful, but its command-line interface gets really complicated rather quickly - especially when working with signal graphs and doing anything more than trivial.
Take for example a signal graph that looks like this:
@@ -49,7 +54,7 @@ ffmpeg -i input.mp4 -i overlay.png -filter_complex "[0]trim=start_frame=10:end_f
Maybe this looks great to you, but if you're not an FFmpeg command-line expert, it probably looks alien.
-If you're like me and find Python to be powerful and readable, it's easy with `ffmpeg-python`:
+If you're like me and find Python to be powerful and readable, it's easier with `ffmpeg-python`:
```python
import ffmpeg
@@ -68,29 +73,45 @@ overlay_file = ffmpeg.input('overlay.png')
)
```
-`ffmpeg-python` takes care of running `ffmpeg` with the command-line arguments that correspond to the above filter diagram, and it's easy to see what's going on and make changes as needed.
+`ffmpeg-python` takes care of running `ffmpeg` with the command-line arguments that correspond to the above filter diagram, in familiar Python terms.
-Real-world signal graphs can get a heck of a lot more complex, but `ffmpeg-python` handles them with ease.
-
+Real-world signal graphs can get a heck of a lot more complex, but `ffmpeg-python` handles arbitrarily large (directed-acyclic) signal graphs.
## Installation
-The latest version of `ffmpeg-python` can be acquired via pip:
+### Installing `ffmpeg-python`
-```
+The latest version of `ffmpeg-python` can be acquired via a typical pip install:
+
+```bash
pip install ffmpeg-python
```
-It's also possible to clone the source and put it on your python path (`$PYTHONPATH`, `sys.path`, etc.):
-
+Or the source can be cloned and installed from locally:
```bash
-$ git clone git@github.com:kkroening/ffmpeg-python.git
-$ export PYTHONPATH=${PYTHONPATH}:ffmpeg-python
-$ python
->>> import ffmpeg
+git clone git@github.com:kkroening/ffmpeg-python.git
+pip install -e ./ffmpeg-python
+```
+
+> **Note**: `ffmpeg-python` makes no attempt to download/install FFmpeg, as `ffmpeg-python` is merely a pure-Python wrapper - whereas FFmpeg installation is platform-dependent/environment-specific, and is thus the responsibility of the user, as described below.
+
+### Installing FFmpeg
+
+Before using `ffmpeg-python`, FFmpeg must be installed and accessible via the `$PATH` environment variable.
+
+There are a variety of ways to install FFmpeg, such as the [official download links](https://ffmpeg.org/download.html), or using your package manager of choice (e.g. `sudo apt install ffmpeg` on Debian/Ubuntu, `brew install ffmpeg` on OS X, etc.).
+
+Regardless of how FFmpeg is installed, you can check if your environment path is set correctly by running the `ffmpeg` command from the terminal, in which case the version information should appear, as in the following example (truncated for brevity):
+
```
+$ ffmpeg
+ffmpeg version 4.2.4-1ubuntu0.1 Copyright (c) 2000-2020 the FFmpeg developers
+ built with gcc 9 (Ubuntu 9.3.0-10ubuntu2)
+```
+
+> **Note**: The actual version information displayed here may vary from one system to another; but if a message such as `ffmpeg: command not found` appears instead of the version information, FFmpeg is not properly installed.
## [Examples](https://github.com/kkroening/ffmpeg-python/tree/master/examples)
@@ -100,26 +121,20 @@ Here are a few:
- [Convert video to numpy array](https://github.com/kkroening/ffmpeg-python/blob/master/examples/README.md#convert-video-to-numpy-array)
- [Generate thumbnail for video](https://github.com/kkroening/ffmpeg-python/blob/master/examples/README.md#generate-thumbnail-for-video)
- [Read raw PCM audio via pipe](https://github.com/kkroening/ffmpeg-python/blob/master/examples/README.md#convert-sound-to-raw-pcm-audio)
+
- [JupyterLab/Notebook stream editor](https://github.com/kkroening/ffmpeg-python/blob/master/examples/README.md#jupyter-stream-editor)
-Feel free to report any bugs or submit feature requests.
+One of the best things you can do to help make `ffmpeg-python` better is to answer [open questions](https://github.com/kkroening/ffmpeg-python/labels/question) in the issue tracker. The questions that are answered will be tagged and incorporated into the documentation, examples, and other learning resources.
+
+If you notice things that could be better in the documentation or overall development experience, please say so in the [issue tracker](https://github.com/kkroening/ffmpeg-python/issues). And of course, feel free to report any bugs or submit feature requests.
-It's generally straightforward to use filters that aren't explicitly built into `ffmpeg-python` but if there's a feature you'd like to see included in the library, head over to the [issue tracker](https://github.com/kkroening/ffmpeg-python/issues).
+Pull requests are welcome as well, but it wouldn't hurt to touch base in the issue tracker or hop on the [Matrix chat channel](https://riot.im/app/#/room/#ffmpeg-python:matrix.org) first.
-Pull requests are welcome as well.
+Anyone who fixes any of the [open bugs](https://github.com/kkroening/ffmpeg-python/labels/bug) or implements [requested enhancements](https://github.com/kkroening/ffmpeg-python/labels/enhancement) is a hero, but changes should include passing tests.
+
+### Running tests
+
+```bash
+git clone git@github.com:kkroening/ffmpeg-python.git
+cd ffmpeg-python
+virtualenv venv
+. venv/bin/activate # (OS X / Linux)
+venv\bin\activate # (Windows)
+pip install -e .[dev]
+pytest
+```
+
+```python
+audio_left = (
+ ffmpeg
+ .input('audio-left.wav')
+ .filter('atrim', start=5)
+ .filter('asetpts', 'PTS-STARTPTS')
+)
+
+audio_right = (
+ ffmpeg
+ .input('audio-right.wav')
+ .filter('atrim', start=10)
+ .filter('asetpts', 'PTS-STARTPTS')
+)
+
+input_video = ffmpeg.input('input-video.mp4')
+
+(
+ ffmpeg
+ .filter((audio_left, audio_right), 'join', inputs=2, channel_layout='stereo')
+ .output(input_video.video, 'output-video.mp4', shortest=None, vcodec='copy')
+ .overwrite_output()
+ .run()
+)
+```
+
## [Jupyter Frame Viewer](https://github.com/kkroening/ffmpeg-python/blob/master/examples/ffmpeg-numpy.ipynb)
@@ -123,3 +153,111 @@ out.run()
+## [Tensorflow Streaming](https://github.com/kkroening/ffmpeg-python/blob/master/examples/tensorflow_stream.py)
+
+
+
+- Decode input video with ffmpeg
+- Process video with tensorflow using "deep dream" example
+- Encode output video with ffmpeg
+
+```python
+process1 = (
+ ffmpeg
+ .input(in_filename)
+ .output('pipe:', format='rawvideo', pix_fmt='rgb24', vframes=8)
+ .run_async(pipe_stdout=True)
+)
+
+process2 = (
+ ffmpeg
+ .input('pipe:', format='rawvideo', pix_fmt='rgb24', s='{}x{}'.format(width, height))
+ .output(out_filename, pix_fmt='yuv420p')
+ .overwrite_output()
+ .run_async(pipe_stdin=True)
+)
+
+while True:
+ in_bytes = process1.stdout.read(width * height * 3)
+ if not in_bytes:
+ break
+ in_frame = (
+ np
+ .frombuffer(in_bytes, np.uint8)
+ .reshape([height, width, 3])
+ )
+
+ # See examples/tensorflow_stream.py:
+ out_frame = deep_dream.process_frame(in_frame)
+
+ process2.stdin.write(
+ out_frame
+ .astype(np.uint8)
+ .tobytes()
+ )
+
+process2.stdin.close()
+process1.wait()
+process2.wait()
+```
+
+
+
+## [FaceTime webcam input (OS X)](https://github.com/kkroening/ffmpeg-python/blob/master/examples/facetime.py)
+
+```python
+(
+ ffmpeg
+ .input('FaceTime', format='avfoundation', pix_fmt='uyvy422', framerate=30)
+ .output('out.mp4', pix_fmt='yuv420p', vframes=100)
+ .run()
+)
+```
+
+## Stream from a local video to HTTP server
+
+```python
+video_format = "flv"
+server_url = "http://127.0.0.1:8080"
+
+process = (
+ ffmpeg
+ .input("input.mp4")
+ .output(
+ server_url,
+ codec = "copy", # use same codecs of the original video
+ listen=1, # enables HTTP server
+ f=video_format)
+ .global_args("-re") # argument to act as a live stream
+ .run()
+)
+
+```
+
+to receive the video you can use ffplay in the terminal:
+
+```
+$ ffplay -f flv http://localhost:8080
+```
+
+## Stream from RTSP server to TCP socket
+
+```python
+packet_size = 4096
+
+process = (
+ ffmpeg
+ .input('rtsp://%s:8554/default')
+ .output('-', format='h264')
+ .run_async(pipe_stdout=True)
+)
+
+while process.poll() is None:
+ packet = process.stdout.read(packet_size)
+ try:
+ tcp_socket.send(packet)
+ except socket.error:
+ process.stdout.close()
+ process.wait()
+ break
+```
diff --git a/examples/facetime.py b/examples/facetime.py
new file mode 100644
index 00000000..58d083ec
--- /dev/null
+++ b/examples/facetime.py
@@ -0,0 +1,8 @@
+import ffmpeg
+
+(
+ ffmpeg
+ .input('FaceTime', format='avfoundation', pix_fmt='uyvy422', framerate=30)
+ .output('out.mp4', pix_fmt='yuv420p', vframes=100)
+ .run()
+)
diff --git a/examples/graphs/dream.png b/examples/graphs/dream.png
new file mode 100644
index 00000000..090b530b
Binary files /dev/null and b/examples/graphs/dream.png differ
diff --git a/examples/graphs/mono-to-stereo.png b/examples/graphs/mono-to-stereo.png
new file mode 100644
index 00000000..92ce0e5b
Binary files /dev/null and b/examples/graphs/mono-to-stereo.png differ
diff --git a/examples/graphs/tensorflow-stream.png b/examples/graphs/tensorflow-stream.png
new file mode 100644
index 00000000..e5047a16
Binary files /dev/null and b/examples/graphs/tensorflow-stream.png differ
diff --git a/examples/split_silence.py b/examples/split_silence.py
index a889db10..90b46d95 100755
--- a/examples/split_silence.py
+++ b/examples/split_silence.py
@@ -27,10 +27,10 @@
parser.add_argument('--end-time', type=float, help='End time (seconds)')
parser.add_argument('-v', dest='verbose', action='store_true', help='Verbose mode')
-silence_start_re = re.compile(' silence_start: (?P
\ - Sort by:\ - best rated\ - newest\ - oldest\ -
\ -\ -
Add a comment\ - (markup):
\ -``code``, \ - code blocks:::and an indented block after blank line