AviSource

From Avisynth wiki
(Difference between revisions)
Jump to: navigation, search
(SegmentedAviSource won't work (lead to desync) - see http://forum.doom9.org/showthread.php?p=1155530#post1155530)
Line 57: Line 57:
 
::In other words, if you don't specify anything, it will try to output [[YV24]]; if that isn't possible it tries [[YV16]], and if that isn't possible it tries [[YV12]], etc ...
 
::In other words, if you don't specify anything, it will try to output [[YV24]]; if that isn't possible it tries [[YV16]], and if that isn't possible it tries [[YV12]], etc ...
  
::For [[Planar|planar]] color formats, adding a '''<tt>'+'</tt>''' prefix, ''e.g.'' <code>AviSource(..., pixel_type="'''+'''YV12")</code>, tells AviSynth the video rows are DWORD aligned in memory instead of packed. '''This can fix skew or tearing of the decoded video''' when the width of the picture is not divisible by 4.
+
::For [[Planar|planar]] color formats, adding a '''<tt>'+'</tt>''' prefix, ''e.g.'' <code>AviSource(..., pixel_type="'''+'''YV12")</code>, tells AviSynth the video rows are DWORD aligned in memory instead of packed. '''This can fix skew or tearing of the decoded video''' with bad codecs when the width of the picture is not divisible by 4.
  
 
:{{Par2|fourCC|string|(auto from source)}}
 
:{{Par2|fourCC|string|(auto from source)}}
Line 126: Line 126:
 
  A = AviSource("FileA.avi") # with audio
 
  A = AviSource("FileA.avi") # with audio
 
  B = AviSource("FileB.avi") # no audio stream
 
  B = AviSource("FileB.avi") # no audio stream
  A ++ [[AudioDub]]B, [[BlankClip]](A)) # insert silent audio with same format
+
  A ++ [[AudioDub]](B, [[BlankClip]](A)) # insert silent audio with same format
  
 
  # Disable audio and request RGB32 decompression
 
  # Disable audio and request RGB32 decompression
Line 137: Line 137:
 
  AviSource("cap.avi", ''fourCC''="XVID")
 
  AviSource("cap.avi", ''fourCC''="XVID")
  
  # Open a YV12 video where the width is not a multiple of four
+
  # Open a YV12 video with a bad codec where the width is not a multiple of four
 
  Avisource("D:\Projects\test.avi", ''pixel_type''="+YV12")
 
  Avisource("D:\Projects\test.avi", ''pixel_type''="+YV12")
  

Revision as of 22:25, 21 January 2016

AviSource takes one or more files - not only AVI but also WAV (audio), AVS (AviSynth scripts), and VDR (VirtualDub frameserver) files. There is built-in Audio Compression Manager support for decoding compressed audio tracks (MP3, AAC, AC3, etc). If more than one file name is given, AviSource returns a clip with all files joined end to end. For the files to be joined, the media properties must be compatible - see Notes below.

AviSource tries to read the file(s) using Avisynth's built-in OpenDML interface (derived from VirtualDub code) where possible; if the file(s) are not in OpenDML format, it uses the Video for Windows AVIFile interface.

If AviSource has trouble with one or the other interface, OpenDMLSource and AviFileSource will force the use of OpenDML or AVIFile, respectively. Only OpenDML can read files larger than 2 GB, but only AVIFile can read non-AVI files (odd, given its name) like the ones listed in the opening paragraph. It can read any file (under 2 GB) for which there exist appropriate stream handlers.

WavSource will open a WAV file, or an audio-only AVI file. It will also return the audio stream from a normal video+audio AVI. This might be useful if your video stream is damaged or unreadable, or simply as a shortcut for AviSource(...).KillVideo.

Contents

Syntax and Parameters

AviSource(string filename [, ...] [, bool "audio", string "pixel_type", string "fourCC", int "vtrack", int "atrack"])
AviFileSource(string filename [, ...] [, bool "audio", string "pixel_type", string "fourCC", int "vtrack", int "atrack"])
OpenDMLSource(string filename [, ...] [, bool "audio", string "pixel_type", string "fourCC", int "vtrack", int "atrack"])

WavSource(string filename [, ...])

string  filename =
One or more file names. The files will be joined into a single clip with UnalignedSplice. To join files, the media properties must be compatible - see Notes below.
bool  audio = true
If true, load the first audio stream, or the stream specified by atrack if present. If false, audio is disabled.
string  pixel_type = "FULL"
Chooses the output color format of the decompressor. Valid values are listed in the table below. This argument has no effect if the video is uncompressed, as no decompressor will be used in that case.
  • If omitted or "FULL", AviSynth will use the first format supported by the decompressor, in the order shown in the table below.
  • If "AUTO", AviSynth will use the alternate (older) order as shown.
pixel_type   Color formats, listed by decoding priority (high to low)
FULL YV24 YV16 YV12 YV411 YUY2 RGB32 RGB24 Y8
AUTO YV12 YUY2 RGB32 RGB24 Y8
In other words, if you don't specify anything, it will try to output YV24; if that isn't possible it tries YV16, and if that isn't possible it tries YV12, etc ...
For planar color formats, adding a '+' prefix, e.g. AviSource(..., pixel_type="+YV12"), tells AviSynth the video rows are DWORD aligned in memory instead of packed. This can fix skew or tearing of the decoded video with bad codecs when the width of the picture is not divisible by 4.
string  fourCC = (auto from source)
Forces AviSynth to use a specific decoder instead of the one specified in the source file. See FourCC for more information.
int  vtrack = 0
Specifies a numbered video track. Track numbers start from zero, and are guaranteed to be continuous (i.e. there must be a track 1 if there is a track 0 and a track 2). If no video stream numbered vtrack exists, an error will be raised.
int  atrack = 0
Specifies a numbered audio track. Track numbers start from zero, and are guaranteed to be continuous (i.e. there must be a track 1 if there is a track 0 and a track 2). If no audio stream numbered atrack exists, no error will be raised, and no audio will be returned.


Notes

Joining clips

  • There is a limit (of about 50, sometimes fewer) AviSource calls in script - see discussion. If the limit is exceeded, you will see the error message
AVISource: couldn't locate a decompressor for fourcc ....
If you need to join more AVIs, try VirtualDub (limit > 700)
  1. the same height and width;
  2. the same color format (as presented by the decoder);
  3. the same frame rate (precisely the same, not approximately); and
  4. the same audio sample rate, bit depth and number of channels.

Windows 7 users

  • WavSource under Windows 7 is unable to load WAV files with 32-bit IEEE Float samples having the WAVEFORMAT structure. You can use ffmpeg to rewrite the header to an extensible format (just do a stream copy; it always writes extensible headers)
ffmpeg -i "bad.wav" -c copy "good.wav"

Helpful hints

  • Sometimes the colors will be distorted when loading a DivX clip in AviSynth v2.5 (the chroma channels U and V are swapped), due to a bug in DivX (5.02 and older). You can use SwapUV to correct it.
  • AVISource can also open DV type 1 video input (only video, not audio).
  • Some video files get decoded with the wrong color standard ('Rec601'/'Rec709') or luma range ('Full'/'TV'). This problem can arise if the input and output color formats are different, forcing a conversion. To avoid this conversion, try to specify another, compatible output format - for example:
  • If the video was encoded as RGB, try pixel_type="RGB24" or "RGB32";
  • If the video was encoded as YUV, try pixel_type="YV12", "YUY2" or "YV24".
If that does not work, try FFmpegSource, LSMASHSource or (if absolutely necessary) DirectShowSource.

More on codecs

Some reference threads:


Examples

# C programmers note: backslashes are not doubled; forward slashes work too
AVISource("d:\capture.avi")
WAVSource("f:/soundtrack.wav")
# Splice two clips together; the following statements do the same thing.
AviSource("cap1.avi") + AviSource("cap2.avi")
AVISource("cap1.avi", "cap2.avi")
# Splice two clips together where frame rates do not match.
A = AviSource("FileA.avi") # "29.97" fps (30000/1001)
B = AviSource("FileB.avi") # 30.0000 fps
A ++ B.AssumeFPS(A)
# Splice two clips together where one of them contains no audio.
A = AviSource("FileA.avi") # with audio
B = AviSource("FileB.avi") # no audio stream
A ++ AudioDub(B, BlankClip(A)) # insert silent audio with same format
# Disable audio and request RGB32 decompression
AVISource("cap.avi", audio=false, pixel_type="RGB32")
# Open a DV, forcing the Canopus DV Codec
AviSource("cap.avi", fourCC="CDVC")
# Open a file, forcing the XviD Codec
AviSource("cap.avi", fourCC="XVID")
# Open a YV12 video with a bad codec where the width is not a multiple of four
Avisource("D:\Projects\test.avi", pixel_type="+YV12")
# opens the first video and second audio stream of a clip
AviSource("D:\Projects\test_multi10.avi", vtrack=0, atrack=1)


Changes

v2.60 Added new color formats, "AUTO" and "FULL".
Added multiple video and audio stream support
Add '+' to pixel_type for padded planar support.
v2.55 Added fourCC option.
Personal tools