VapourSynth Editor

[PRAGMA]

How does one print debugging messages into the Log window?
`logging` module never prints anything regardless of level.

[_Al_]

I use couple of methods , maybe there is something within vsedit to allow it, someone might add something, or you can do couple of workarounds:

First workaround, you name your script as *.py and instead of clip.set_output() (although you can leave it there in your script), you add:

for frame in range(0, len(clip)):
    clip.get_frame(frame)

this will just make quick request for frames and script will go thru, much faster then actual previewing. You can request a specific frames , one frame or different range. You do not use vsedit, but your favorite python console.
So you do not use vsedit at all

SECOND, you can use *.py and code your previewer, it is not that difficult as you'd think, you just need to pick up a modul - openCV, PIL (using with tkinter) or PyQt (Qt in python). I use openCV or PyQt. vsedit uses Qt.

THIRD, you re-direct sys.stdout.
I use openCV and this script: outputwindow.py. All you do is just import it in your vapoursynth scrip:
import outputwindow
and your print (sys.stdout) is automatically redirected to extra tkinter GUI window. I found it a while ago on web and adjusted some lines so it even works with vsedit. So this is most comfortable method I guess.

#Python 3
"""
    named errorwindow originally
    Import this module into graphical Python apps to provide a
    sys.stderr. No functions to call, just import it. It uses
    only facilities in the Python standard distribution.

    If nothing is ever written to stderr, then the module just
    sits there and stays out of your face. Upon write to stderr,
    it launches a new process, piping it error stream. The new
    process throws up a window showing the error messages.
    
   Code derived from Bryan Olson's source posted in this related Usenet discussion:
   https://groups.google.com/d/msg/comp.lang.python/HWPhLhXKUos/TpFeWxEE9nsJ
   https://groups.google.com/d/msg/comp.lang.python/HWPhLhXKUos/eEHYAl4dH9YJ

   martineau - Modified to use subprocess.Popen instead of the os.popen
               which has been deprecated since Py 2.6. Changed so it
               redirects both stdout and stderr. Also inserted double quotes around paths
               in case they have embedded space characters in them, as
               they did on my Windows system.
               
   to use it with Preview() for openCV player:            
   -changed subprocess.Popen command to list instead of string , so it works under linux
   -added exception to catch window canceled by user and deleting pipe, so new GUI is automatically constructed again if needed,
   -added st.ScrolledText instead of Text
   -made sure that subprocess.Popen executable is python executable (or pythonw under windows),
    under windows, running it from Mystery Keeper's vsedit, sys.executable returned 'vsedit',
"""
import subprocess
import sys
import _thread as thread
import os

ERROR_FILENAME_LOG = 'error_printing_to_gui.txt'

if __name__ == '__main__':  # When spawned as separate process.
    # create window in which to display output
    # then copy stdin to the window until EOF
    # will happen when output is sent to each OutputPipe created
    import tkinter as tk
    import tkinter.scrolledtext as st
    from tkinter import BOTH, END, Frame, TOP, YES
    import tkinter.font as tkFont
    import queue as Queue

    Q_EMPTY = Queue.Empty  # An exception class.
    queue = Queue.Queue(1000)  # FIFO, first put first get

    def read_stdin(app, bufsize=4096):
        while True:
            queue.put(os.read(sys.stdin.fileno(), bufsize)) 

    class Application(Frame):
        def __init__(self, master, font_size=10, family='Courier', text_color='#0000AA', rows=25, cols=128):
            super().__init__(master)
            self.master = master
            if len(sys.argv) < 2:
                title = "Output stream from unknown source"
            elif len(sys.argv) < 3: 
                title = "Output stream from {}".format(sys.argv[1])
            else:  # Assume it's a least 3.
                title = "Output stream '{}' from {}".format(sys.argv[2], sys.argv[1])
            self.master.title(title)
            self.pack(fill=BOTH, expand=YES)
            font = tkFont.Font(family=family, size=font_size)
            width = font.measure(' ' * (cols+1))
            height = font.metrics('linespace') * (rows+1)
            self.configure(width=width, height=height)
            self.pack_propagate(0)  # Force frame to be configured size.

            self.logwidget = st.ScrolledText(self, font=font) 
            self.logwidget.pack(side=TOP, fill=BOTH, expand=YES)
            self.logwidget.configure(foreground=text_color)
            self.after(200, self.start_thread, ())  # Start polling thread.

        def start_thread(self, _):
            thread.start_new_thread(read_stdin, (self,))
            self.after(200, self.check_q, ())

        def check_q(self, _):
            go = True
            while go:
                try:
                    data = queue.get_nowait().decode()
                    if not data:
                        data = '[EOF]'
                        go = False
                    self.logwidget.insert(END, data)
                    self.logwidget.see(END)
                except Q_EMPTY:
                    self.after(200, self.check_q, ())
                    go = False
                    
    root = tk.Tk(baseName='whatever_name')
    app = Application(master=root)
    app.mainloop()

else: # when module is first imported
    import traceback

    class OutputPipe(object):
        def __init__(self, name=''):
            self.lock = thread.allocate_lock()
            self.name = name

        def flush(self):  # NO-OP.
            pass

        def __getattr__(self, attr):
            if attr == 'pipe':  # Attribute doesn't exist, so create it.
                # Launch this module as a separate process to display any output it receives
 
                executable = sys.executable                
                try:
                    basename = os.path.basename(executable)
                    name, _ = os.path.splitext(basename)
                    if not name.lower().startswith('python'):
                        executable = self.get_executable()                    
                except:
                    executable = self.get_executable()
                    
                argv1 = __file__
                    
                try:
                    argv2 = os.path.basename(sys.argv[0])
                except:
                    argv2 = ''
                argv3 = self.name
                
                command = [executable]
                for arg in [argv1, argv2, argv3]:
                    if arg:
                        command.append(arg)
                try:
                    # Had to also make stdout and stderr PIPEs too, to work with pythonw.exe
                    self.pipe = subprocess.Popen(command,
                                                 bufsize=0,
                                                 stdin=subprocess.PIPE,
                                                 stdout=subprocess.PIPE,
                                                 stderr=subprocess.PIPE).stdin
                except Exception:
                    # Output exception info to a file since this module isn't working.
                    exc_type, exc_value, exc_traceback = sys.exc_info()
                    msg = '{} exception in {}\n'.format(exc_type.__name__, os.path.basename(__file__))
                    with open(ERROR_FILENAME_LOG, 'wt') as info:
                        info.write('fatal error occurred spawning output process')
                        info.write('exeception info:' + msg)
                        traceback.print_exc(file=info)

                    sys.exit('fatal error occurred')

            return super(OutputPipe, self).__getattribute__(attr)
        
        def get_executable(self):
            #if running this within vsedit under windows sys.executable name is 'vsedit'
            return 'pythonw'
            
        def write(self, data):
            with self.lock:
                try:
                    data = data.encode()
                    self.pipe.write(data)  # First reference to pipe attr will cause an
                                           # OutputPipe process for the stream to be created.                                      
                except Exception:
                    #gui was canceled by user, piping would cause error
                    #pipe attr can be deleted so new is constructed with __getattr__() and therefore new GUI pops up if needed
                    del self.pipe
                    #pass

    try:
        os.remove(EXC_INFO_FILENAME)  # Delete previous file, if any.
    except Exception:
        pass

    # Redirect standard output streams in the process that imported this module.
    sys.stderr = OutputPipe('stderr')
    sys.stdout = OutputPipe('stdout')

[PRAGMA]

Originally Posted by _Al_

I use couple of methods , maybe there is something within vsedit to allow it, someone might add something, or you can do couple of workarounds:

First workaround, you name your script as *.py and instead of clip.set_output() (although you can leave it there in your script), you add:

for frame in range(0, len(clip)):
    clip.get_frame(frame)

this will just make quick request for frames and script will go thru, much faster then actual previewing. You can request a specific frames , one frame or different range. You do not use vsedit, but your favorite python console.
So you do not use vsedit at all

SECOND, you can use *.py and code your previewer, it is not that difficult as you'd think, you just need to pick up a modul - openCV, PIL (using with tkinter) or PyQt (Qt in python). I use openCV or PyQt. vsedit uses Qt.

THIRD, you re-direct sys.stdout.
I use openCV and this script: outputwindow.py. All you do is just import it in your vapoursynth scrip:
import outputwindow
and your print (sys.stdout) is automatically redirected to extra tkinter GUI window. I found it a while ago on web and adjusted some lines so it even works with vsedit. So this is most comfortable method I guess.

#Python 3
"""
    named errorwindow originally
    Import this module into graphical Python apps to provide a
    sys.stderr. No functions to call, just import it. It uses
    only facilities in the Python standard distribution.

    If nothing is ever written to stderr, then the module just
    sits there and stays out of your face. Upon write to stderr,
    it launches a new process, piping it error stream. The new
    process throws up a window showing the error messages.
    
   Code derived from Bryan Olson's source posted in this related Usenet discussion:
   https://groups.google.com/d/msg/comp.lang.python/HWPhLhXKUos/TpFeWxEE9nsJ
   https://groups.google.com/d/msg/comp.lang.python/HWPhLhXKUos/eEHYAl4dH9YJ

   martineau - Modified to use subprocess.Popen instead of the os.popen
               which has been deprecated since Py 2.6. Changed so it
               redirects both stdout and stderr. Also inserted double quotes around paths
               in case they have embedded space characters in them, as
               they did on my Windows system.
               
   to use it with Preview() for openCV player:            
   -changed subprocess.Popen command to list instead of string , so it works under linux
   -added exception to catch window canceled by user and deleting pipe, so new GUI is automatically constructed again if needed,
   -added st.ScrolledText instead of Text
   -made sure that subprocess.Popen executable is python executable (or pythonw under windows),
    under windows, running it from Mystery Keeper's vsedit, sys.executable returned 'vsedit',
"""
import subprocess
import sys
import _thread as thread
import os

ERROR_FILENAME_LOG = 'error_printing_to_gui.txt'

if __name__ == '__main__':  # When spawned as separate process.
    # create window in which to display output
    # then copy stdin to the window until EOF
    # will happen when output is sent to each OutputPipe created
    import tkinter as tk
    import tkinter.scrolledtext as st
    from tkinter import BOTH, END, Frame, TOP, YES
    import tkinter.font as tkFont
    import queue as Queue

    Q_EMPTY = Queue.Empty  # An exception class.
    queue = Queue.Queue(1000)  # FIFO, first put first get

    def read_stdin(app, bufsize=4096):
        while True:
            queue.put(os.read(sys.stdin.fileno(), bufsize)) 

    class Application(Frame):
        def __init__(self, master, font_size=10, family='Courier', text_color='#0000AA', rows=25, cols=128):
            super().__init__(master)
            self.master = master
            if len(sys.argv) < 2:
                title = "Output stream from unknown source"
            elif len(sys.argv) < 3: 
                title = "Output stream from {}".format(sys.argv[1])
            else:  # Assume it's a least 3.
                title = "Output stream '{}' from {}".format(sys.argv[2], sys.argv[1])
            self.master.title(title)
            self.pack(fill=BOTH, expand=YES)
            font = tkFont.Font(family=family, size=font_size)
            width = font.measure(' ' * (cols+1))
            height = font.metrics('linespace') * (rows+1)
            self.configure(width=width, height=height)
            self.pack_propagate(0)  # Force frame to be configured size.

            self.logwidget = st.ScrolledText(self, font=font) 
            self.logwidget.pack(side=TOP, fill=BOTH, expand=YES)
            self.logwidget.configure(foreground=text_color)
            self.after(200, self.start_thread, ())  # Start polling thread.

        def start_thread(self, _):
            thread.start_new_thread(read_stdin, (self,))
            self.after(200, self.check_q, ())

        def check_q(self, _):
            go = True
            while go:
                try:
                    data = queue.get_nowait().decode()
                    if not data:
                        data = '[EOF]'
                        go = False
                    self.logwidget.insert(END, data)
                    self.logwidget.see(END)
                except Q_EMPTY:
                    self.after(200, self.check_q, ())
                    go = False
                    
    root = tk.Tk(baseName='whatever_name')
    app = Application(master=root)
    app.mainloop()

else: # when module is first imported
    import traceback

    class OutputPipe(object):
        def __init__(self, name=''):
            self.lock = thread.allocate_lock()
            self.name = name

        def flush(self):  # NO-OP.
            pass

        def __getattr__(self, attr):
            if attr == 'pipe':  # Attribute doesn't exist, so create it.
                # Launch this module as a separate process to display any output it receives
 
                executable = sys.executable                
                try:
                    basename = os.path.basename(executable)
                    name, _ = os.path.splitext(basename)
                    if not name.lower().startswith('python'):
                        executable = self.get_executable()                    
                except:
                    executable = self.get_executable()
                    
                argv1 = __file__
                    
                try:
                    argv2 = os.path.basename(sys.argv[0])
                except:
                    argv2 = ''
                argv3 = self.name
                
                command = [executable]
                for arg in [argv1, argv2, argv3]:
                    if arg:
                        command.append(arg)
                try:
                    # Had to also make stdout and stderr PIPEs too, to work with pythonw.exe
                    self.pipe = subprocess.Popen(command,
                                                 bufsize=0,
                                                 stdin=subprocess.PIPE,
                                                 stdout=subprocess.PIPE,
                                                 stderr=subprocess.PIPE).stdin
                except Exception:
                    # Output exception info to a file since this module isn't working.
                    exc_type, exc_value, exc_traceback = sys.exc_info()
                    msg = '{} exception in {}\n'.format(exc_type.__name__, os.path.basename(__file__))
                    with open(ERROR_FILENAME_LOG, 'wt') as info:
                        info.write('fatal error occurred spawning output process')
                        info.write('exeception info:' + msg)
                        traceback.print_exc(file=info)

                    sys.exit('fatal error occurred')

            return super(OutputPipe, self).__getattribute__(attr)
        
        def get_executable(self):
            #if running this within vsedit under windows sys.executable name is 'vsedit'
            return 'pythonw'
            
        def write(self, data):
            with self.lock:
                try:
                    data = data.encode()
                    self.pipe.write(data)  # First reference to pipe attr will cause an
                                           # OutputPipe process for the stream to be created.                                      
                except Exception:
                    #gui was canceled by user, piping would cause error
                    #pipe attr can be deleted so new is constructed with __getattr__() and therefore new GUI pops up if needed
                    del self.pipe
                    #pass

    try:
        os.remove(EXC_INFO_FILENAME)  # Delete previous file, if any.
    except Exception:
        pass

    # Redirect standard output streams in the process that imported this module.
    sys.stderr = OutputPipe('stderr')
    sys.stdout = OutputPipe('stdout')

I cant get this to work on my end, when I do import outputwindow and then print("Test", file=sys.stdout) or stderror, nothing at all happens.

[_Al_]

that outputwindow.py takes care of it all, there is only import of that script needed and then just using print(), no arguments needed

from vapoursynth import core
import outputwindow
clip = core.std.BlankClip()
print('width for that blank clip is:', clip.width)

[PRAGMA]

Originally Posted by _Al_

that outputwindow.py takes care of it all, there is only import of that script needed and then just using print(), no arguments needed

from vapoursynth import core
import outputwindow
clip = core.std.BlankClip()
print('width for that blank clip is:', clip.width)

Yeah when I use that even, it doesn't work, it literally does nothing, no errors or anything.
Im on KDE Plasma (Linux), perhaps there's something to do with that?

[_Al_]

Possible, I tested it on Win7 and Ubuntu 18.04 though.
If pipe is not created it should write error into "error_printing_to_gui.txt".
I'd check it with just some simple *.py file first, like example above, not using vsedit. Then you can go further and try to write all variables into that error txt log as well thru out that outputwindow.py like for example for executable variable:

with open(ERROR_FILENAME_LOG, 'a') as info:
    info.write('sys.executable:')
    info.write(executable) #executable should be 'python' or 'pythonw' for windows

etc, because this is the only way to find out values when sys.stdout is redirected, or try to get rid of that:
sys.stderr = OutputPipe('stderr')
at the end of outputwindow script, it might start print errors into python consol, IDLE etc.

edit: corrected 'wt' into 'a' so it just adds to log

[PRAGMA]

Originally Posted by _Al_

Possible, I tested it on Win7 and Ubuntu 18.04 though.
If pipe is not created it should write error into "error_printing_to_gui.txt".
I'd check it with just some simple *.py file first, like example above, not using vsedit. Then you can go further and try to write all variables into that error txt log as well thru out that outputwindow.py like for example for executable variable:

with open(ERROR_FILENAME_LOG, 'a') as info:
    info.write('sys.executable:')
    info.write(executable) #executable should be 'python' or 'pythonw' for windows

etc, because this is the only way to find out values when sys.stdout is redirected, or try to get rid of that:
sys.stderr = OutputPipe('stderr')
at the end of outputwindow script, it might start print errors into python consol, IDLE etc.

edit: corrected 'wt' into 'a' so it just adds to log

No idea why, but it only works if its not in the site-packages directory and loaded elsewhere, I got it working by doing the following inside VS-Editor script:

sys.path.append(os.path.dirname(os.path.realpath(__file__)))
import outputwindow

Which will add the directory of the vpy file to temporary PATH, allowing it to import outputwindow.py if I put it next to my vpy script.
Thanks a ton!
Wish there was a built in way to do this.

P.S. I was doing raise Exception("BlaBla") as a logger but of course it would end the script so it wasnt perfect. Is there perhaps a way we can figure out how raise Exception works and just pony up a similar thing except raise Log("Msg") that doesnt exit()?

[_Al_]

I'm so sorry, I forgot about that already, adding that dir to sys.path, I'm not using vsedit now. vsedit needed that. If importing it just from some other *.py running script it was fine having that in site-packages dir.

That outputwindow.py works with any python app so it is usable elsewhere. Printing, sys.stdout.write(str(some value)) or just print(...), into tkinter gui, which is python standard library. And even that app crashes that windows stays on to report error.

[lansing]

I'm trying to build the project in Qt Creator in Windows 10 and I'm getting the error "vapoursynth/VapourSynth.h file not found" from the #include lines. It worked without a problem the last time I ran it in Windows 7. Had the path changed?

[Selur]

You probably need to adjust

INCLUDEPATH += 'C:/Program Files (x86)/VapourSynth/sdk/include/'

to match where Vapoursynth is located on your system.

[lansing]

Originally Posted by Selur

You probably need to adjust

INCLUDEPATH += 'C:/Program Files (x86)/VapourSynth/sdk/include/'

to match where Vapoursynth is located on your system.

Looks like the vs installation path did changed. I have the 64 bit version, before, the 64 bit version was stuffed inside "c:/program files(x86)/" folder, now it's in "c:/program files/".

In the vsedit.pro file, I tried changing the INCLUDEPATH under the win32{}, but it still reporting file not found.

[Selur]

You did:
1. edit the .pro file
2. Build->Run qmake
3. Build->Rebuild All
if you skipped the second step the changes of the .pro file might not have any effect.

[lansing]

Originally Posted by Selur

You did:
1. edit the .pro file
2. Build->Run qmake
3. Build->Rebuild All
if you skipped the second step the changes of the .pro file might not have any effect.

I tried that too, still not working.

Here's what I did:

- Opened the pro.pro file in Qt
- Under "projects" tab, set the "Desktop Qt 5.13.1 MingGW 64-bit" compiler as default
- Modified the "includepath" line in vsedit.pro to:

win32 {
	QT += winextras

        INCLUDEPATH += 'C:/Program Files/VapourSynth/sdk/include/'

- go build->run qmake
- go build->rebuild all

The problem still persist. It still complains about the vapoursynth.h file not found.

[lansing]

Okay I finally figured out the problem. I need to modify the other two .pro files in the project that has this includepath line, since I'm rebuilding all of them.

[Jukus]

When I use for preview:

haf.QTGMC(clip, Preset='Very Slow', Sharpness=0.8, FPSDivisor=1, TFF=True)
clip = core.std.Crop(clip, 0, 0, 2, 0)

That’s all right.
But when I use:

haf.QTGMC(clip, Preset='Very Slow', Sharpness=0.3, FPSDivisor=1, SourceMatch=3, Lossless=2, MatchEnhance=0.75, TFF=True)
clip = core.std.Crop(clip, 0, 0, 2, 0)

Then I get the error:

Error on frame 0 request:
Resize error 1027: image dimensions must be divisible by subsampling factor

[Selur]

seems to me like you are missing a 'clip = ' before the 'haf.QTGMC' part

[Jukus]

Originally Posted by Selur

seems to me like you are missing a 'clip = ' before the 'haf.QTGMC' part

No, I just did not copy it into the message.

[Selur]

Seems related to 'Lossless' using Lossless=0 removes the error.

[Selur]

It does, still this should be properly handled in QTGMC itself.

[Tohno_Neil]

Originally Posted by ~ VEGETA ~

I installed vapoursynth version 46 the x64 bit on windows 10 pro, and I have latest x64 version of vs editor. However I get this:



I used VS repo GUI to get the path for scripts and plugins and put it in vs edit but still no use. Here are the paths:



first one for plugins and the other for scripts.

anything to solve it?

thanks!

UPDATE: I solved it by installing python in C:\ and for all users then installing VS for all users and in C.

A better solution.

https://bitbucket.org/mystery_keeper...apoursynth-r48

Originally Posted by kypec

Sure, here is my build but I would recommend to have it built on your own Mint system. I don't think there are any special non-default dependencies needed, just follow my instructions and install any missing packages from official repositories.

Is this also works on windows?