Avisynth+

[pinterf]

Originally Posted by StainlessS

EDIT: Perhaps 'n' is name, ie function name, variable name, device name.

'n' is for functioN

[StainlessS]

'n' is for functioN

Thanx.

Last one on list,

YPlaneMinMaxDifference    "c[threshold]fi"

Un-named optional final arg of type int ?

Whereas NEO

YPlaneMinMaxDifference    "c[threshold]f[offset]i"

EDIT: After ANY optional arg, so all following args have to be optional (standard amongst many languages) and I think that is enforced by Avisynth
even though 'i' type arg is neither specified as named optional, nor un-named optional, '[offset]i' or '[]i', however, how does it magically know that the arg name should be accepted as 'offset', when is not specified.
(I think some clever magic spell is being cast, but maybe should comply with standard prototype spec stuff, or some things will not work proper [eg AvsPMod]).
Avisynth+ may know what clever magic is involved, but as AVS exports function names in "$InternalFunctions$" (and "$PluginFunctions$"), and parameters via "$Plugin!" and "!Param$",
so would be best if software that interrogates those infos, get the un-magic'ed versions.

EDIT: From Wiki:- http://avisynth.nl/index.php/Interna...time_functions

 Color plane median, min, max, range

    YPlaneMedian(clip [, int offset = 0])   
    UPlaneMedian(clip [, int offset = 0])   
    VPlaneMedian(clip [, int offset = 0])   
    BPlaneMedian(clip [, int offset = 0])   AVS+ 
    GPlaneMedian(clip [, int offset = 0])   AVS+ 
    RPlaneMedian(clip [, int offset = 0])   AVS+ 

    YPlaneMin(clip [, float threshold = 0, int offset = 0])   
    UPlaneMin(clip [, float threshold = 0, int offset = 0])   
    VPlaneMin(clip [, float threshold = 0, int offset = 0])   
    BPlaneMin(clip [, float threshold = 0, int offset = 0])   AVS+ 
    GPlaneMin(clip [, float threshold = 0, int offset = 0])   AVS+ 
    RPlaneMin(clip [, float threshold = 0, int offset = 0])   AVS+ 

    YPlaneMax(clip [, float threshold = 0, int offset = 0])   
    UPlaneMax(clip [, float threshold = 0, int offset = 0])   
    VPlaneMax(clip [, float threshold = 0, int offset = 0])   
    BPlaneMax(clip [, float threshold = 0, int offset = 0])   AVS+ 
    GPlaneMax(clip [, float threshold = 0, int offset = 0])   AVS+ 
    RPlaneMax(clip [, float threshold = 0, int offset = 0])   AVS+ 

    YPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   
    UPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   
    VPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   
    BPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   AVS+ 
    GPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   AVS+ 
    RPlaneMinMaxDifference(clip [, float threshold, int offset = 0])   AVS+

[pinterf]

Originally Posted by StainlessS

Thanx.

Last one on list,

YPlaneMinMaxDifference    "c[threshold]fi"

Un-named optional final arg of type int ?

Whereas NEO

YPlaneMinMaxDifference    "c[threshold]f[offset]i"

Avisynth+ is also OK.

Avisynth+:
  {  "YPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_Y },
  {  "UPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_U }, // AVS+! was before: missing offset parameter
  {  "VPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_V }, // AVS+! was before: missing offset parameter
  {  "RPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_R },
  {  "GPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_G },
  {  "BPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_B },

[StainlessS]

Not sure that I'm understanding properley, think that you are saying that it works ok.

PLG="YPlaneMinMaxDifference"
PARAM=RT_PluginParam(PLG)
RT_Debugf("PLUG=%s PARAM=%s",PLG,PARAM)
return MessageClip("DONE")

result ( '[offset]' missing )

00000011	13:16:13.462	[2900] RT_DebugF: PLUG=YPlaneMinMaxDifference PARAM=c[threshold]fi

{ "YPlaneMinMaxDifference", BUILTIN_FUNC_PREFIX, "c[threshold]f[offset]i", MinMaxPlane::Create_minmax, (void *)PLANAR_Y },

"[offset]" has gone AWOL.

Source to RT_PluginParam [ env->AddFunction("RT_PluginParam", "s",RT_PluginParam, 0); ]

AVSValue __cdecl RT_PluginParam(AVSValue args, void* user_data, IScriptEnvironment* env) {
    char * myName="RT_PluginParam: ";
    const char *s=args[0].AsString();
    int len = int(strlen(s));
    char *bf = new char[len + 16];
    if(bf==NULL)    env->ThrowError("%sCannot allocate memory",myName);
    strcpy(bf,"$Plugin!");
    strcat(bf,s);
    strcat(bf,"!Param$");
    char *ps;
    try {
        AVSValue var  = env->GetVar(bf);
        if(!var.IsString()) {
            delete [] bf;
            env->ThrowError("%s Var '$Plugin!%s!Param$' is not a string",myName,s);
        }
        delete [] bf;
        ps = (char*)var.AsString();
    } catch (IScriptEnvironment::NotFound) {
        ps="";
    }
    return env->SaveString(ps);
}

Also, is there a way to extract all alterative parameter lists for a plugin name ? [there did not used to be for avs standard).

[pinterf]

When there are too many data, e.g. output is fast, OutputDebugString can lose characters.

[Groucho2004]

Originally Posted by pinterf

When there are too many data, e.g. output is fast, OutputDebugString can lose characters.

That's nasty.

[StainlessS]

Originally Posted by pinterf

When there are too many data, e.g. output is fast, OutputDebugString can lose characters.

Cannot possibly be the cause, the RT_Stats function list extractors do not use OutputDebugString at all, same problem.
I've never seen swallowed output in OutputDebugString (and I've often sent biggish text files to OutputDebugString for capture and checking)

function list extractors do not use OutputDebugString at all

At least not for extraction or processing or writing to file. [there is a problem somewhere]

EDIT: There are many others with exact same symptoms in the functionlist comparitor prev posted, YPlaneMinMaxDifference just happened to be the easiest to locate, being the last one in list.
From first few

                                                        Differences between Avisynth NEO and Avisynth+
                                                        ##############################################


AviSynth_Neo_0.1_(r2822,_Neo,_i386)_ORDERED_Function_List                                           AviSynth+_0.1_(r2772,_MT,_i386)_ORDERED_Function_List


AddProp                   "csn"
AlignedSplice             "cci"                                                                     AlignedSplice             "cc+"
AlignedSplice             "cci"


AverageChromaU            "c[offset]i"                                                              AverageChromaU            "ci"
AverageChromaV            "c[offset]i"                                                              AverageChromaV            "ci"

AverageLuma               "c[offset]i"                                                              AverageLuma               "ci"

Actually, It ALWAYS seems to happen with "[offset]i", every single one of them (and there are quite a few).

EDIT:
Wiki, seems to be missing offset args

    LumaDifference(clip1, clip2)   
    ChromaUDifference(clip1, clip2)   
    ChromaVDifference(clip1, clip2)   
    RGBDifference(clip1, clip2)   

    BDifference(clip1, clip2)   AVS+ 
    GDifference(clip1, clip2)   AVS+ 
    RDifference(clip1, clip2)   AVS+ 
    BDifferenceFromPrevious(clip)   AVS+ 
    GDifferenceFromPrevious(clip)   AVS+ 
    RDifferenceFromPrevious(clip)   AVS+ 

    YDifferenceFromPrevious(clip)   
    UDifferenceFromPrevious(clip)   
    VDifferenceFromPrevious(clip)   
    RGBDifferenceFromPrevious(clip)

Swallowed optional offset names [RHS text column, hope I got them all].

ChromaUDifference         "cc"                                                                      ChromaUDifference         "cci"
ChromaVDifference         "cc"                                                                      ChromaVDifference         "cci"

LumaDifference            "cc"                                                                      LumaDifference            "cci"

RGBDifference             "cc"                                                                      RGBDifference             "cci"
RGBDifferenceFromPrevious "c"                                                                       RGBDifferenceFromPrevious "ci"
RGBDifferenceToNext       "c[offset]i"                                                              RGBDifferenceToNext       "ci"

UDifferenceFromPrevious   "c"                                                                       UDifferenceFromPrevious   "ci"
UDifferenceToNext         "c[offset]i"                                                              UDifferenceToNext         "ci"
UPlaneMax                 "c[threshold]f[offset]i"                                                  UPlaneMax                 "c[threshold]fi"
UPlaneMedian              "c[offset]i"                                                              UPlaneMedian              "ci"
UPlaneMin                 "c[threshold]f[offset]i"                                                  UPlaneMin                 "c[threshold]fi"
UPlaneMinMaxDifference    "c[threshold]f[offset]i"                                                  UPlaneMinMaxDifference    "c[threshold]fi"

VDifferenceFromPrevious   "c"                                                                       VDifferenceFromPrevious   "ci"
VDifferenceToNext         "c[offset]i"                                                              VDifferenceToNext         "ci"
VPlaneMax                 "c[threshold]f[offset]i"                                                  VPlaneMax                 "c[threshold]fi"
VPlaneMedian              "c[offset]i"                                                              VPlaneMedian              "ci"
VPlaneMin                 "c[threshold]f[offset]i"                                                  VPlaneMin                 "c[threshold]fi"
VPlaneMinMaxDifference    "c[threshold]f[offset]i"                                                  VPlaneMinMaxDifference    "c[threshold]fi"

YDifferenceFromPrevious   "c"                                                                       YDifferenceFromPrevious   "ci"
YDifferenceToNext         "c[offset]i"                                                              YDifferenceToNext         "ci"
YPlaneMax                 "c[threshold]f[offset]i"                                                  YPlaneMax                 "c[threshold]fi"
YPlaneMedian              "c[offset]i"                                                              YPlaneMedian              "ci"
YPlaneMin                 "c[threshold]f[offset]i"                                                  YPlaneMin                 "c[threshold]fi"
YPlaneMinMaxDifference    "c[threshold]f[offset]i"                                                  YPlaneMinMaxDifference    "c[threshold]fi"

[pinterf]

Originally Posted by Groucho2004

That's nasty.

Probably the output is O.K., I experienced the missing lines in the viewer (DebugView)

[StainlessS]

Maybe old DebugView, update.

DebugView v4.81(Published: December 4, 2012):- https://docs.microsoft.com/en-us/sys...oads/debugview

[Maybe you use Kernel mode a lot, I dont]

EDIT: I switch off option WIN32 PIDS, and always Force Carriage Returns [If you dont, some text disappears off RHS when not sending '\n' at end of lines].

[wonkey_monkey]

I use DebugView too and have never had any problem with missing lines.

Clear-output string: When DebugView sees the special debug output string "DBGVIEWCLEAR" it clears the output.

Holy cow that's going to save me a lot of trouble.

[StainlessS]

Holy cow that's going to save me a lot of trouble.

I've never seen that before, making a note of it.

I used to use a Sinclair QL via RS232 for showing debug stuff (used to just switch on/plug in and auto started showing any debug stuff on 9" green screen), I kinda forgot bout that when I gave to charity shop 2 QL's
+ £1000's of softwares, next day I thought, what the hell do I do for debugging now. Maybe I set up debugview for same type output on laptop or 10" tablet laptoppy thingy.
I got a USB to Ethernet whatsit, but wonder if USB alone could be persuaded to work.
EDIT: I also used to view Debug stuff on 2nd monitor, but have not had dual display for a few years now, guess we needs one.

[StainlessS]

Seems to be a pretty serious bug in avisynth NEO, Main avs level local variables not visible inside Scriptclip.
Works fine even in Avs v2.58, Fails in AVS NEO < "I dont know what S Means. ([Scriptclip] Line 1)" >

    S=RT_TxtAddStr(S,S2)    # Some String
    W=1280 H=640 LINES=RT_TxtQueryLines(S)
    L= LINES*20 + H + 100
    RT_Debug(S)
    BlankClip(Width=W,Height=H,Length=L).ScriptClip("""RT_Subtitle("%s",S,x=10,y=height+100-current_frame,expx=true,expy=true)""")

Fixes in Avs NEO

    S=RT_TxtAddStr(S,S2)    # Some String
    W=1280 H=640 LINES=RT_TxtQueryLines(S)
    L= LINES*20 + H + 100
    RT_Debug(S)
    Global GLB_S=S
    BlankClip(Width=W,Height=H,Length=L).ScriptClip("""RT_Subtitle("%s",GLB_S,x=10,y=height+100-current_frame,expx=true,expy=true)""")

The Fix was already incorporated into the zip of FunctionLists/FunctionList_creator_scripts previously posted.

Perhaps we need a NEO thread.
Could probably be fixed using Grunt style ScriptClip(args="S") instead of Global.

EDIT: Alternative script OK in v2.58, but error in NEO [EDIT: Thanx Grouchy, its a breeze to switch avs back and forth now with Universal Avisynth Installer ]

S="HELLO"
BlankClip.ScriptClip("""Subtitle(S)""")

Maybe some GitHub guy could post in Neo whotsit.

EDIT: To post #4629

Maybe old DebugView, update.

DebugView v4.81(Published: December 4, 2012):- https://docs.microsoft.com/en-us/sys...oads/debugview

[Maybe you use Kernel mode a lot, I dont]

EDIT: I switch off option WIN32 PIDS, and always Force Carriage Returns [If you dont, some text disappears off RHS when not sending '\n' at end of lines].

EDIT: Above in BLUE why I now always use the below with auto add NewLine (rather than OutputDebugString directly).

int __cdecl dprintf(char* fmt, ...) {
    char printString[2048]="Shuffle: ";                             // Must be nul Termed, eg "Test: " or ""
    char *p=printString;
    for(;*p++;);
    --p;                                                            // @ null term
    va_list argp;
    va_start(argp, fmt);
    vsprintf(p, fmt, argp);
    va_end(argp);
    for(;*p++;);
    --p;                                                            // @ null term
    if(printString == p || p[-1] != '\n') {
        p[0]='\n';                                                  // append n/l if not there already
        p[1]='\0';
    }
    OutputDebugString(printString);
    return int(p-printString);                                      // strlen printString
}

[Stereodude]

Is it not possible to use two instances of the same filter that each use conditional reader in AVIsynth+ with unique conditional arguments for each?

This does not give the results I would expect.

LoadPlugin("C:\HDTV Tools\DGMPGDec\DGDecode-x86.dll")

MPEG2Source("YMMB_2.d2v", idct=5, moderate_h=40, moderate_v=40, cpu2="oooooo")

ConvertBits(16)

ColorYUV(conditional=true, levels="TV", analyze=false)
ConditionalReader("offu_t.txt", "coloryuv_off_u", false)
ConditionalReader("offv_t.txt", "coloryuv_off_v", false)

#some processing here

ColorYUV(conditional=true, levels="TV", analyze=false)
ConditionalReader("off_y.txt", "coloryuv_off_y", false)
ConditionalReader("gain_y.txt", "coloryuv_gain_y", false)

ConvertBits(8, dither = 1)

The output is the same as this:

LoadPlugin("C:\HDTV Tools\DGMPGDec\DGDecode-x86.dll")

MPEG2Source("YMMB_2.d2v", idct=5, moderate_h=40, moderate_v=40, cpu2="oooooo")

ConvertBits(16)

ColorYUV(conditional=true, levels="TV", analyze=false)
ConditionalReader("offu_t.txt", "coloryuv_off_u", false)
ConditionalReader("offv_t.txt", "coloryuv_off_v", false)
ConditionalReader("off_y.txt", "coloryuv_off_y", false)
ConditionalReader("gain_y.txt", "coloryuv_gain_y", false)

#some processing here

ColorYUV(conditional=true, levels="TV", analyze=false)
ConditionalReader("offu_t.txt", "coloryuv_off_u", false)
ConditionalReader("offv_t.txt", "coloryuv_off_v", false)
ConditionalReader("off_y.txt", "coloryuv_off_y", false)
ConditionalReader("gain_y.txt", "coloryuv_gain_y", false)

ConvertBits(8, dither = 1)

It seems that all 4 ConditionalReader arguments in the script are seen by each of the ColorYUV instances. Is there some way to prevent each instance of the ColorYUV filter from seeing all 4 conditional reader arguments?

[TheFluff]

Is that your actual script? To me it looks like you're reading the same files in both instances.

[Stereodude]

Originally Posted by TheFluff

Is that your actual script? To me it looks like you're reading the same files in both instances.

That's a simplified version of the actual script that shows the issue without having a pile of extra stuff in it. The actual script has other filters in between the two ColorYUV instances. Hence the #some processing here line.

I'm not sure what you're trying to say about the same files. Those two scripts have the exact same output. I'm asking if that's intentional. If so, can we have some sort of additional feature to segregate conditional reader statements in Avisynth+. Like a line that will be parsed by Avisynth+ and prevent all 4 ConditionalReader lines from being seen by both instances of the ColorYUV filter. Something that will cause conditional reader statements above the line to not be seen by filters below the line and vice versa.

[wonkey_monkey]

I have a filter which takes an array of floats as one of its arguments. It's instantiated like this:

AVSValue __cdecl Create_pyramid_mul(AVSValue args, void* user_data, IScriptEnvironment* env) {
	return new pyramid_mul(
		args[0].AsClip(),
		args[1],
		args[2].AsBool(false),
		env
	);
}

I want to create another function which only takes a single float argument, and constructs the array itself before calling the same constructor (below is just some test code which doesn't use the parameter):

AVSValue __cdecl Create_pyramid_blur(AVSValue args, void* user_data, IScriptEnvironment* env) {
	AVSValue multipliers[3] = { 0, 0, 1 };

	return new pyramid_mul(
		args[0].AsClip(),
		AVSValue(multipliers, 3),
		false,
		env
	);
}

But for some reason this doesn't work. Instead of receiving an array (0,0,1), the values the filter finds in the array are either 0, QNAN, or some random very large number.

Clearly my understanding of AVSValue objects and arrays is lacking. Can someone tell me the right way to do this?

---

Edit: is it because the AVSValue I'm sending to the filter - which I use in every GetFrame - is a pointer which becomes invalid after the constructor is called and the instantiator ends?

[TheFluff]

Originally Posted by Stereodude

That's a simplified version of the actual script that shows the issue without having a pile of extra stuff in it. The actual script has other filters in between the two ColorYUV instances. Hence the #some processing here line.

I'm not sure what you're trying to say about the same files. Those two scripts have the exact same output. I'm asking if that's intentional. If so, can we have some sort of additional feature to segregate conditional reader statements in Avisynth+. Like a line that will be parsed by Avisynth+ and prevent all 4 ConditionalReader lines from being seen by both instances of the ColorYUV filter. Something that will cause conditional reader statements above the line to not be seen by filters below the line and vice versa.

What I mean is that both sets of calls to conditionalreader seem to be reading from the same input files. I'm asking if this is a mistake in your example script or not. If it's not a mistake, well, you would expect two function calls with the same parameters to give the same results, no?

[pinterf]

Originally Posted by wonkey_monkey

AVSValue __cdecl Create_pyramid_blur(AVSValue args, void* user_data, IScriptEnvironment* env) {
	AVSValue multipliers[3] = { 0, 0, 1 };

	return new pyramid_mul(
		args[0].AsClip(),
		AVSValue(multipliers, 3),
		false,
		env
	);
}

What happens if you try with floats?
AVSValue multipliers[3] = { 0.0f, 0.0f, 1.0f };

[wonkey_monkey]

Same thing. It seems the object only exists during construction of the filter instance, whereas actual proper filter parameters persist. I made the filter take a copy during construction, and now it works as intended.

[Stereodude]

Originally Posted by TheFluff

What I mean is that both sets of calls to conditionalreader seem to be reading from the same input files. I'm asking if this is a mistake in your example script or not. If it's not a mistake, well, you would expect two function calls with the same parameters to give the same results, no?

They're not the same in the script with two and two. They have similar names, but they're not the same.