Re: Stuck at FILE_ATTRIBUTE_REPARSE_POINT attribute - C / C++ / MFC Discussion Boards

Re: how to know file type and its protection type for office files?

Richard MacCutchan25-Dec-13 23:57

25-Dec-13 23:57

The file type can be guessed from the extension, although that does not guarantee the content. It is not possible to guess the encryption type; after all that is part of the reason for encryption, to hide information about the file's contents.

Veni, vidi, abiit domum

DirectShow - video resize interpolation

sdancer7523-Dec-13 22:09

sdancer75

23-Dec-13 22:09

Hi,

I am using DirectShow for playing video and VRM9 as a renderer. I want to resize my video but the result is a choppy with a kind of pixelated effect.

I want a smooth interpolated picture as you can see on different media players. Does VRM9 have something to enable or is something to implement by my own ?

Regards,

sdancer75

Re: DirectShow - video resize interpolation

sdancer757-Jan-14 23:20

sdancer75

7-Jan-14 23:20

Any suggestion here ? Is there any anti-aliasing method for the video playback ?

sdancer75

Re: DirectShow - video resize interpolation

sdancer754-Mar-14 21:52

sdancer75

4-Mar-14 21:52

Solved ! VMR9 under vista and later OS does not support interpolation. Microsoft removed this functionality to enforce programmers to use the new EVR renderer.

Shity thoughts from a company like MS!

sdancer75

Stuck at FILE_ATTRIBUTE_REPARSE_POINT attribute

vishalgpt23-Dec-13 19:39

vishalgpt

23-Dec-13 19:39

Hi to all, I am writing a program which will copy a directory tree and delete the source, every thing works as planned, but given that a reparse point could point anywhere (even network drives), one need to be careful when deleting recursively, it could end up deleting way more than on a target system.

How to deal with this situation when dealing with FILE_ATTRIBUTE_REPARSE_POINT, whether it is a recursive copy routine or delete routine.

Regards,
Vishal

Re: Stuck at FILE_ATTRIBUTE_REPARSE_POINT attribute

Richard MacCutchan23-Dec-13 22:11

Richard MacCutchan

23-Dec-13 22:11

It is not quite straightforward, you should review http://msdn.microsoft.com/en-us/library/windows/desktop/aa365503(v=vs.85).aspx[^].

Veni, vidi, abiit domum

Re: Stuck at FILE_ATTRIBUTE_REPARSE_POINT attribute

Randor 24-Dec-13 19:00

Randor

24-Dec-13 19:00

Hi,

1.) Call DeviceIoControl[^] with FSCTL_GET_REPARSE_POINT[^] and get a REPARSE_DATA_BUFFER[^].
2.) Call IsReparseTagMicrosoft[^] and check that the the REPARSE_DATA_BUFFER.ReparseTag[^] member is IO_REPARSE_TAG_SYMBOLIC_LINK[^] or IO_REPARSE_TAG_MOUNT_POINT[^].
3.) If so... begin copying the characters beginning at REPARSE_DATA_BUFFER.PrintNameOffset[^] and stop when you reach REPARSE_DATA_BUFFER.PrintNameLength[^].
4.) Now that you have the target path calculate with text comparisons whether or not operating on the target would cause infinite recursion. If your input is not normalized then you can apply C14N by calling PathCanonicalize function to remove navigation elements such as "\..\.\..\".

Best Wishes,
-David Delaune

Re: Stuck at FILE_ATTRIBUTE_REPARSE_POINT attribute

jschell25-Dec-13 8:03

jschell

25-Dec-13 8:03

vishalgpt wrote:
I am writing a program which will copy a directory tree and delete the source

Do you have an actual business case for this?

vishalgpt wrote:
How to deal with this situation when dealing with FILE_ATTRIBUTE_REPARSE_POINT

Depends on your business case but one very obvious and easy solution involves the following steps.
- Stop
- Report an error.

Some more complex solutions would be to span the tree before starting and refuse to continue if any problems are found.

Just curious what is your application going to do if the the files/directories are in use when you are deleting (which is probably much more likely than what you are asking.)

Little OT, but I am really puzzled by this.

Vaclav_23-Dec-13 15:12

Vaclav_

23-Dec-13 15:12

Been told by gurus that this is legal. It does compile in VC.
But since when?
It makes no sense to me. SO how big is the memory footprint?
Thanks for your time.

long unsigned int foo;

Re: Little OT, but I am really puzzled by this.

enhzflep23-Dec-13 21:02

enhzflep

23-Dec-13 21:02

I've read that the position of the word unsigned isn't important, though I'm not sure if this is an addition made over the years, or if it has been a part of c (&c++) since day 0.

sizeof(long) and sizeof(int) are both the same on a 32 bit system - 4. If memory serves me correctly, in the old days of my experience with 16-Bit turbo C++ 3.1, a long was 4 bytes long and an int was 2 bytes. Not sure if the sizeof operator returns the largest of the two or something else. sizeof(long long) for instance, is 8. So, I don't know if long int is a type in and of itself, as long long is, or if its seen as being a long modification of the int data type.

The StackOverflow community generally have someone online with the c/c++ specs seemingly committed to memory. Laugh | :laugh:

Interesting question. Thumbs Up | :thumbsup:

Re: Little OT, but I am really puzzled by this.

Richard MacCutchan23-Dec-13 22:09

Richard MacCutchan

23-Dec-13 22:09

I just tried this:

C++

long long unsigned int llui;
unsigned int ui;
unsigned long ul;
cout << "long long unsigned int: " << sizeof(llui) << endl;
cout << "          unsigned int: " << sizeof(ui) << endl;
cout << "         unsigned long: " << sizeof(ul) << endl;

resulting in:

long long unsigned int: 8
          unsigned int: 4
         unsigned long: 4

It appears that the compiler accepts all specifications and creates the largest item. The results are the same whether unsigned or not.

Veni, vidi, abiit domum

Re: Little OT, but I am really puzzled by this.

CPallini26-Dec-13 7:17

CPallini

26-Dec-13 7:17

Why do you think it is a nonsense?
long is a shortcut for long int, (like short for short int).
unsigned is a legal attribute and foo is a noble choice for a variable name.

You may also find it here: "Data Type Ranges" at MSDN[^].

Veni, vidi, vici.

Re: Little OT, but I am really puzzled by this.

Vaclav_26-Dec-13 8:20

Vaclav_

26-Dec-13 8:20

Because the "unsigned" should be the first specification. We do not write "int unsigned", so why is "long unsigned int" OK?

Re: Little OT, but I am really puzzled by this.

CPallini26-Dec-13 9:04

CPallini

26-Dec-13 9:04

Vaclav_Sal wrote:
We do not write "int unsigned", so why is "long unsigned int" OK?

We do, actually (at least we can do) (MSDN[^]):

The optional keywords signed and unsigned can precede or follow any of the integral types, except enum, and can also be used alone as type specifiers, in which case they are understood as signed int and unsigned int, respectively. When used alone, the keyword int is assumed to be signed. When used alone, the keywords long and short are understood as long int and short int.

Veni, vidi, vici.

Can't call a VB6 function from C

Arm5523-Dec-13 5:42

Arm55

23-Dec-13 5:42

Hello everybody, I'm new in this forum. I don't know where to place this thread, because it's about VB6 and C, but I think the problem is in the C code so here I am. I'm working with C with no special focus, so the solution can be in C++ too Smile | :)

Well, this is the problem. I have a VB6 program that simply calls a function when the Command1 is pressed. This one:

Public Function SampleSub(ByVal x As Integer, ByVal y As Integer) As Boolean
    MsgBox "Hello world!"
    SampleSub = True
End Function

I'm trying to call it from C like this:

C++

#include "main.h"

typedef BOOL (__stdcall *SampleSubPtr)(WORD, WORD);
static SampleSubPtr SampleSub;

DLL_EXPORT BOOL APIENTRY DllMain(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved)
{
    BOOL ret;

    if (fdwReason == DLL_PROCESS_ATTACH)
    {
            SampleSub = (SampleSubPtr)(0x00401B90);
            if (SampleSub == NULL)
                MessageBox(0, "SampleSub is NULL.", "M", 0);
            ret = SampleSub(5, 5);
    }

    return TRUE;
}

0x00401B90 is the pointer to the VB6 function. Then I place this DLL into the VB6 program memory space and the function is called. But it simply doesn't work. Debugging this with OllyDbg gives me this error when calling the function:

Access violation when reading [00000076]

I'm 100% sure that the function is at the address 0x00401B90. VB6 uses __stdcall and I use __stdcall from C. VB6 uses 16-bit integers and I use 16-bit integers (WORD) from C.

I can't solve this. Any ideas?

Thanks in advance! Smile | :)

Re: Can't call a VB6 function from C

David Crow23-Dec-13 5:44

David Crow

23-Dec-13 5:44

Is the SampleSub() function in a DLL?

"One man's wage rise is another man's price increase." - Harold Wilson

"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons

"Show me a community that obeys the Ten Commandments and I'll show you a less crowded prison system." - Anonymous

Re: Can't call a VB6 function from C

Arm5523-Dec-13 7:01

Arm55

23-Dec-13 7:01

Hello. Thanks for the answer. No, the SampleSub() function is inside a VB6 program, an executable. Then I build a C DLL with the code above and inject it on the VB6 process. Sorry if I don't explain that very well.

Thanks again. Smile | :)

Re: Can't call a VB6 function from C

Richard Andrew x6423-Dec-13 9:13

Richard Andrew x64

23-Dec-13 9:13

How are you determining the address of the VB6 function?

The difficult we do right away...
...the impossible takes slightly longer.

Re: Can't call a VB6 function from C

Arm5523-Dec-13 10:08

Arm55

23-Dec-13 10:08

Hi, thanks for your answer. I'm using:

AddressOf SampleSub

Also OllyDbg gives me that address.

Re: Can't call a VB6 function from C

Richard Andrew x6423-Dec-13 16:43

Richard Andrew x64

23-Dec-13 16:43

I would experiment with calling other VB6 functions to see if they give the same result. Especially try calling a function that takes no parameters to see if the problem might be the parameter passing.

Have you ever had success calling a VB6 function from C?

The difficult we do right away...
...the impossible takes slightly longer.

Re: Can't call a VB6 function from C

Arm5524-Dec-13 4:41

Arm55

24-Dec-13 4:41

Hi, thanks for your answer. I tried calling a function without parameters and without return value and it doesn't work. This is the VB function now:

Public Sub SampleSub()
    MsgBox "Hello world!"
End Sub

And the new C prototype:

C++

typedef void (__stdcall *SampleSubPtr)(void);

But the same Access violation error (that says OllyDbg).

In the past I was able to call a VB6 function from C, that's why I'm so angry! Frown | :(

But I just can't remeber how I did it. I'm sure I did it like I'm doing now, but I don't know why it doesn't work Hmmm | :|

. Now SampleSub() is at 0x00401BC0.

Do you think that the way of compiling, the language (if C or C++) or something like that is causing errors? Because I'm 90% sure that the code is the same that the one I used in the past to call a VB6 function, so I need to think that I'm compiling wrong or something like that. I'm compiling like this (or that's what CodeBlocks does for me Poke tongue | ;-P

mingw32-g++.exe  -Wall -DBUILD_DLL -O2     -c "C:\Path\main.cpp" -o obj\Release\main.o

And using extern "C" in the main function.

Thanks again Smile | :)

modified 24-Dec-13 10:49am.

Re: Can't call a VB6 function from C

Richard Andrew x6424-Dec-13 6:51

Richard Andrew x64

24-Dec-13 6:51

I'm afraid I don't know what it could be. However, I had always believed that it was not possible to arbitrarily call into a VB6 executable.

I was under the impression that the old VB worked on a, sort of, virtual execution scheme, and the VB program was under the control of a supervisor thread.

I wish I could be more helpful!

The difficult we do right away...
...the impossible takes slightly longer.

Re: Can't call a VB6 function from C

Arm5524-Dec-13 7:11

Arm55

24-Dec-13 7:11

Ok thank you very much, I'll keep trying and tell you if I could make the trick. If anybody have any idea please tell me!!

Thanks again! Smile | :)

Re: Can't call a VB6 function from C

Randor 24-Dec-13 18:12

Randor

24-Dec-13 18:12

Hmmm,

Btw, your original code has several bugs. The Visual Basic 6 Boolean is actually a VARIANT_BOOL[^] from the C++ programmers point of view. This is a difference of sizeof(short) versus sizeof(int). Also... most VB6 data types are actually VARIANTS ^

Best Wishes,
-David Delaune

[edit]
I remembered helping someone with this many years ago... searched a bit and found it:
VC++ & VB Inter Process Communication Plus Introduction To VB Subclassing[^]

Re: Can't call a VB6 function from C

Arm5525-Dec-13 9:25

Arm55

25-Dec-13 9:25

Hi Randor, thank you very much for your reply. I'm actually just using a VB6 "function" with no return value and that takes no parameters. This one:

Public Sub SampleSub()
    MsgBox "Hello world!"
End Sub

With the typedef void (__stdcall *SampleSubPtr)(void); prototype. What should I use instead of void? I think it will be better to start calling a simple function with no parameters, that's why I delete the old function that returns a VB boolean and takes two Integers as it params. Poke tongue | ;-P

I checked out the link your provided but that's for a Variant type in VB6, do you mean I should use VARIANT in C++ for any VB6 type?

I was thinking... I always get the Access violation when reading... error, do you think that a type incompatibility can cause this?

What would you do in my place? This problem is driving me crazy! Frown | :(

Thanks again! Smile | :)

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.