|
I also use IE but not all my users do so I need to do it for both IE and NN. I already did it for IE using browser helper objects. I tried to research how downloaders catch clicks in NN and I found the following:
HKEY_CURRENT_USER\Software\Netscape\Netscape Navigator\Automation Protocols
I have FlashGet(also known as JetCar) installed and it is set to catch clicks in NN. So I have two string values under this key:
*Name/Type/Data*
ftp/REG_SZ/JetCar.Netscape
http/REG_SZ/JetCar.Netscape
I looked up JetCar.Netscape under HKEY_CLASSES_ROOT and found out ist GUID - {FB5DA723-162B-11D3-8B9B-AA70B4B0B524}. Then I looked up what interface has that GUID using OLE/COM Object Viewer and found out that it is IJetCarNetscape. So now it is clear that in order to catch clicks in NN I need to write a COM server. But to do that I need to find description of such interface (what methods and events should it incapsulate). But I cannot find this information anywhere on the internet
|
|
|
|
|
Can you please tell me or provide some links as to how you did this for IE. I need to do the exact same thing for IE.
|
|
|
|
|
As chris mentioned, the only way for this to work is by declaring your class member function as static.
|
|
|
|
|
Hi!
I'm trying to set the mouse to a custom cursor while it's in the client window.
I made a cursor with myhcur = CreateCursor(...)
I tried SetCursor(myhcur), but it got immediately changed back to the normal arrow cursor.
So then I tried changing the system cursor from arrow to my custom cursor, which I did with SetSystemCursor(myhcur, 32512). I used 32512 instead of OCR_NORMAL because the compiler thought it was an undefined symbol even when I'd included winuser.h. So now I have my own custom cursor instead of the arrow and everything works great.
The problem is that I can't switch back. I've tried each of
hnormal = (HCURSOR) LoadImage(0, (char *) 32512/*OCR_NORMAL*/, IMAGE_CURSOR, 0, 0, LR_DEFAULTSIZE);
and
hnormal = LoadCursor(0, IDC_ARROW);
But when I try to set these back when my program ends, SetSystemCursor(hnormal, 32512) returns no error but doesn't change the cursor back.  ... ... ...
|
|
|
|
|
Handle WM_SETCURSOR and call SetCursor() in response to that message.
--Mike--
http://home.inreach.com/mdunn/
"Holding the away team at bay with a non-functioning phaser was an act of unmitigated gall. I admire gall."
-- Lt. Cmdr. Worf
|
|
|
|
|
Thanks, I just finally found that.
I guess DefWindowProc was the culprit that kept changing the cursor back!
|
|
|
|
|
Currently I am writing the error catching and message displaying portions of my code. Now, although I know my supertight code will never fail and therefore is redundant and will never be used, I still wish to test it.
I am hoping there is some way I can skip chunks of code in the debugger and get directly to the error stuff I want, so I don't have to piss around with changing data or code just to do the testing.
From memory this was possible using Microsft C 6 under OS/2, but what about VC++ 6 under Windows 2000?
Michael Martin
Pegasystems Pty Ltd
Australia
martm@pegasystems.com
+61 413-004-018
|
|
|
|
|
I would like my program to know at all times:
- Is the mouse in the client area?
- If so then where is it?
And I want to do this without polling GetCursorPos.
A windows message like WM_MOUSEHASJUSTLEFT_YOURCLIENTAREA would be nice because then I could just use the last value I got from WM_MOUSEMOVE without worrying that the mouse has left the premises. Any suggestions?
Thanks!
-Alex
|
|
|
|
|
|
OK... I basically understand how the 'get' portion of a HTTP connection works... but now I have another question.
I am writing a program that will automaticalliy go through various CGI pages, entering the necessary input and parsing the output.
I should be able to handle the parsing, but my question lies in how exactly to 'post' the variables from a form to the CGI page.
EX: The first page has a form that posts a text field named 'Account' and a password field named 'Password' to "main.cgi".
I assume you would get a CHttpFile* that is to the main.cgi and do:
http_file->SendRequest
But it is here that I am not sure what to fill in...
What headers would I need to send? And MSDN says that the lpOptional would usually be the information for a POST operation... but what exactly would I put there?
Thanks for any help you can give!
Adam
cabadam@houston.rr.com
|
|
|
|
|
in.open(m_pFileName, ios::in, filebuf::sh_none);
CString strTemp;
while (!in.eof())
{
char ch;
in.get(ch);
strTemp += ch;
}
.. why does strTemp get an extra FFh char at the end?
-ryan
|
|
|
|
|
eof is one past the last character, equal to -1.
Christian
#include "std_disclaimer.h"
People who love sausage and respect the law should never watch either one being made.
The things that come to those who wait are usually the things left by those who got there first.
|
|
|
|
|
groove, thanks for the speedy response, ..
-ryan
|
|
|
|
|
I'd just like to make sure of something that I'm kind of worried about.
This is not ok:
char *Foo() {
char ret[100] = "hello";
return ret;
}
This is not ok:
char *Foo() {
char ret[100] = "hello";
return ret;
}
This is ok:
const char *Foo() {
return "hello world";
}
main() {
printf(Foo());
}
Is this right?
Thanks!
|
|
|
|
|
return "hello world"; is fine because the string literal "hello world" is stored in the executable image, and stays allocated for the life of your program.
--Mike--
http://home.inreach.com/mdunn/
"Holding the away team at bay with a non-functioning phaser was an act of unmitigated gall. I admire gall."
-- Lt. Cmdr. Worf
|
|
|
|
|
When you say
<br />
char anArray[200]<br />
you are saying "Grab me 200 characters worth of local space and call it "anArray". You could them go on to put "Hello World" into it if you wanted, but only by copying it in character by character, for example with:
<br />
strcpy(anArray, "Hello World!");<br />
However because the space for anArray is local it's discarded as soon as your function exits. A pointer returned to it will point to garbage, because that pointer looses it's validity when the routine exits.
The string constant "Hello World" is a different kind of beasty. The string is stored by the compiler in static storage. It's address remains valid for the life of the program.
When you say, for example,
<br />
const char *p = "Hello World";<br />
You are allocating a pointer. A memory location containing the address of that static string. Not a new block of memory containing the string itself (which is why it's ok to do a simple assignment in such a case rather than a more complex copy).
Does that make things clearer?
|
|
|
|
|
I need the directory in which the executable is in. I can't seem to find a function that just gets the path for the directory.
If the file path is... c:\program files\cool program\program.exe
I need c:\program files\cool program\
I was hoping to do this by a single function call rather than a hideuos solution by getting the whole path and deleting characters from the end until reaching a \
|
|
|
|
|
Why not call GetCurrentDirectory on startup, and store it ?
Christian
#include "std_disclaimer.h"
People who love sausage and respect the law should never watch either one being made.
The things that come to those who wait are usually the things left by those who got there first.
|
|
|
|
|
Because GetCurrentDirectory doesn't necessarily return the directory the exe is in. (i think)
I used the debugger and that function, and it showed c:\my documents, while the exe was in c:\my documents\C++ work\Application\Debug\
|
|
|
|
|
How odd - this is how we do it, and have done for as long as I have been here, and it's always worked fine. Are you running it apart from VC, or from within VC ?
Christian
#include "std_disclaimer.h"
People who love sausage and respect the law should never watch either one being made.
The things that come to those who wait are usually the things left by those who got there first.
|
|
|
|
|
I was running it from within VC, which is probably the problem then.
|
|
|
|
|
GetModuleFileName or GetModuleFileNameEx should work, or you can parse argv[0].
Isn't this in the FAQ yet? 
|
|
|
|
|
As I previously mentioned I wanted to get the directory the file is in, not the file name and its full path.
|
|
|
|
|
I think its a simple parse. Once you have the full path just strrchr for the last '\\', then the next to last, and what remains, however improbable, should be your directory.
later
oops - I think just one strrchr would do it - just read your msg again...
|
|
|
|
|
Or better yet, use stl strings.
string dirAndName = "c:\\Test\\SubDir\\File.Ext";
string directoryOnly = dirAndName.substr( 0, dirAndName.findLastOf('\\') );
Of course, this does no error checking, so if you think there might be no '\' character at all then you'd need to rework this slightly.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
