|
Dear Group,
I am trying to get the source (src) of images in web pages. I am using IHTMLElement2 (IE's API) to get the information regarding the html elements. I successfully extracted images, but i dont know how to extract the src of images.
for example: if you have,
then how would you extract "/intl/en_uk/images/logo.gif" using the IE's API !!
I know i can do that by searching through the html source code, and do a similarity match for the word (src) ... that would be too time consuming
does anyone know an easy better way ???
cheers
llp00na
|
|
|
|
|
llp00na wrote:
I am using IHTMLElement2 (IE's API) to get the information regarding the html elements. I successfully extracted images, but i dont know how to extract the src of images.
If you got the IHTMLElement2 for the image tag, try querying for IHTMLElement interface and then using getAttribute() on it.
--
jlr
http://jlamas.blogspot.com/[^]
|
|
|
|
|
llp00na wrote:
I know i can do that by searching through the html source code, and do a similarity match for the word (src) ... that would be too time consuming
It depends on the size of the page (i.e. the HTML) and whether image links are coded in plain ol' HTML vs. being injected into the document using JavaScript.
See the LinkChecker example in this[^] article. You may be able to use it with almost no code changes.
/ravi
My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com
|
|
|
|
|
|
Hi. Iwant to get the owner name of a file or directory. I hope to get the creators name.
Thus I used the GetFileSecurity to get the file security id and then the GetSecurityOwnerDescriptor to get the account security id. Then, I stack.
According to the manual you have to use the LookupAccountSide specifing two buffers as parameters: One to get the account name and another to get the domain name. Also according to them you need to give the current size of the buffers. The call will fail if one of buffer is not large enough and it will return the correct buffer sizes. It will also fail for other reasons like network failure etc.
Does anyone knows what is the error returned due to a buffer failure? Is it the same error for both buffers or a different one? How many calls do I need until I correct the buffers (by reading the sizes and allocating the buffers) until I get anyothe legitimate failure??
Why MicroSoft, NEVER, NEVER, NEVER documents the possible ways that a function may fail??? Is this a developers friendly company???
|
|
|
|
|
I use something like:
PSECURITY_DESCRIPTOR pSecurityDescriptor;
PSID pSidOwner;
SID_NAME_USE rSidNameUse;
DWORD dwAccountSize,
dwDomainSize;
char *pszAccount,
*pszDomain;
if (GetNamedSecurityInfo(
(LPSTR) lpszObjectName,
SE_FILE_OBJECT,
OWNER_SECURITY_INFORMATION,
&pSidOwner,
NULL,
NULL,
NULL,
&pSecurityDescriptor) == ERROR_SUCCESS)
{
dwAccountSize = 0;
dwDomainSize = 0;
LookupAccountSid(NULL, pSidOwner, NULL, &dwAccountSize, NULL, &dwDomainSize, &rSidNameUse);
pszAccount = new char[dwAccountSize];
pszDomain = new char[dwDomainSize];
if (LookupAccountSid(NULL, pSidOwner, pszAccount, &dwAccountSize, pszDomain, &dwDomainSize, &rSidNameUse) == TRUE)
...
delete [] pszAccount;
delete [] pszDomain;
LocalFree(pSecurityDescriptor);
}
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
Thanks. I do not think that the GetNamedSecurityInfo versus a combination GetFileSecurity,
GetSecurityOwnerDescriptor will make much difference. To my view if I have two functions I localise slightly the failure. In the GetFileSecurity I know that I faild to get the file SID whilst with GetSecurityOwnerDescriptor I know I faild to get the owner account SID. Unless the GetNamedSecurityInfo returns a different account SID from the proposed combination.
My real question is again the two calls of the LookupAccountSid. Like your case how do I know that the first one faild due to zero buffer size and not for any other reason (e.g. the file is across the network and the network went down when I was examining, or the file is been deleted and been left with a duggling pSidOwner). Also will one call set both buffer sizes or one of them and then will fail again because the second buffer is still NULL?
Any detailed documentation on the failures will be great.
Its an art to keep things simple
|
|
|
|
|
Dimitris Vikeloudas wrote:
Like your case how do I know that the first one faild due to zero buffer size and not for any other reason
It returns ERROR_INSUFFICIENT_BUFFER.
Dimitris Vikeloudas wrote:
Also will one call set both buffer sizes or one of them and then will fail again because the second buffer is still NULL?
They are both set to the necessary size.
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
At a breakpoint I see:
m_pParent = 0x003330d60 {CMainFrame hWnd=0x00060643 {unused=???}}
////
I am passing the MainFrame CWnd object into a constructor and assigning the object to m_pParent.
What I want to know is what the hex values assigned to m_pParent and hWnd mean...and why the values are different? Isn't hWnd a pointer to the Windows window??
Thanks.
Jerry
|
|
|
|
|
HWND is the handle of the window, a reference that is used by Windows to keep track of open windows. m_pParent is a pointer to the CWnd object, I see no reason why they should be the same?
Regards
Senthil
_____________________________
My Blog | My Articles | WinMacro
|
|
|
|
|
m_pParent is a pointer to an MFC CWnd-derived object. That object encapsulates a window and allows you to manipulate it through various methods and properties. One of those properties (member variables) is the handle to the window (hwnd), which is simply a unique id assigned by Windows to each window created. Therefore, your CWnd pointer and its hwnd value are never going to be the same.
Cheers,
Tom Archer - Visual C++ MVP
Archer Consulting Group
"So look up ahead at times to come, despair is not for us. We have a world and more to see, while this remains behind." - James N. Rowe
|
|
|
|
|
Hi,
I created an executable with Visual c++. At some point I had to change the
icon of the file (\res subdirectory). Now, the new icon was immediately
shown on the computer I use to run Visual c++. The problem is on two
computers that are used to evaluate the tool I am developing. No matter
what I do the old icon is used by Windows Explorer and is shown in the
top left corner of the application window.
Now, if I copy the executable file to a server on the network, the new
icon is used by Windows Explorer on some workstations and the old one
on the others. My guess is that a computer gets somehow infected by the
old icon by I can not prove that becose I have run out of Windows
workstations in our network
Regards
Andrej
|
|
|
|
|
voda wrote:
At some point I had to change the
icon of the file (\res subdirectory).
Exactly how did you go about changing it? My guess is that you only changed one of the resolutions (e.g., 16x16) rather than all of them (16x16, 32x32, 48x48).
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
Here is a start from MSDN "Customizing Icons":
In this example, the icon is in the c:\MyDir\MyProgram.exe file, with an index of two. The registry entry that assigns the custom icon to all .myp files is as shown here.
HKEY_CLASSES_ROOT
.myp
(Default) = MyProgram.1
MyProgram.1
(Default) = MyProgram Application
DefaultIcon
(Default) = C:\MyDir\MyProgram.exe,2
Finally, call SHChangeNotify to notify the Shell to update its cache.
I am not sure what parameters to use with the SHChangeNotify function. Maybe you can enlighten us when you figure it out?
|
|
|
|
|
I am trying to port a largish VC6 project to VC7/VC8 beta. Problem:
I have an .idl file with an interface definition using a PROPVARIANT structure (which is now defined in propidl.idl)
The resulting type library is imported in the client using an #import statement.
In VC7, the interface gets declared with a nnamespace_TLB::__tag_inner_PROPVARIANT type,which is not related to the PROPVARIANT coming from windows headers (which in turn breaks all my code)
Pandoras Gift #44: Hope. The one that keeps you on suffering.
aber.. "Wie gesagt, der Scheiss is' Therapie"
boost your code || Fold With Us! || sighist | doxygen
|
|
|
|
|
I'm writing application to connect one client to multiple servers (to send the same data to several servers).
I now how to make server to manipulate with many clients, but no how to send data to many servers in the proper way.
So, Is there a way to connect and send data to multiple servers from one client.
Do you have any example to help me?
Thank you.
|
|
|
|
|
You could round-robin the connections from your client to each server in turn, or create multiple threads, each of which sends data to a single client.
/ravi
My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com
|
|
|
|
|
Hi,
I have this problem I can't find much information about. Suppose I have the following function:
void test(float arg[][3])
{
}
...
float ** vertices = Allocate2DArray(10,3);
test(vertices);
...
float vertices2[10][3];
test(vertices2);
So basically in the first case it can't convert ** to [][3]. What's the difference? How can I cast it? I can't change the function because the real one is already written that way.
Finally a question about freeing the memory. Suppose I have
BYTE * temp = new BYTE[TotalSize]
float ** result = (float **) temp;
..followed by a loop to assign result[i] = (float *) to a location in temp[i*cols*sizeof(float)];
Now if I type delete[] result. Will that delete all the memory I allocated? After all I allocated a BYTE[], but then just did a bunch of assignments.
Thanks.
|
|
|
|
|
Budric B. wrote:
What's the difference?
Even though they can sometimes be used interchangeably, pointers and arrays are not the same thing. One way to achieve what you want is like:
double (*vertices)[3] = (double (*)[3]) malloc(10 * sizeof(*vertices));
for (int x = 0; x < 10; x++)
{
for (int y = 0; y < 3; y++)
vertices[x][y] = 0.0;
}
test(vertices);
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
FIRST PART:
If test(float* in){...}
vertices2 // is a pointer
test(vertices2) // ok
test(&vertices2) // error
vertices // is a pointer to a pointer
test(*vertices) // ok
test(vertices) // error
In this case a pointer to array of pointers, it gets a bit confusing.
float* pr = vertices[r];
float f = pr[c];
Therefore,
float f = (vertices[r])[c];
SECOND PART:
(*result)[i] == temp[i]
result[i] == &temp[i]
No the == are not typos.
I hope that helps.
INTP
"The more help VB provides VB programmers, the more miserable your life as a C++ programmer becomes."
Andrew W. Troelsen
|
|
|
|
|
So, what is the difference between ** and [x][y]? Well, one is a static memory location and the other are pointers.
float **x;
float y[10][3];
x is a pointer to a pointer. So, what the compiler does it it goes to the address of the variable X. It then reads this address and then goes to the address stored at that location. It then uses this address as the place to store the float!
If you understand single pointers you can see how the double pointer works:
First Address = *x;
Second Address = *First Address;
The Actual Float Value = *Second Address;
At any level you can actually allocate a "block" of pointers that you can index into at that time.
So what's the difference with an array?
y[1][2] = 12;
Is actually
y[3 + 2] = 12;
It's a single block of memory! There are no pointers! In the case of even a single float *, the address of the variable stores the address that then stores the actual value! In the case of an array, the address of the variable stores the actual value!
So, a multi-dimentional pointer is actually a single block of memory that is then index by the compiler as one block. So if you have: rows by columns:
float x[row][column]
each row is maxcolumns size. So if you want to go to rowx, coly, you could simply do rowx*maxcolumns + coly. This would index that block of memory.
The compiler knows how to generate the correct code if it's an array or pointer. Remember, an array's address stores the actual data. The pointer's address, stores another address. That address then stores the data (in a single indirection case).
8bc7c0ec02c0e404c0cc0680f7018827ebee
|
|
|
|
|
Hey. I'm new to MFC. Here's the situation... I've got a DLL that's using shared MFC. I just want a my dialog to appear when a method in the DLL is called.
I've declared
AFX_MANAGE_STATE(AfxGetStaticModuleState());
in the method. The dialog has a class Dialog that extends CDHtmlDialog. In the method I have
Dialog dlg;<br />
dlg.DoModal();
What else is needed to make it appear? Does there need to be an object extending CWinApp, or can I just use the dialog by itself? If there has to be a CWinApp, do I need to declare it in the method or global in the DLL (I've tried both ways and neither works.)?
Many thanks for any help.
|
|
|
|
|
You need to start your project as a "regular DLL", not as "extension DLL". Then you can, simly spoken, work as it would be a normal Application. It has a CWinApp derived application object and you can use it with any kind of application.
An extension Dll does not have a CWinApp derived object. It uses the one from the client application, which leads to soms restraints.
Serch MSDN for more information, my english is too poor to explain it better.
MS
|
|
|
|
|
It is a regular DLL. What I'm wondering is if I NEED the CWinApp object at all. Can I not just do something like
Dialog dlg;<br />
dlg.DoModal();
and have that dialog appear? I've also tried making the dialog appear when InitInstance() is called in a globally declared CWinApp.
Dialog dlg;<br />
m_pMainWnd = &dlg;<br />
int result = dlg.DoModal();<br />
This doesn't make the dialog appear either. This works in an MFC application I've made, but when the same code is used in the DLL it doesn't work.
Thanks.
|
|
|
|
|
1) You need a CWinApp derived object because it provides the main message pump, the idle loop, and so on.
2) If "Dialog" in your example is a CDialog derived class (with dialog resource assigned) it should work.
MS
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
