|
|
After each action that can be undone, push the current state of the document onto a stack. Each time an "undo" is requested, pop the most recent document state off the stack and notify the view that a "new" document needs to be rendered.
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
If this is a grphic oriented application, I agree that every primitive has a counter to track the refernece number. This is text processing, MS-Word like application. Is it resonable to change the document state everytime when you type a character?
Thank you for response
|
|
|
|
|
qudayong wrote:
Is it resonable to change the document state everytime when you type a character?
I don't know if it's a good idea or not. I've never had the need to implement such a feature so I'm not sure how I'd actually do it. It was just a guess.
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
Just think of the stack:
character: rowpos, colnumber
For graphics you could have something like:
OperationType, MetaData
For example. So it's just a stack of commands. You could even determine that a "word" is basically a state, so once an entire word is typed you put it on the stack.
So, some considerations could be:
1. How much data are you storing per-letter/state operation?
2. How long of a history buffer do you want?
If you store a letter as being character, followed by rowposition, column number you are actually storing 2 bytes (unicode let's say), then 2 bytes (say you can't have a line longer than 65536), then say 4 bytes (you have up to 4 billion lines in the document).
2+2+4 = 8 bytes per character operation. If you type 10,000 words and say each word is an average of 7 characters, that's 70,000 characters, that's 560,000 bytes or about 560k of undo changes. That's also 70,000 entries in your undo list. Not to mention possible optimizations in referencing and things or combining operations that cancel each other out.
How many does word go back to?
8bc7c0ec02c0e404c0cc0680f7018827ebee
|
|
|
|
|
One option (which is required for usability) is to group words into single undo entries. Otherwise, just undoing single letters gets to be really annoying. Anyway, most people use backspace to correct single letter errors, not undo.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
Thank you. You opinion ia valuable
|
|
|
|
|
for text, it's pretty easy:
when the user performs an action, save the action in a stack, then perform the action on the document. for undo, pop the last action off the stack and reverse the action.
the trick is to define 'action' in such a way that you can undo it.
ex. for deleting text, save the deleted text and its position as the action (to undo, insert the deleted text at the position). for inserting text, save the position and the number of characters added (to undo, remove the number of characters at the position).
Cleek | Image Toolkits | Thumbnail maker
|
|
|
|
|
Suppose you have 20 pages file.
You last action is delete 19 pages, and the perform undo. If you perform this action servral times, your stack will be overloaded.
|
|
|
|
|
that's why programs put up messages like "Insufficient space in the Undo buffer to save this action. Do you wish to proceed anyway?" (from the Visual Studio editor)
Cleek | Image Toolkits | Thumbnail maker
|
|
|
|
|
that's not usually true - you can't undo an undo - when you do the first undo, the stack is emptied
|
|
|
|
|
Having a test on MS-WORD, I find that you delete 20 page then perform undo. After 30 times there's no complain about memory and the action performs very fast. So I assume, there must has other way than "command pattern" to perform such an action.
cheers
|
|
|
|
|
I might as well say it before the flaming starts.... if you have a specific question to ask, I am sure that someone here would be more than happy to help.... The first thing you will hear is "do your own homework"...
So why can't YOU get it to run in VC++?
Artificial intelligence is no match for natural
stupidity.
Never argue with an idiot. They will drag you down to their level where they are an expert.
|
|
|
|
|
My VC++6 & MS DEV is somewhat out of order, I tried to re-load but was unsuccessful; I need to change the TEST file html code so that the student name and # appears beneath the clock; if I do and try to save it I get just under 20 errors when I attempt to run(!); I am supposed to copy and paste the TEST html code to Notepad or Wordpad, save as html to a deskcop icon; html code to change and save:
ERROR
Betty Hope<h 2=""> ERROR
898654712<h 2=""> ERROR
Betty Hope
898654712
ALSO WINZIPPED it and another friend could not unzip it properly to have a look. thanks for any help.
|
|
|
|
|
I have created an executive file in the form of CMemFile In an MFC application, which is directly copied in memory and is not on the hard disk. Now, how to execute this file directly from the memory? I tried to use the ShellExecute method, however, it does not accept the CMemFile giving an error saying that it cannot convert CMemFile to LPCSTR.
Is there any method to directly execute the exe file from the memory?
Regards
|
|
|
|
|
How does CMemFile store its data? Most likely it stores a pointer to the memory that contains the actual data. Look in the Microsoft documentation for CMemFile to see how it stores the data. Hopefully, CMemFile has a method to provide this pointer. An LPCSTR is a pointer to a string, so method could be called and the result could be cast to an LPCSTR if it is not already returned that way. Then this could be provided to the ShellExecute function.
David Spain, C++ Applications Programmer
|
|
|
|
|
David Spain wrote:
How does CMemFile store its data? Most likely it stores a pointer to the memory that contains the actual data. Look in the Microsoft documentation for CMemFile to see how it stores the data. Hopefully, CMemFile has a method to provide this pointer. An LPCSTR is a pointer to a string, so method could be called and the result could be cast to an LPCSTR if it is not already returned that way. Then this could be provided to the ShellExecute function.
I don't think so. You might obtain the CMemFile data using Detach, which returns a BYTE* and also closes the file. The return type is a BYTE* because a file may contain any type of data, not just text. Casting it to an LPCTSTR makes no sense: it may fool the compiler into accepting it, but the content will still be the same, and if it wasn't text, no one interpreting it as such will have correct results. In this case, in fact, it's an executable (binary) file.
Besides, ShellExecute expects to receive the name of the file path and name, not the content of the file.
--
jlr
http://jlamas.blogspot.com/[^]
|
|
|
|
|
Jose Lamas Rios wrote:
I don't think so. You might obtain the CMemFile data using Detach, which returns a BYTE* and also closes the file. The return type is a BYTE* because a file may contain any type of data, not just text. Casting it to an LPCTSTR makes no sense: it may fool the compiler into accepting it, but the content will still be the same, and if it wasn't text, no one interpreting it as such will have correct results. In this case, in fact, it's an executable (binary) file.
You are right. What I wrote actually made no sense. Obviously the ShellExecute function wants a string pointer, that points to the name of an actual file on disk.
As simply another idea, an executable in memory seems similar to a running process. Would using CreateProcess or something similar be a better avenue to pursue instead of ShellExecute?
David Spain, C++ Applications Programmer
|
|
|
|
|
Hi all,
I need to authenticate myself to a proxy. I have the domain\username and password.
I set it like this:
<br />
InternetSetOption((HINTERNET)(*pHttpFile), INTERNET_OPTION_USERNAME, <br />
sUsername, sUsername.GetLength());<br />
InternetSetOption((HINTERNET)(*pHttpFile), INTERNET_OPTION_PASSWORD, <br />
sPassword, sPassword.GetLength());
However, I still get the 407 error code.
Any idea how to set username and password for a file when behind a proxy?
Thanks!
Geert
Want to spread the newest version of your software automatically? Use Updater!
Visit my website: www.gvhsoftware.org
|
|
|
|
|
I am trying to get a rebar added to an existing MDI window, I am using the following code:
<br />
INITCOMMONCONTROLSEX icex;<br />
<br />
icex.dwSize = sizeof(INITCOMMONCONTROLSEX);<br />
icex.dwICC = ICC_COOL_CLASSES|ICC_BAR_CLASSES;<br />
InitCommonControlsEx(&icex);<br />
<br />
HINSTANCE hInst = (HINSTANCE)GetWindowLong(hParentHandle, GWL_HINSTANCE);<br />
hToolbar = CreateWindowEx(WS_EX_TOOLWINDOW,REBARCLASSNAME,NULL,WS_VISIBLE | WS_BORDER | WS_CHILD | WS_CLIPCHILDREN | WS_CLIPSIBLINGS | RBS_VARHEIGHT | RBS_BANDBORDERS | CCS_NODIVIDER | CCS_NOPARENTALIGN,0,0,0,0,hParentHandle,NULL,hInst,NULL);<br />
if( !hToolbar ) {<br />
MessageBox(0, "Could not create toolbar control", "Error", MB_OK | MB_ICONHAND);<br />
return;<br />
}<br />
<br />
SetWindowLong(hWndBar, GWL_STYLE, WS_VISIBLE | WS_CHILD | TBSTYLE_TOOLTIPS | CCS_NOPARENTALIGN | CCS_NORESIZE);<br />
SetWindowPos(hWndBar, 0, 0, 0, 0, 0, SWP_NOMOVE | SWP_NOSIZE | SWP_NOZORDER | SWP_FRAMECHANGED);<br />
<br />
RECT rect;<br />
GetWindowRect (hWndBar, &rect);<br />
<br />
REBARINFO rbi;<br />
rbi.cbSize = sizeof(REBARINFO);<br />
rbi.fMask = NULL;<br />
SendMessage(hToolbar, RB_SETBARINFO, 0, (LPARAM)&rbi);<br />
<br />
REBARBANDINFO rbb;<br />
rbb.cbSize = sizeof(REBARBANDINFO);<br />
rbb.fMask = RBBIM_SIZE | RBBIM_CHILD | RBBIM_CHILDSIZE | RBBIM_ID | RBBIM_STYLE | RBBIM_TEXT | 0;<br />
rbb.hwndChild = hWndBar;<br />
rbb.cxMinChild = rect.right - rect.left + 2;<br />
rbb.cyMinChild = rect.bottom - rect.top + 2;<br />
rbb.cx = 250;<br />
rbb.fStyle = RBBS_BREAK | RBBS_GRIPPERALWAYS;<br />
rbb.wID = 100085;<br />
rbb.lpText = TEXT("Toolbar");<br />
<br />
SendMessage(hToolbar, RB_SETPARENT, (WPARAM) (HWND) data->hParentHandle, (LPARAM)0);<br />
SendMessage(hToolbar, RB_INSERTBAND, (WPARAM)-1, (LPARAM)&rbb);<br />
SendMessage(hToolbar, RB_SHOWBAND, (WPARAM)(UINT)0, (LPARAM)(BOOL)TRUE);<br />
SetWindowPos(hToolbar, 0, 0, 0, 0, 0, SWP_NOMOVE | SWP_NOSIZE | SWP_NOZORDER | SWP_FRAMECHANGED);<br />
Now using that code, the bar appears to flash on the screen but doesn't stay there, what am I missing?
|
|
|
|
|
Hi,
I have one executable say program.exe which is linked with first.dll.
first.dll contains TestClass. Now if i replace the first.dll with the
dll of same name and same exported function of class TestClass, but
having different implimentation, then the program.exe is executed with
replaced first.dll.
Is there any way to know that the first.dll is replaced or not.
Regards
AbdulMunaf Chhatra
Software Engineer,
C-Sam Solutions Pvt. Ltd.
Phone : +91-265-2324018/19
Email : a.chhatra@popnet.co.in
|
|
|
|
|
Abdul Munaf Chhatra wrote:
Is there any way to know that the first.dll is replaced or not.
How about looking for an exported function or global variable that exists only in the original DLL?
"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown
|
|
|
|
|
Here is something I did a long ago.
Create a global string in the DLL like
char* pCpde = "My Secret Code";
Before loading the DLL from the EXE, open the DLL in binay mode from the EXE.
Search for your secret code.
If it exists, close the DLL file and load it.
I call it signing code. 
« Superman »
|
|
|
|
|
If you look at MS Word, with a new blank document.. and then choose View->Print Layout menu.. Now, you'll see the white page on the brown/gray background.. on a 100% view
I know that on normal circumstances in Doc/View architecture, the client area of CView entirely occupies the client area of the CCHildFrame..
Now with the Print-Layout View, Is it possible that the white page is drawn on the CView client area but it is NOT occupying the whole client area of CCHIldFrame so thats why we are seeing the background color of CChildFrame?
Or is the white page and the background color are all drawn together in the client area of CView?
Maverick
"watch the birdie!..."
|
|
|
|
|
I used Spy++ and using the window finder tool suggests that the 'white paper' that you see is a window that is not part of the background that the paper 'sits' on.
I'd be surprised to learn that the MFC document/view classes are in use in Word.
Chris Meech
I am Canadian. [heard in a local bar]
Remember that in Texas, Gun Control is hitting what you aim at. [Richard Stringer]
Nice sig! [Tim Deveaux on Matt Newman's sig with a quote from me]
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
