|
On the contrary. MFC Doc/View is way too file centric and way to restrictive. I like the Smalltalkish MVC model a LOT better, where the model isn't file centric.
--
Please state the nature of your medical emergency.
|
|
|
|
|
I am interested in the MVC model, do you have any good links, or should I just google to find some?
Also, Microsoft strongly believes in the Document centeric model. If you look at all of their current software, Office being the main model, they are all document centric. As a general rule they do not do MDI interfaces anymore (development environments excluded).
If you have two documents, you will have two top level windows open on the task bar to allow you to switch between the two documents a lot easier.
So possibly this will explain the strong file centric development model in MFC.
(thats actually a bunch of BS because WTL does this the proper way with the Multi-threaded SDI model  )
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life!
|
|
|
|
|
Paul Watt (kilowatt) wrote:
I am interested in the MVC model, do you have any good links, or should I just google to find some?
I don't have any links on the top of my mind. I was acquainted with the MVC paradigm in a OO-course using Smalltalk (I chose not to say model to avoid confusion - the M in MVC stands for Model  ). The entire GUI was built using this paradigm. Anyhow, MVC is a fairly well-known OO-concept in GUI modelling. I think a Google search would give you more than you can read in a life time.
MVC stands for Model View Controller. The View(s) observe(s) a Model and updates itself for each change in the Model. The Controller(s) control the Model and the View. The Controller(s) also observe(s) the Model and the View and may update accordingly.
Example scenario: Model = geographical map data. View = graphical representation of map data. Controllers = scrollbars for panning and buttons for zooming in and out. The scrollbar Controllers observes the Model and the View - thus it is able to determine how large the thumbs should be. When the thumbs are moved, the scrollbar Controller notifies the View that it should depict another portion of the map. It's all very simple really, it's just a division of responsibility to make it easier and more flexible.
If you haven't noticed, Microsofts Document is equivalent to the Model, and the View is equivalent to View+Controller. It's easy to see that the Document/View isn't half has flexible as "pure" MVC. I don't know why Microsoft settled for a less powerful paradigm - nobody bothered to question it perhaps?
Paul Watt (kilowatt) wrote:
Also, Microsoft strongly believes in the Document centeric model. If you look at all of their current software, Office being the main model, they are all document centric. As a general rule they do not do MDI interfaces anymore (development environments excluded).
Yes, but Office is per se file-centric. Everything you do in MS Office results in a file. What if the "document" is reading data from the network never intended to be docified? In situations like that, the doc-template stuff is really mind-boggling. And all those weird frame/view/document "protocols"? Not very intuitive at all. I get the shivers everytime I do it.. 
Paul Watt (kilowatt) wrote:
thats actually a bunch of BS because WTL does this the proper way with the Multi-threaded SDI model
WTL does very little - but what it does is proper and well done. MFC does a whole lot, but misses the point 90% of the time. But that's just my not too objective oppinion.
--
Please state the nature of your medical emergency.
|
|
|
|
|
Heath,
Here is another good link, especially with documentation if you are going to be doing anything on WTL. The WTL Documentation Project[^] here at Code Project.
Nick Parker
The greatest lesson in life is to know that even fools are right sometimes. - Winston Churchill
|
|
|
|
|
Can somebody show me how should I work with an CComPtr pointer. Are there any web links wich I can follow.
|
|
|
|
|
CComPtr is a SmartPointer wrapper class specifically designed for COM objects. So when your smart pointer goes out of scope, it will automatically call Release on your object. It is very simple to use.
First declare a CComPtr to wrap one of your objects:
<br />
CComPtr<IDataObject> spDataObj;<br />
Now you can use spDataObj anywhere that you would normally use a pointer to the interface. You do not need to worry about casting or dereferencing or anything like that it is completely transparent.
<br />
IUnknown *pUnk;<br />
...<br />
pUnk->QueryInterface(IID_IDataObject, &spDataObject);
spDataObj->GetData(...);
Some other cool things to know are the Attach and Detach members. This will allow you to attach an existing interface pointer to the CComPtr without addrefing it internally, and the Detach method will allow you to remove the internal interface pointer without releasing it.
Other than that it should be pretty straight forward, and you may also want to checkout CComQIPtr. That smart pointer object will allow you to query for the interface on assignment, so it will save you one step when you are setting up your pointers.
Good Luck
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life!
|
|
|
|
|
Also, check out CComPtr::CoCreateInstance (or was it CreateInstance only?)
--
Please state the nature of your medical emergency.
|
|
|
|
|
it is CoCreateInstance. CreateInstance is with _com_ptr_t.
Michel
It is a lovely language, but it takes a very long time to say anything in it, because we do not say anything in it, unless it is worth taking a very long time to say, and to listen to.
- TreeBeard
|
|
|
|
|
|
That does look cool.
Have you used it?
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life!
|
|
|
|
|
Not yet. I browsed through the documentation and found that it's excellent in theory.
What's really excellent is that you don't need a separate tool like yacc or bison - you just need C++/STL! And the quality seems very high. It's being submitted to the boost library/committers so it has to be high quality.
They're also working on a strict LL(1) parser (more efficient in terms of speed and memory) for every day grammars. What's lovely with LL(k) is that it can take on complex grammars like that of C++.
Also, if I understood the docs correctly, it also has support for symbol management with complete scoping! They even claim that they'll implement the different scoping rules found in C++ - including König lookup.
If I need to write a parser/compiler again, I'll sure try this library. It looks very promising.
--
Please state the nature of your medical emergency.
|
|
|
|
|
from the page:
... The Spirit framework enables a target grammar to be written exclusively in C++. Inline EBNF grammar specifications can mix freely with other C++ code and, thanks to the generative power of C++ templates, are immediately executable. In retrospect, conventional compiler-compilers or parser-generators have to perform an additional translation step from the source EBNF code to C or C++ code. ...
WOW !!!!! Writing the grammars in C++ !!! Inline EBNF grammar specifications mixed freely with other C++ code !!!!!!!!!
Using C++,STL and template metaprogramming !!!!!!!! 
Now , when I have the time I'll test it, but it appears that it ROCKS !!!
Thanks a lot for the link !!! Added instantly to my Favourites
p.s: I didn't have compilers in college , but I studied myself a bit
Cheers,Joao Vaz
And if your dream is to care for your family, to put food on the table, to provide them with an education and a good home, then maybe suffering through an endless, pointless, boring job will seem to have purpose. And you will realize how even a rock can change the world, simply by remaining obstinately stationary.-Shog9
|
|
|
|
|
Great ain't it?
Joao Vaz wrote:
p.s: I didn't have compilers in college , but I studied myself a bit
Compiler construction was the single best course I ever took at the university. I learnt a lot of good things in that course - everything from lexical scanning, parsing, symbol management, code generation/virtual machine-basics and some optimization. If your local university ever offers such a course - don't blink! Apply for it right away, you won't regret it.
Our lab assignment was to write a compiler for a simplified version of Pascal. The compiler then generated output in a three address code. This 3ac should then be executed in a virtual machine. Best lab assignment I ever had. I miss school sometimes.
--
Please state the nature of your medical emergency.
|
|
|
|
|
Jörgen Sigvardsson wrote:
Compiler construction was the single best course I ever took at the university. I learnt a lot of good things in that course - everything from lexical scanning, parsing, symbol management, code generation/virtual machine-basics and some optimization.
Cool course . 
Jörgen Sigvardsson wrote:
Our lab assignment was to write a compiler for a simplified version of Pascal. The compiler then generated output in a three address code. This 3ac should then be executed in a virtual machine. Best lab assignment I ever had.
Even cooler.
Jörgen Sigvardsson wrote:
I miss school sometimes.
Me too, but at the time I hated it 
Now I always thinking in taking a masters degree and do research ...
Cheers,Joao Vaz
And if your dream is to care for your family, to put food on the table, to provide them with an education and a good home, then maybe suffering through an endless, pointless, boring job will seem to have purpose. And you will realize how even a rock can change the world, simply by remaining obstinately stationary.-Shog9
|
|
|
|
|
Jörgen Sigvardsson wrote:
Compiler construction was the single best course I ever took at the university. I learnt a lot of good things in that course - everything from lexical scanning, parsing, symbol management, code generation/virtual machine-basics and some optimization. If your local university ever offers such a course - don't blink! Apply for it right away, you won't regret it.
I had two courses like that last year. Those were the most difficult courses in college and not all students take it.
Jörgen Sigvardsson wrote:
Our lab assignment was to write a compiler for a simplified version of Pascal. The compiler then generated output in a three address code. This 3ac should then be executed in a virtual machine. Best lab assignment I ever had. I miss school sometimes
Similar assignement here also!
Best regards,
Alexandru Savescu
|
|
|
|
|
Alexpro wrote:
Those were the most difficult courses in college and not all students take it.
Challenges inspire!
The single most difficult course I took was "Distributed Systems". And now we're talking truly distributed systems - not simple client/server-systems. It really starts freaking out when you take causality into account - it feels all of a sudden like a bad mix between Star Trek and theory of relativity. All of a sudden you're in this time-quantified universe where you cannot always be sure if one event occured before another. Not having a common clock make things very complicated!
--
Please state the nature of your medical emergency.
|
|
|
|
|
Jörgen Sigvardsson wrote:
The single most difficult course I took was "Distributed Systems".
I am studying this course this semester!
The first lab describes the Synchronizing Resources Language (the home page can be found here).
Best regards,
Alexandru Savescu
|
|
|
|
|
I'm doing a COM/ATL dll and I want to add XP themes to it. I did the manifest thigny but it does not seem to work...
any hints?
Thanks guys!
---------------
Concentrating on Ideas
http://www.edovia.com
|
|
|
|
|
From what I've read, .dll must have the RT_MANIFEST resource defined as 2 for libraries, instead of 1 for applications. See if that works.
"Well, I wouldn't say I've been missing it, Bob." - Peter Gibbons
|
|
|
|
|
Didn't work for me... do you have some documentation about how to do it?
Thanks!
---------------
Concentrating on Ideas
http://www.edovia.com
|
|
|
|
|
All the documentation is available in MSDN[^]. That would be a good place to look.
"Well, I wouldn't say I've been missing it, Bob." - Peter Gibbons
|
|
|
|
|
Dudes,
I had a small MFC app based on propertysheet and pages that I am porting to WTL. All is well except I never get the WM_INITDIALOG message for the propertysheet. Any clue?
|
|
|
|
|
The property sheet template does not process the WM_INITDIALOG message. You'll need to create your own Init routine and call it after the sheet is created.
|
|
|
|
|
Can somebody help me eliminate the following error. I get this error when I compile for Release and not for debug.
--------------------Configuration: esContaX - Win32 Release--------------------
Linking...
LIBCMT.lib(crt0.obj) : error LNK2001: unresolved external symbol _main
Release/esContaX.exe : fatal error LNK1120: 1 unresolved externals
Error executing link.exe.
esContaX.exe - 2 error(s), 0 warning(s)
I Know this error is related to that the function _main is somehow unrichable. But why could this happen only in release mode.
|
|
|
|
|
Try changing the C/C++ run-time you link with : VC++ / C++ / CodeGeneration / (single threaded(libc) | multithreadedstatic(libcmd) | multithreadeddll(msvcrt)).
She's so dirty, she threw a boomerang and it wouldn't even come back.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
