I'm starting to develop a Windows service. I want to use some classes from my own, that has little dependencies to some MFC classes like CString, CSocket, CArchive, CMemFile and CObject. MSDN says you need to be very careful about which pieces of MFC you use in the Windows service, but don't specifies it and don't describes the problems that can occur.
My questions are:
what pieces of MFC can be used?
what problems can I expect, by using MFC?
which parts of Windows service are critical for MFC use?
is it advisable to use ATL instead of MFC for Windows service?
I'm not sure what they mean in teh MSDN article. As long as you don't use any of the GUI functionality you'll be fine - but that's a general design issue when developing services.
That being said, ATL has functionality specifically designed for building services IIRC so you may be better off using that.
To answer your questions (to the best of my knowledge):
1) the ones you specify are no problem.
2) I guess they mean synchronization issues with UI components. As long as you don't use any CWnd-derived classes you'll be fine.
3) don't understand the question.
4) See before, plus ATL is more lightweight so you'll have to distribute less, and provides build-in functionality that'll make it less of a pain to develop the service. See e.g. CAtlServiceModuleT. You'll still be able to mostly use your own classes, as CString is shared between MFC and ATL nowadays and ATL has classes for socket programming and memory file mapping itself. It doesn't have an equivalent for CArchive, and I'm not sure what functionality you use in CObject so I can't say whether there's an equivalent in ATL. So to conclude, I'd say 'yes' to this question.
(I know this answer is a bit late and this question was already answered but MFC in services is a sore spot for me...)
CSockets, far as I recall, require a Window. It makes an invisible one in the background. I found out this the hard way when I tried include some pre-exisiting MFC code into a windows service. Maybe this was only required if you accepted socket connection - I can't recall? But it did not work! (How exactly I wasted so much time doing this w/o realizing this limitation is a long story)
CObject? If you need the runtime class id stuff use RTTI (dynamic_cast, etc...)
CString, I like CString, I know it's shared with ATL now, not sure if you pull it in w/o MFC or ATL included... You could use std::string. Also, I recall someone created a derived std::string that provided the same methods as CString.
(EDIT: found the code - man!! that's a blast from the past...)
CArchive, CMemFile: do you really need these?
Anyway, as Roel said, ATL may be more helpful. I wouldn't use MFC in a server-side application (ever!) ATL? Maybe. If I needed COM, defiantly. No COM but for CAtlServiceModuleT, etc... maybe....
And another bad thing about MFC in services that I have just experienced while trying to turn a regular MFC-ATL app into a service: The use of AfxConnectionAdvise() is actually useless without a Window procedure. The threads in my service are just regular non message-pumping threads. I believe this is why I never get events fired from another COM server I have developed. That other COM server hangs on Fire_xxxEvent(), causing a big mess in the whole system.
Related
Was wondering if this: https://code.visualstudio.com/docs/editor/codebasics could be implemented with WinApi or by DLL calls/injection globally in every application?
Which api call could be relevant to get me started?
This cannot be done. There are too many problems that need to be solved for which there is no general solution.
The standard carets have a hard limit of one caret per queue. With that in mind you would now have to solve not one, but two problems: Getting a custom caret implementation, and fighting the system-provided one.
That might still sound doable, but even just getting your custom-rendered carets injected into foreign windows isn't possible. There is no infrastructure in the system that allows you to safely tap into the rendering of arbitrary (or even standard) controls you do not own.
Now even if you got a solution to all of the above, how would you communicate those multiple selection marks back to client code? EM_GETSEL is strictly limited to one selection mark at most.
So far, this was mostly just about standard controls. Things won't get any easier with custom control implementations. WPF, to my knowledge, uses a pretty much closed down control library, that doesn't even provide the customization points of Windows' common controls. Same goes for UI toolkits like Qt. While open source, Qt doesn't allow for any external customization.
I'm sure there are more problems that don't have a general solution. While not an exhaustive list, the above problems prevent implementation of multi-selection in arbitrary UIs outside your control.
I am developing socket reading on an ActiveX EXE (i.e on a seperate thread).
How many sockets i can safely read independently?
I am working on windows XP OS.
I think this might be an operating system limit - I think I remember running up against a limit of 80 sockets on the XP machine I was using several years ago.
I would recommend that you abandon your effort and go with a commercial solution. I remember going down this path back in the 90s and running into a brick wall with ActiveX EXEs as far as threading goes. The thing is that ActiveX EXEs are apartment threaded, not free-threaded, so you don't get completely independent threads.
And doing server side threads properly is hard enough in modern languages, let alone ones that weren't designed for this purpose.
I ended up purchasing Server Sockets from Dart. Easily the best investment for that project. The performance is truly great - you are only limited by the system resources.
MSWINSCK.OCX is a very old way of doing things; it came with Visual Basic 6.0 and i remember using it way back when. i'm not sure the licensing on it... apparently it registers fine under 32-bit win7, but not 64-bit; here's a link to how to get it to register on 64-bit systems: http://angrybyte.com/windows-hacks/mswinsck-ocx-for-64-bit-windows-7-vista/
if you have an MSDN subscription or similar that gives you the ability to download the developer tools (bizSpark, etc. will do it too) then i believe that will also give you a license to redistribute the .ocx.
(btw, i don't actually remember the interface, but i seem to remember it being at least slightly more intuitive than the berkeley socket() interfaces.)
however, personal recommendation given your requirements: learn the APIs, there are lots of examples out there, and just write yourself a class that encapsulates them in a similar way as, say, the .NET Socket class... the APIs aren't that hard and i'm sure there's lots of help to be had here as well, and that's probably better than relying on something that's out-of-date like the control...
There are many applications for Windows these days that don't use native windows controls, don't have standard window frames and generally look different. What are some recommended techniques for creating such interfaces?
There are good reasons not to. Like that you will most likely not do a better job than Windows does. (Maybe it will look better (in your opinion), but will it behave?). Or that it's not what most users expect. Or that it will look like s**** on Windows 2011.
That said, it's not hard. You simply handle the WM_NC* events like WM_NCPAINT or WM_NCHITTEST. NC stands for Non Client (window area). And of course, there is a trick on Vista/Win7 (you have to announce it to the DWM).
From an implementation aspect, you could employ WPF (Windows Presentation Foundation) assuming you code for .NET :) It has pretty bunch of skinnable controls, that may look like native and may not.
From a design aspect, if your interface isn't going to follow documented standards (like the Windows UI guidelines), it has to be intuitive. I think the new generation of Windows applications will go through a growing phase in a manner similar to the early days of the Web. After a time, some standards or common themes will evolve.
Can you give us some sample applications? Some apps that don't use native windows controls use cross-platform GUI libraries, like Qt for C++ or Tkinker. These maintain the same look across different platforms.
I wouldn't really recommend making your user interface different deliberately. You don't stand to gain much. Your controls are almost always going to be buggier than native controls, and you are requiring the user to learn something new. Now, if you're controls add a large enough value to be worth the users' time it can be okay. But making them get used to different looking buttons is rarely worth it.
I`m not sure if this answer your question.
You can use third party skinning controls like from Infragistics, or SkinSoft for example.
But like Bubba said I`d recommend going for WPF.
Model-View-Controller! It's as valuable here as in web apps or anywhere else. Be sure to keep the part of your program that generates the custom UI separate from the part of your program that flashes the BIOS.
I know this question is 10 years old but none of the answers mention using an option in visual studio, dont know if it existed at the time.
Theres an option to remove the border of the window in visual studio (called borderStyle). Thats the easiest way to do it, using C#. After removing the border, all you have to do is create a new interface. If you're looking to do it in C++, i think you need to use DWM. I will let an example i found here.
https://github.com/melak47/BorderlessWindow
Another example (maybe without DWM? didnt test):
https://social.msdn.microsoft.com/Forums/vstudio/en-US/b98c4c06-9581-44d3-8e5a-4adb2316e653/win32-about-styles-how-can-i-do-a-borderless-window?forum=vclanguage
There is a lot of people disencouraging to do it in this thread but there's no reason to not do it, if you know what you're doing your application can look great.
It takes at least 7 assemblies and restricting my AddIn's data model to data types that remoting can deal with before the appdomain isolation features begin to work. It is so complex! The System.AddIn teams blog implies to me they were trying to re-create a mental model of COM, a model I never understood very well in the first place and am not sold on the benefits. (If COM is so good why's it dead?-rhetorical question.) If I don't need to mirror or interop with legacy COM (like VSTO does using System.AddIn), is it possible to just create some classes that load load in a new AppDomain?
I can write the discovery code my self, I've done it before and a naive implementation is pretty fast because I'm not like iterating over the assemblies in the GAC!
So my specific question is, can I get the AppDomain isolation that AddIns provide with a few code Remoting snippets, and what would those be?
I'm not entirely sure that that any answer to your question meets the terms of the site - there is no solution.
Yes, remoting is easier as it is done for you. However, it is highly controlled and as you identified, requires a little work to plumb it all together. The cache file spewed out by the discovery process is hardly welcome either.
System.AddIn excels at isolation, which is actually a bit of an arse to put together from scratch in a robust, flexible way. It supports cross process hosting and fairly simple passage of user WPF elements from one domain to another.
One thing to remember however is that MAF's target audience is not those who are trying to connect two applications together. It is targeting developers wanting pluggable yet secure systems (cross process hosting protects the root application from unhandled exceptions, appdomains allow for executing potentially foreign code with defined security). From most communication, direct yourself straight towards System.Runtime.Remoting or WCF.
If you want to continue with System.AddIn, consider the pipeline builder plugin for visual studio!
In conclusion - you can get System.AddIn isolation using Remoting but to get a decent system you will require more than a few snippets. I am trying to replicate it myself and am tripping up all over remote interface component - something System.AddIn does without a hitch.
After messing around with System.Add for quite a while, I'm convinced that it was added as a one-off special purpose solution for Microsoft use. I'm surprised it got elevated to a core part of the .NET framework. It doesn't seem to have the refinement and polish needed for a general .NET framework component.
I'd like to find an alternative way to create .NET managed add-ins that doesn't require so much effort.
I'm about to take on a project that requires a Firefox addon to issue call to COM dll's installed on the client Windows machine and I'm having a hard time estimating the complexity of this undertaking.
I have quite a bit of experience with COM, so I'm not frightened by it. I have less experience with Firefox addons, but I don't think that's where my problems are going to be.
Has anybody done anything like that?
Does Firefox allow its addons to communicate freely with the outside world?
Is there a plugin or sample code somewhere that does something like this (Google turned up little useful results)?
Update: Naturally, I would prefer a solution that doesn't require building an extension in C++, if that's at all possible.
Create an XPCOM in C++ to talk to your COM objects as usual. The XPCOM extensions can then be made available to Javascript to do the rest of the extension (GUI mainly, I guess). However, that's about all I understand about it :-)
You might be interested by IE Tab extension, which is open source. I haven't looked but I guess it does Com access to use IE's display.
If you do this you'll almost certainly need to use XPCOM. Take a look at Shanti Rao's JSDB, which supports COM/ActiveX within Javascript. The ActiveX stuff is in a file called wrap_com.cpp. It supports most IDispatch interfaces; the Invoke method of IDispatch is the magic that makes this work.
How complex this is depends on how "easy" you want this to be from Javascript. If you implemented an XPCOM thing called IDispatchObject where you had to pass in the names of methods & an array of arguments, then it's probably not too hard. If you want to create a general method for doing dynamic bindings to COM objects & referring to them directly, then that's tougher... not sure if the techniques used in JSDB will carry over to XPCOM.