If necessary to test the GUI, I would rather use QTP or some kind of automation software to test my GUI. If you seperate your business/data logic from the presentation layer, you should be able to write unit test against the business/data layer. For GUI testing, you can take a look at Fitnesse[^]. I haven't tried it, but have seen some demos and heard it could really work for GUI applications.

I couldn't find the right information about white box testing for vc++ application so thought of taking help from CP.

What you are describing is a System Locale and it cannot be changed programatically. Programming Blog utf8-cpp

That's not as "easy" as it may seems, because sometimes even the hardware is involved... Let me go back in the early days when the fist IBM PC comes to light: there was the ASCII charset (7 bit) that IBM extended adding 128 more character, fitting 1 byte, and adding gliphs to satisfy many latin-derived alphabets (accented vovels, diacritics. etc, pus some "semigraphics"). This did't in fact satisfy any of the langags since all of the supported ones always had something missing. When windows comes along, a number of "localizations" had been done by basically replacing some useless glyphs (for a specific language) with more useful ones. This led to a number of different character maps (the so called "codepages") some of them still fitting the 256 glyphs space and some other going ahead of that number by using particular "escapes" (and hence requiring MBCS or DBCS support from the string manipulation functions and from the font definitions, as well from the graphics drivers. Unicode then standardized all that stuff by giving to any existing "glyph" a specific and unique ID. That makes -in theory- possible to depict all the glyphs in any font but - beware - that requires lot of space so not all font makers does that (or does that in different packaging). So: non-unicode applications to properly display characters must not only rely on the specification of the langID, but also require the corresponding codepage in the system, and since they cannot be contemporary used (a same code means a different glyph within a different codepage) the kernel (better: the user-side of it: user.dll and gdi.dll) must know which codepage has to use. But this requires dismounting a driver and mounting another one, affecting the system globally. That's makes your problem not solvable as an "application local problem". 2 bugs found. > recompile ... 65534 bugs found. :doh:

Hi, I create a table with ServerID and BagID and insert records like this: INSERT INTO tbServer VALUES (1,1), (2,1), (3,2), (4,3), (5,3), (6,4) Here is the query: SELECT DISTINCT BagID FROM tbServer GROUP BY BagID HAVING COUNT(BagID) = 1; I hope this will help you ...

Now I am able to get the correct value's The problem is with resource id's, I haven't defined same resource id in dll and in mfc application.

use function "afx_msg void OnTimeChange( )"

Thnx a-lot I will go with drawing text.

Thank you chandra for the link

Thank you

Super Hornet wrote: Is it possible to make a dialog based application behave as domodal dialog when it is invoked by shellexecute??? Yes. The fact that the dialog based application was invoked via ShellExecute() is irrelevant. "Love people and use things, not love things and use people." - Unknown "The brick walls are there for a reason...to stop the people who don't want it badly enough." - Randy Pausch

Super Hornet wrote: displaying them using "MoveFile" method Huh? MoveFile displays images? Super Hornet wrote: is it better to access the images directly from remote path and display instead of copying it to local machine? Why don't you try it? Mark Salsbery Microsoft MVP - Visual C++ :java:

Thanks Mihai.

Thank you Mark. The links are very useful.

Super Hornet wrote: Is the above code correct??? That all depends on what you were trying to accomplish. The use of cout appears to be the only thing C++ that I see. "Love people and use things, not love things and use people." - Unknown "The brick walls are there for a reason...to stop the people who don't want it badly enough." - Randy Pausch

Please post your replies

Mihai Moga wrote: Don Kackman's MFC classes for multiple monitors[^] 2) Jon Newman's Multiple Monitor Support[^] 3) Asim Goheer's Dual / Multiple Monitor Support in Win32[^] Don't these all require multiple monitors? OT: I found it funny that all three articles were conceived out of the need for a screensaver. "Love people and use things, not love things and use people." - Unknown "The brick walls are there for a reason...to stop the people who don't want it badly enough." - Randy Pausch

You are welcome!! Rating the reply will tell answerers how much it was helpful to questioners regards, suman -- "Programming is an art that fights back!"