UNICODE conversion
in [MFC]
|
From: Rami on 11 Mar 2008 09:14 I have a project in VC 6.0 which composed of a one main and several dll's. I want to convert all the projects gradually to UNICODE. However, since its a gradual process, through the debugging and integration, I may need to work with ASCII as well. Is it possible to do so? Is there some constrain on the interface in this case(or some other)? Regards Rami
From: Ajay Kalra on 11 Mar 2008 09:47 I would move all or nothing. It would be futile to keep some of it in UNICODE and the rest in ASCII. It will be painful and a wasted effort to keep both going at the same time. You should take the plunge and do the whole thing in one shot. Also, why are you converting to UNICODE? Is this a new requirement? --- Ajay "Rami" <rami(a)giron.com> wrote in message news:uFWvSn3gIHA.6032(a)TK2MSFTNGP03.phx.gbl... >I have a project in VC 6.0 which composed of a one main and several dll's. > I want to convert all the projects gradually to UNICODE. However, since > its > a gradual process, through the debugging and integration, I may need to > work > with ASCII as well. > Is it possible to do so? > Is there some constrain on the interface in this case(or some other)? > > Regards > Rami > >
From: Joseph M. Newcomer on 11 Mar 2008 10:25 This would be a Bad Idea. You have to convert all of it to use Unicode. Note that you can do part of the conversion gradually, but leave it as ANSI. For example, make sure all uses of 'char' are replaced by 'TCHAR', all uses of 'char *' by 'LPTSTR' and all uses of 'const char *' by LPCTSTR. Make sure all "..." and '.' literals have _T() around them. Replace all str... functions with their corresponding _tcs... functions. Watch out for confusions caused by thinking strlen or .GetLength() tell you the number of bytes, and fix them to do *sizeof(TCHAR) as part of the computation (and the inverse: byte counts disguised as character counts). Ditto for sizeof(). When you have all the code converted like this, then just build a new configuration in VS6 for Unicode Debug and one for Unicode Release, build in the Unicode Debug and start testing. This is by far the best approach. Don't try to convert one DLL to Unicode-only, and try to integrate it to an ANSI app; that way madness lies. joe On Tue, 11 Mar 2008 15:14:27 +0200, "Rami" <rami(a)giron.com> wrote: >I have a project in VC 6.0 which composed of a one main and several dll's. >I want to convert all the projects gradually to UNICODE. However, since its >a gradual process, through the debugging and integration, I may need to work >with ASCII as well. >Is it possible to do so? >Is there some constrain on the interface in this case(or some other)? > >Regards >Rami > Joseph M. Newcomer [MVP] email: newcomer(a)flounder.com Web: http://www.flounder.com MVP Tips: http://www.flounder.com/mvp_tips.htm
From: Rami on 11 Mar 2008 13:06 Thanks, Following your detailed answer I looked again in the projects settings and saw that all of them have "_MBCS" preprocessor definition. Should I delete it or just add the "UNICODE" definition? Thanks Rami "Joseph M. Newcomer" <newcomer(a)flounder.com> wrote in message news:m95dt3l47s9gbigdkqj6qvv2jrlb61tgdi(a)4ax.com... > This would be a Bad Idea. You have to convert all of it to use Unicode. > Note that you > can do part of the conversion gradually, but leave it as ANSI. For > example, make sure all > uses of 'char' are replaced by 'TCHAR', all uses of 'char *' by 'LPTSTR' > and all uses of > 'const char *' by LPCTSTR. Make sure all "..." and '.' literals have _T() > around them. > Replace all str... functions with their corresponding _tcs... functions. > Watch out for > confusions caused by thinking strlen or .GetLength() tell you the number > of bytes, and fix > them to do *sizeof(TCHAR) as part of the computation (and the inverse: > byte counts > disguised as character counts). Ditto for sizeof(). When you have all > the code converted > like this, then just build a new configuration in VS6 for Unicode Debug > and one for > Unicode Release, build in the Unicode Debug and start testing. This is by > far the best > approach. Don't try to convert one DLL to Unicode-only, and try to > integrate it to an > ANSI app; that way madness lies. > joe
From: Ajay Kalra on 11 Mar 2008 13:12
If you want UNICODE, you should not have MBCS or _MBCS at all in your settings. You should use UNICODE and _UNICODE instead. -- Ajay "Rami" <rami(a)goren.org.ir> wrote in message news:eNaHDp5gIHA.5164(a)TK2MSFTNGP03.phx.gbl... > Thanks, > Following your detailed answer I looked again in the projects settings and > saw that all of them have "_MBCS" preprocessor definition. > Should I delete it or just add the "UNICODE" definition? > Thanks > Rami > > "Joseph M. Newcomer" <newcomer(a)flounder.com> wrote in message > news:m95dt3l47s9gbigdkqj6qvv2jrlb61tgdi(a)4ax.com... >> This would be a Bad Idea. You have to convert all of it to use Unicode. >> Note that you >> can do part of the conversion gradually, but leave it as ANSI. For >> example, make sure all >> uses of 'char' are replaced by 'TCHAR', all uses of 'char *' by 'LPTSTR' >> and all uses of >> 'const char *' by LPCTSTR. Make sure all "..." and '.' literals have >> _T() around them. >> Replace all str... functions with their corresponding _tcs... functions. >> Watch out for >> confusions caused by thinking strlen or .GetLength() tell you the number >> of bytes, and fix >> them to do *sizeof(TCHAR) as part of the computation (and the inverse: >> byte counts >> disguised as character counts). Ditto for sizeof(). When you have all >> the code converted >> like this, then just build a new configuration in VS6 for Unicode Debug >> and one for >> Unicode Release, build in the Unicode Debug and start testing. This is >> by far the best >> approach. Don't try to convert one DLL to Unicode-only, and try to >> integrate it to an >> ANSI app; that way madness lies. >> joe > > |