Multilingual software..

[Egbert Zijlema]

Hello Edwin,

If I understand you correctly, you mean that my application does not use such languages like Russian, Greek or Chinese?
In that case you are right. It's all Latin, indeed.

------------------
Egbert Zijlema, journalist and programmer (zijlema at basicguru dot eu)
http://zijlema.basicguru.eu
*** Opinions expressed here are not necessarily untrue ***

[Michael Mattias]

>No, it is not enough.
>You need control created as unicode class too.

Really?
(PB/WIn 8.03)

' test_ddtw.bas
#COMPILE EXE
#DIM ALL
#INCLUDE "Win32API.INC"

%ID_LABEL      = 101&
%ID_TIMER      =  678&
%SUB_COUNTER   = 7

DECLARE FUNCTION SetWindowTextW LIB "User32.dll" ALIAS "SetWindowTextW" (BYVAL hWnd AS LONG, BYREF sz AS ANY) AS LONG

FUNCTION PBMAIN () AS LONG
    
    LOCAL hDLG AS LONG

    DIALOG NEW 0, "Test Unicode call on DDT dialog", 10,10, 200,100  TO hDLG
    ' add ANSI static control
    CONTROL ADD LABEL, hDlg, %ID_LABEL, "", 10,10,180,12
    
    DIALOG SHOW MODAL hDlg CALL DialogProc
    

END FUNCTION

CALLBACK FUNCTION DialogProc () AS LONG

 LOCAL iCount AS LONG, s AS STRING
 
   SELECT CASE AS LONG CBMSG
       CASE %WM_INITDIALOG
           CONTROL SET TEXT CBHNDL, %ID_LABEL, "Initial Text"
           SetTimer CBHNDL, %ID_TIMER, 1000&, %NULL
         
           
       CASE %WM_TIMER
            CONTROL GET USER CBHNDL, %ID_LABEL, %SUB_COUNTER TO iCount
            INCR iCount
            CONTROL SET USER CBHNDL, %ID_LABEL, %SUB_COUNTER, icount
            
            s = USING$ ("Counters:  #,  ", iCount)
            
            IF Icount MOD 2 THEN
            ' use ansi string
               S = S & "Using Control Set Text"
               CONTROL SET TEXT CBHNDL, %ID_LABEL, s
            ELSE
                ' use unicode string
               S = S & "Using SetWindowTextW"
               S = S & $NUL     ' required to make it nts
               S = UCODE$(s)
               SetWindowTextW GetDlgITem(CBHNDL, %ID_LABEL), BYVAL STRPTR(S)
            END IF
            
      CASE %WM_DESTROY
             KillTimer CBHNDL, %ID_TIMER
            
  END SELECT
  
END FUNCTION

MCM

[Edwin Knoppert]

That's not a real test imo since you are not really setting unicode characters, i mean using a 100% two byte based characterset.
Converting to ansi to uni might not be the same for this.
We really need to test this on such Windows imo.


------------------
http://www.hellobasic.com

[Aleksandr Dobrev]

Originally posted by Edwin Knoppert:
would an unicode SDK window viable for ansicode use?
(I mean vice versa?)

Yes, no problem with that.

Originally posted by Michael Mattias:
Really?

really

if you change your text to russian you will see surprize
becouse control is ANSI-class and OS converted it again to ANSI
with default regional setting.


------------------
-=Alex=-



[This message has been edited by Aleksandr Dobrev (edited January 07, 2007).]

[Edwin Knoppert]

Geeeeh

------------------
http://www.hellobasic.com

[Edwin Knoppert]

I was thinking, under NT it should be reasonably simple to hook calls like:

CreateDialogIndirectParamA() and such..

maybe this is an option?


------------------
http://www.hellobasic.com

[Michael Mattias]

>That's not a real test imo since you are not really setting unicode characters

Well here in flyover country we rubes kinda 'spect that UCODE$ creates, like, unicode, you know?

> if you change your text to russian

Wait just a minute.. Russian.. Egbert and Edwin are in Holland, Schullian's in Greece...

Damn foreign conspiracy, that's what it is. Let me call Homeland Security, we'll put an end to this.

[Edwin Knoppert]

>Egbert and Edwin are in Holland

Holland?? nah.. we are in the Netherlands
(So i've heard)

------------------
http://www.hellobasic.com

[Tyrone W. Lee]

Geez...

I'm lost and there are too many cross conversations for me to understand
anything going on here.. I'll start with finding out what UNICODE
actually is.. I can't see why you need unicode to change English to Dutch.

Why do you need UNICODE to change:

"Help" to "Hulp" to "Hilfe"

Browsers do it.. And its been done on paper and computer long before
Windows and all this crap..

Thanks anyway..

Tye

------------------
Explorations v7.96 RPG Development System http://www.explore-rpg.com

[This message has been edited by Tyrone W. Lee (edited January 07, 2007).]

[Aleksandr Dobrev]

Originally posted by Michael Mattias:
Wait just a minute.. Russian.. Egbert and Edwin are in Holland, Schullian's in Greece...
Damn foreign conspiracy, that's what it is. Let me call Homeland Security, we'll put an end to this.

Michael !!!!

The best phrase, thank you !!! :-|






------------------
-=Alex=-

[Edwin Knoppert]

>I can't see why you need unicode to change English to Dutch.

Indeed but at some point Windows wanted to access the azian market and supporting their local language as well (instead of Englisch only).
The ascii set is very limited and therefore, since Windows 95, they implemented unicode support.
This means instead of a single byte they extended it to two bytes to so you'll get 65k combinations instead of 256.
(Chinese have more than 1000 symbols??)

To remain backwards compatible they kept support for single byte as well.

>Browsers do it.. And its been done on paper and computer long before Windows and all this crap.
Hmm, not so.. afaik the original html language was 127 chars aka 7bit, not even 8bit!
Nowadays there is utf8 and utf16 and so on.
BROWSERS are prepared to handle the unicode.
But then there is the language issue, the unicode value must 'fit' the users language.
Imo that's the reason i can't correctly read azian websites.
I'm not so clever with all of this but maybe someone can explain this better.



------------------
http://www.hellobasic.com

[Aleksandr Dobrev]

Originally posted by Tyrone W. Lee:
I can't see why you need unicode to change English to Dutch.
Why do you need UNICODE to change:
"Help" to "Hulp" to "Hilfe"

No. Lets wait for german people, they have umlaute

If seriously, then european languages has a few chars
with different view, but others has a lot.
IMHO ignoring fast growing countries it is loosing market.
Few years ago salary in Russia was ~$100-$200/month.
Right now to be able afford live in a Moscow people need
at least $1000-$1200/month. Thats mean, if a few years ago
russians need make a cracks to be able afford some programs,
but right now they will prefer to pay for programs than spend
sleepless nights time for cracking some programs.

So its up to developer ignore that countries or not.

Browsers do it..

Browsers is unicode application...

------------------
-=Alex=-

[Aleksandr Dobrev]

Originally posted by Edwin Knoppert:
Imo that's the reason i can't correctly read azian websites.

To be able display correctly foreign languages you just need
go to Start->Control Panel->Regional and Language options
then switch to Advenced tab and select as many languages as
you want support.

I'm not commercial specialist, but IMHO ignoring
countries as Japan,Korean,China,Turkish,Finland,Sweden,France, Russia....

Ok, thats up to ... software maker, M$ made their choice,
and then less compititors on that markets, than better for
that who understand it



------------------
-=Alex=-

[Edwin Knoppert]

Agreed but it's also not so trivial.
You'll need translators and when things are added these needs to be translated as well.
Iow, it's not simple translate and sell issue.

------------------
http://www.hellobasic.com

[Don Schullian]

Tyrone (if you're still with us),

Assuming that the program you write is being used in the country
that uses the language you want to display in then you don't need
unicode. What looks like gobbledegook on your computer will read
correctly on theirs. That's the good news.

The bad news is that you may not be able to see the text correctly
on your computer but.... even if you could see it, could you read it
and be able to correct it? The word 'conundrum' comes to mind here.

Michael: Yes, you've caught us! Call the cops! Ya' just have to
keep an eye on us 59 year old, Caucasian males wearing sneakers.



------------------
--
C'ya
Don
don at DASoftVSS dot com
http://www.DASoftVSS.com

[Lothar Pink]

Does anyone still wonder why I suggest all PB programmers learn to write applications 'SDK-style' at some point?

It is still a difference if you work with unicode APIs or if you have a compiler like MSVC++ which provides you with
unicode variables. It is quite boring to use unicode with ASCIIZ strings.

I also had to translate a PB application to unicode, it had to be japanese. For the strings I have created a unicode CSV file
which contains the strings (I use that way because the CSV file is exported by Excel which can be easily edited by
some non-programmer which makes the translation).

For the unicode support, I have created a layer which dynamically replaces the calls to ansi APIs to unicode APIs if
unicode is available on the system (this isn't the case on most Win9x systems). This means that I can use unicode with DDT.

This unicode tool is not complete or bug-free, but my application doesn't have so many dialogs so with some manual corrections
it works quite well.

Lothar

------------------

[Lothar Pink]

i have posted my unicode ddt tool at:
http://www.powerbasic.com/support/pb...ad.php?t=25069


------------------