Problem in Arabic writing

[NMMA]

I use Qt 4.6.2 in fedora "linux", I create array with static values contains Arabic strings.

but when I tried to retrieve any value from my array the string returns with value

= NULL.

I use "char*" to save my values.

Is there problem in the encoding?

can any one help me?

[danhicksbyron]

Have you tried regular Roman/ASCII characters in the same scenario?

[rahulvala]

I use Qt 4.6.2 in fedora "linux", I create array with static values contains Arabic strings.

but when I tried to retrieve any value from my array the string returns with value

= NULL.

I use "char*" to save my values.

Is there problem in the encoding?

can any one help me?

There is only a small set of characters that you can use in a GET request. Other than A-Z, a-z, 0-9 and a few other characters, anything else needs to be URL encoded ("%xx", where "xx" is the hex code for the byte).

Since a "%xx" code can represent only one byte, it cannot be used to encode Unicode characters. You must convert the Unicode characters into bytes first.

The best way to encode Unicode as a sequence of bytes is to encode as UTF-8.

UTF-8 will encode Arabic characters as two bytes each.



For more information visit below disscusion :
http://discussion.forum.nokia.com/fo...d.php?t=168291
http://discussion.forum.nokia.com/fo...ad.php?t=59463
http://discussion.forum.nokia.com/fo...d.php?t=165996


best regards,

[divanov]

I think your problem might be related to an editor you use

#include <QtGui>

int main(int argc, char *argv[])
{
    QApplication app(argc, argv);
    const char *str = "رسالة اختبار";
    QString string = QString::fromUtf8 (str);
    qDebug() << QString("Unicode length %1 ASCII length %2").arg(string.length()).arg(strlen(str));
    for (int i = 0; str[i]; i++)
        qDebug() << QString("[%1]").arg(str[i]);
    for (int i = 0; i < string.length(); i++)
        qDebug() << QString("[%1]").arg(string.at(i));
}

[NMMA]

Thanx all for ur reply,

I want to know how to convert my text from utf-8 to ANSI manually without using Qstring?

[divanov]

Thanx all for ur reply,

I want to know how to convert my text from utf-8 to ANSI manually without using Qstring?

There is no way to convert UTF-8 to Windows-1252 as the former is a superset of the latter.

[NMMA]

I cannot do function that do mapping between UTF-8 representation and ANSI?

[divanov]

I cannot do function that do mapping between UTF-8 representation and ANSI?

Well, how do you define mapping? If it's about conversion then you can check how iconv deal with the similar task
http://www.gnu.org/software/libiconv/

[NMMA]

Thanx for ur quick reply

but I have Question..

If I use UTF-8...I need to store my output in wchar? or I can continue to use my char*?

[divanov]

If I use UTF-8...I need to store my output in wchar? or I can continue to use my char*?

To answer the Question I will need to know how are you planning to use the output.

[NMMA]

I want to store arabic character in "char*" and use "switch" to make

some comparison like that:

Code :

char* one="صفر"
int len=strlen(one);
int i = 0;
for(i = 0;i<len;i++)
{
//
//
switch(one[i])
{
case 'َ':
one[i]= 'a';
break;
case 'ِ':
one[i]='i';
break;
case 'ُ':
one[i]='u';
break;
case 'ّ':
one[i]='+';
break;
case 'ْ':
one[i]='o';
break;
case 'ً':
one[i]='&';
break;
case 'ٌ':
one[i]='$';
break;
case 'ٍ':
one[i]='%';
break;
case 'ء':
one[i]='?';
break;
case 'ا':
one[i]='A';
break;
case 'إ':
one[i]='I';
break;
case 'ؤ':
one[i]='U';
break;
case 'ئ':
one[i]='Y';
break;
case 'أ':
one[i]='P';
break;
case 'ب':
one[i]='b';
break;
case 'ت':
one[i]='t';
break;
case 'ة':
one[i]='e';
break;
case 'ث':
one[i]='c';
break;
case 'ج':
one[i]='j';
break;
case 'ح':
one[i]='H';
break;
case 'خ':
one[i]='K';
break;
case 'د':
one[i]='d';
break;
case 'ذ':
one[i]='>';
break;
case 'ر':
one[i]='r';
break;
case 'ز':
one[i]='z';
break;
case 'س':
one[i]='C';
break;
case 'ش':
one[i]='s';
break;
case 'ص':
one[i]='S';
break;
case 'ض':
one[i]='D';
break;
case 'ط':
one[i]='T';
break;
case 'ظ':
one[i]='Z';
break;
case 'ع':
one[i]='E';
break;
case 'غ':
one[i]='g';
break;
case 'ف':
one[i]='f';
break;
case 'ق':
one[i]='q';
break;
case 'ك':
one[i]='k';
break;
case 'ل':
one[i]='L';
break;
case 'م':
one[i]='m';
break;
case 'ن':
one[i]='n';
break;
case 'ه':
one[i]='h';
break;
case 'و':
one[i]='w';
break;
case 'ي':
one[i]='y';
break;
case 'آ':
one[i]='@';
break;
case 'ى':
one[i]='V';
break;
default:
break;
}

}
First, when I do that.size of len return with value=6 although I set str with 3 letters
I need to use wchar? how I can deal with this situation ?

[divanov]

I want to store arabic character in "char*" and use "switch" to make
some comparison like that:

First, when I do that.size of len return with value=6 although I set str with 3 letters
I need to use wchar? how I can deal with this situation ?

Arabic letters as being non-Latin take 2 bytes for storing in UTF-8. However, some other alphabets can take even more. In order to produce portable code, you should switch to using QString instead of char* and QChar instead of char. You won't be able to use switch operator for comparing Unicode characters as they are of non elementary type, but rather a strings.
Here is a small example:
http://doc.qt.nokia.com/4.6/qstring.html#operator-5b-5d

[NMMA]

Arabic letters as being non-Latin take 2 bytes for storing in UTF-8. However, some other alphabets can take even more. In order to produce portable code, you should switch to using QString instead of char* and QChar instead of char. You won't be able to use switch operator for comparing Unicode characters as they are of non elementary type, but rather a strings.
Here is a small example:
http://doc.qt.nokia.com/4.6/qstring.html#operator-5b-5d

Thanks too much for your reply
my problem finally solved