Hi you all,
    I've been using Amarok on my Ubuntu for sometime now and I realize that some of my Thai songs (I am Thai by the way) are not put into the right header in the collection view. Please see the attached screen short.
    To be more specific, in the collection view, the artist will be display alphabetically grouped by its first character (without 'the'). This is correct for all of the Latin artist names. However, for my Thai artist names, all of them are grouped into one. The group header shows the first character of the last (in Thai alphabetical order). Basically, it's wrong.
    I tried to look into the code and have put some debuging code in it, right after the CollectionItem is created (the DividerItem (the item of the group header) is created every before that).

file:collectionbrowser.cpp - function:CollectionView::renderTreeModeView - line# ~ 4000 (the line# is shift after I mess with it :p )

Code: Select all


            CollectionItem* item = new CollectionItem( this, m_cat1, unknown );
            item->setExpandable( true );
            item->setDragEnabled( true );
            item->setDropEnabled( false );
            item->setText( 0, *it );
            item->setPixmap( 0, pixmap );

    // NawaMan Added ----------------------------------------------------------------------
    printf("childCount() = %d, dividerCount = %d \n",childCount(), dividerCount );

if(dividerCount < 10) {
    for(int i = 0; i < childCount(); i++) {
        QListViewItem *item = itemAtIndex(i);
        QString a = item->text(0);
        printf("Item Text #%d |%d|: ",i,a.length());
        for(int i = 0; i < a.length(); i++) printf("%c",a.at(i).unicode());
        printf("(");
        for(int i = 0; i < a.length(); i++) printf("%x-",a.at(i).unicode());
        printf(")\n");
    }
} else printf("> 10 now\n");
    // End NawaMan Added ------------------------------------------------------------------------------------------------

            //The structure of the return is different if Year - Album is format
            if ( VisYearAlbum == 1 )
                --it;

            if ( it == begin )
                break;
        }
    }


    Basically, I want to see how the item and the group header is added. The result shows that only the Thai named (but perhaps all non-Latin) are not put in the right position. Here is a beginning part of the result:

Code: Select all

childCount() = 2, dividerCount = 1
Item Text #0 |1|: B(e42-)
Item Text #1 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
childCount() = 4, dividerCount = 2
Item Text #0 |1|: '(e27-)
Item Text #1 |1|: B(e42-)
Item Text #2 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #3 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
childCount() = 6, dividerCount = 3
Item Text #0 |1|: %(e25-)
Item Text #1 |1|: '(e27-)
Item Text #2 |1|: B(e42-)
Item Text #3 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #4 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
Item Text #5 |13|: %0-H- (LaOn)(e25-e30-e2d-e48-e2d-e19-20-28-4c-61-4f-6e-29-)
childCount() = 8, dividerCount = 4
Item Text #0 |1|: !(e21-)
Item Text #1 |1|: %(e25-)
Item Text #2 |1|: '(e27-)
Item Text #3 |1|: B(e42-)
Item Text #4 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #5 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
Item Text #6 |13|: %0-H- (LaOn)(e25-e30-e2d-e48-e2d-e19-20-28-4c-61-4f-6e-29-)
Item Text #7 |20|: !2!H2▒%9 (Mama Blue)(e21-e32-e21-e48-e32-e1a-e25-e39-20-28-4d-61-6d-61-20-42-6c-75-65-29-)
childCount() = 10, dividerCount = 5
Item Text #0 |1|: 1b-)
Item Text #1 |1|: !(e21-)
Item Text #2 |1|: %(e25-)
Item Text #3 |1|: '(e27-)
Item Text #4 |1|: B(e42-)
Item Text #5 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #6 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
Item Text #7 |13|: %0-H- (LaOn)(e25-e30-e2d-e48-e2d-e19-20-28-4c-61-4f-6e-29-)
Item Text #8 |20|: !2!H2▒%9 (Mama Blue)(e21-e32-e21-e48-e32-e1a-e25-e39-20-28-4d-61-6d-61-20-42-6c-75-65-29-)
Item Text #9 |18|: (14L (Pansak)(e1b-e32-e19-e28-e31-e01-e14-e34-e4c-20-28-50-61-6e-73-61-6b-29-)
childCount() = 12, dividerCount = 6
Item Text #0 |1|: 1b-)
Item Text #1 |1|: !(e21-)
Item Text #2 |1|: %(e25-)
Item Text #3 |1|: '(e27-)
Item Text #4 |1|: B(e42-)
Item Text #5 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #6 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
Item Text #7 |13|: %0-H- (LaOn)(e25-e30-e2d-e48-e2d-e19-20-28-4c-61-4f-6e-29-)
Item Text #8 |20|: !2!H2▒%9 (Mama Blue)(e21-e32-e21-e48-e32-e1a-e25-e39-20-28-4d-61-6d-61-20-42-6c-75-65-29-)
Item Text #9 |18|: (14L (Pansak)(e1b-e32-e19-e28-e31-e01-e14-e34-e4c-20-28-50-61-6e-73-61-6b-29-)
Item Text #10 |1|: W(57-)
Item Text #11 |23|: Wolfgang Amadeus Mozart(57-6f-6c-66-67-61-6e-67-20-41-6d-61-64-65-75-73-20-4d-6f-7a-61-72-74-)



For example, when Wasana Poonpol (The second artist) is add (her name in Thai starts with the unicode E27), instead of appearing after the item '(e27-), her item append after B(e42-) which was created for LoSo.

(Code Up close 1)

Code: Select all

..
childCount() = 4, dividerCount = 2
Item Text #0 |1|: '(e27-)
Item Text #1 |1|: B(e42-)
Item Text #2 |11|: B%B
                       (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text #3 |28|: '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
...


But when Wolfgang Amadeus Mozart enters the sense his name appears in the right header (W(57-)).

(Code Up Close 2)

Code: Select all

...
childCount() = 12, dividerCount = 6
Item Text(1): 1b-)
Item Text(1): !(e21-)
Item Text(1): %(e25-)
Item Text(1): '(e27-)
Item Text(1): B(e42-)
Item Text(11): B%B
                   (LoSo)(e42-e25-e42-e0b-20-28-4c-6f-53-6f-29-)
Item Text(28): '2*2 9%% (Wasana Poonpol)(e27-e32-e2a-e19-e32-20-e1e-e39-e25-e1c-e25-20-28-57-61-73-61-6e-61-20-50-6f-6f-6e-70-6f-6c-29-)
Item Text(13): %0-H- (LaOn)(e25-e30-e2d-e48-e2d-e19-20-28-4c-61-4f-6e-29-)
Item Text(20): !2!H2▒%9 (Mama Blue)(e21-e32-e21-e48-e32-e1a-e25-e39-20-28-4d-61-6d-61-20-42-6c-75-65-29-)
Item Text(18): (14L (Pansak)(e1b-e32-e19-e28-e31-e01-e14-e34-e4c-20-28-50-61-6e-73-61-6b-29-)
Item Text(1): W(57-)
Item Text(23): Wolfgang Amadeus Mozart(57-6f-6c-66-67-61-6e-67-20-41-6d-61-64-65-75-73-20-4d-6f-7a-61-72-74-)
...


Despise how much I like this genius, he should get no special treat.

I also try changing the artist name to some other languages (japanese and chinese) and the result is the same.
I also try to change my Ubuntu locale to Thai (I am now in Canada so I am currently use CA_EN as my Locale) but still no use.

I search high-and-low (as a programmer who last use C++ was almost 10 years ago) and found that nothing can in Amarok code that seems to handle this. I bet it is somewhere in KDE or QT code (So why do I post post this in Amarok forum, you may ask).

Since messing up KDE code and re-compile it, it's one of the last thing a non-C++ programmer and a GNOME user (like me) will do so I thought I can post it here and someone else may be in the better position to solve this or at least delegate this problem to someone else better to solve.

If you think, this problem can be solved here in Amarok. You may guide me how and I may try to solve it.

Sorry for my poor English.
Have a very good day.