SymbolEditor moved to KDE Review

[Steve Allewell]

On 16/06/14 22:54, Albert Astals Cid wrote:
> El Dimarts, 3 de juny de 2014, a les 06:34:08, Steve Allewell va escriure:
>> Greetings
>
> Hi there
>
>> The application SymbolEditor was recently imported to KDE Playground
>> under the KDE Incubator process (with KXStitch) having previously been
>> hosted on Sourceforge.
>>
>> SymbolEditor is a QPainterPath symbol creator and was originally written
>> to provide symbol libraries for the KXStitch application which is a
>> cross stitch pattern creator. There is a possibility that these symbol
>> libraries may be useful to other applications that may use similar
>> functionality or requirements. It is a stable and mature application
>> having been in development for a number of years and has had several
>> stable releases.
>>
>> An initial wiki entry has been created here.
>> http://userbase.kde.org/SymbolEditor
>>
>> A description of the file format used is here.
>> http://techbase.kde.org/User:Sallewell/SymbolEditor_File_Format
>> A more suitable home to be determined.
>>
>> Having fulfilled the incubation criteria, SymbolEditor has now been
>> moved to KDE Review with the aim of eventually moving to
>> Extragear/Graphics if the review is approved.
>> http://projects.kde.org/projects/kdereview/symboleditor
>>
>> To cover the requirements for submission:
>> * A Docbook manual is included in the source. Yuri Chornoivan has been
>> making some changes to reorganise the manual folder, fix some typos and
>> some images.
>>
>> * No libraries are supplied but apidox is used to document the code.
>>
>> * There are no krazy issues reported at EBN
>> http://ebn.kde.org/krazy/reports/playground/graphics/symboleditor/index.html
>> (not sure if this will move with the move to KDE Review)
>>
>> * No specific usability tests have been done, but SymbolEditor is being
>> used by a few users with few problems being reported bug or usability wise.
>>
>> * There are no obvious performance problems identified at this time.
>>
>> * SymbolEditor is completely translatable and a number of translations
>> already exist. Albert Astals Cid has moved these to l10n.
>>
>> Please review the SymbolEditor application and let me know what needs
>> attention to reach an acceptable position.
>
> Some comments:
>   * MainWindow::cleanChanged has a bit of a word puzzle, hard for translators to translate correctly, i'd just move the %1 down to the two if branches.

Pushed a fix to redefine the strings.

>   * While playing with it and valgrind i got
>
> ==2487== Conditional jump or move depends on uninitialised value(s)
> ==2487==    at 0x428AAA: Editor::mouseReleaseEvent(QMouseEvent*) (Editor.cpp:874)
> ==2487==    by 0x5CE8539: QWidget::event(QEvent*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x5C98E2B: QApplicationPrivate::notify_helper(QObject*, QEvent*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x5C9F599: QApplication::notify(QObject*, QEvent*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x50AF428: KApplication::notify(QObject*, QEvent*) (kapplication.cpp:311)
> ==2487==    by 0x6E4741C: QCoreApplication::notifyInternal(QObject*, QEvent*) (in /usr/lib/x86_64-linux-gnu/libQtCore.so.4.8.6)
> ==2487==    by 0x5C9ED92: QApplicationPrivate::sendMouseEvent(QWidget*, QMouseEvent*, QWidget*, QWidget*, QWidget**, QPointer<QWidget>&, bool) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x5D13A0A: QETWidget::translateMouseEvent(_XEvent const*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x5D132A8: QApplication::x11ProcessEvent(_XEvent*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0x5D3AB51: x11EventSourceDispatch(_GSource*, int (*)(void*), void*) (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.6)
> ==2487==    by 0xAFEFE43: g_main_context_dispatch (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4100.0)
> ==2487==    by 0xAFF0087: g_main_context_iterate.isra.24 (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4100.0)
>
> Have a look at Editor.cpp:874

Pushed a fix to initialise the variable in the mousePressEvent. No 
longer getting the message above.

>
> Cheers,
>    Albert
>
>>
>> Best Regards
>>
>>
>> Steve Allewell
>

Thanks for the feedback

Regards

Steve