--===============1637011679081167694== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/103313/#review8695 ----------------------------------------------------------- kdeui/itemviews/kcategorizedview.cpp I would say 'no halfway exit always' especially in long functions. Midf= unc exit is too easily overlooked. That said, if most of processing is done= in else loop if you elide the midfunc exit then the function is even more = a refactoring candidate. If readability suffers... yet another reason for r= efactoring. = - Ruurd Pels On Dec. 4, 2011, 8:42 a.m., Jaime Torres Amate wrote: > = > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/103313/ > ----------------------------------------------------------- > = > (Updated Dec. 4, 2011, 8:42 a.m.) > = > = > Review request for kdelibs and Rafael Fern=C3=A1ndez L=C3=B3pez. > = > = > Description > ------- > = > Basically, what I do is: > If there are one or zero columns, hide the horizontalScrollBar until it i= s needed. (it has worked in the past, but in another file). > Apply the same strategy with files. > = > Additional stuff: > Moved the common calculus of itemSize outside of the if then else. > = > = > This addresses bugs 213068 and 287847. > http://bugs.kde.org/show_bug.cgi?id=3D213068 > http://bugs.kde.org/show_bug.cgi?id=3D287847 > = > = > Diffs > ----- > = > kdeui/itemviews/kcategorizedview.cpp 46a1cde = > kutils/kpluginselector.cpp ca0691d = > = > Diff: http://git.reviewboard.kde.org/r/103313/diff/diff > = > = > Testing > ------- > = > Krunner config does not loop (neither kgetnewstuff from kstars). I can no= t test with amarok (I've hit by an amarok start bug). > Please, test with other programs. > = > = > Thanks, > = > Jaime Torres Amate > = > --===============1637011679081167694== Content-Type: text/html; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable
This is an automatically generated e-mail. To reply, visit: http://git.revie= wboard.kde.org/r/103313/

= =
kdeui/itemviews/kcategorizedview.cpp (Diff revision 2)
void KCategorizedView::rowsAboutToBeRemoved(const QModelIndex &=
parent,
1376
        return;
I would say 'no halfway exit always' especially in long func=
tions. Midfunc exit is too easily overlooked. That said, if most of process=
ing is done in else loop if you elide the midfunc exit then the function is=
 even more a refactoring candidate. If readability suffers... yet another r=
eason for refactoring. 

- Ruurd


On December 4th, 2011, 8:42 a.m., Jaime Torres Amate wrote:

Review request for kdelibs and Rafael Fern=C3=A1ndez L=C3=B3pez.
By Jaime Torres Amate.

Updated Dec. 4, 2011, 8:42 a.m.

Descripti= on

Basically, what I do is:
If there are one or zero columns, hide the horizontalScrollBar until it is =
needed. (it has worked in the past, but in another file).
Apply the same strategy with files.

Additional stuff:
Moved the common calculus of itemSize outside of the if then else.

Testing <= /h1>
Krunner config does not loop (neither kgetnewstuff from ksta=
rs). I can not test with amarok (I've hit by an amarok start bug).
Please, test with other programs.
Bugs: 213068, = 287847

Diffs=

  • kdeui/itemviews/kcategorizedview.cpp (46a1= cde)
  • kutils/kpluginselector.cpp (ca0691d)

View Diff

--===============1637011679081167694==--