--===============9146435563989394464== Content-Type: multipart/alternative; boundary="===============4876295939436905722==" --===============4876295939436905722== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable > On March 17, 2012, 9:13 a.m., Thomas Zander wrote: > > The change had to do with a right-to-left layout where +1 would not go = to the visible right, but to the left. > > = > > If this fixes the unit test, then please commit. I'll make a note to f= ind the case where the old code broke for and add a unit test for that one = too, so a solution can be found for all items :) Sounds good. I'll push it, then. - Robert Mathias ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/104272/#review11500 ----------------------------------------------------------- On March 17, 2012, 1:08 a.m., Robert Mathias Marmorstein wrote: > = > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/104272/ > ----------------------------------------------------------- > = > (Updated March 17, 2012, 1:08 a.m.) > = > = > Review request for KOffice and Thomas Zander. > = > = > Description > ------- > = > This one-line change seems to have broken the kword-frames-TestDocumentLa= yout unit test. I don't know whether the behaviour of Qt has changed or if= the cursorToX function simply doesn't work as expected, but the test passe= s after reverting this change. > = > = > Diffs > ----- > = > kword/part/frames/KWTextDocumentLayout.cpp fc5173d = > = > Diff: http://git.reviewboard.kde.org/r/104272/diff/ > = > = > Testing > ------- > = > Everything builds, compiles, runs, and passes the same unit tests (plus o= ne more!) if we revert the change. However, it's possible that it's the te= st that's wrong. I don't understand the layout algorithm well enough to kn= ow for sure. > = > = > Thanks, > = > Robert Mathias Marmorstein > = > --===============4876295939436905722== 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/104272/

On March 17th, 2012, 9:13 a.m., Thomas Zand= er wrote:

The chang=
e had to do with a right-to-left layout where +1 would not go to the visibl=
e right, but to the left.

If this fixes the unit test, then please commit.  I'll make a note to f=
ind the case where the old code broke for and add a unit test for that one =
too, so a solution can be found for all items :)
Sounds good=
.  I'll push it, then.

- Robert Mathias


On March 17th, 2012, 1:08 a.m., Robert Mathias Marmorstein wrote:

Review request for KOffice and Thomas Zander.
By Robert Mathias Marmorstein.

Updated March 17, 2012, 1:08 a.m.

Descripti= on 

This one-line change seems to have broken the kword-frames-T=
estDocumentLayout unit test.  I don't know whether the behaviour of Qt =
has changed or if the cursorToX function simply doesn't work as expecte=
d, but the test passes after reverting this change.

Testing <= /h1> 
Everything builds, compiles, runs, and passes the same unit =
tests (plus one more!) if we revert the change.  However, it's possible=
 that it's the test that's wrong.  I don't understand the layou=
t algorithm well enough to know for sure.

Diffs=

  • kword/part/frames/KWTextDocumentLayout.cpp (fc5173d)

View Diff
--===============4876295939436905722==-- --===============9146435563989394464== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ koffice-devel mailing list koffice-devel@kde.org https://mail.kde.org/mailman/listinfo/koffice-devel --===============9146435563989394464==--