> On jul. 9, 2015, 9:20 p.m., Albert Astals Cid wrote: > > Why are you removing the test? Seems like a value test to me to make sure > > we're differentiating when # is a char of the file and when # refers to a > > page number. > > Jan Kundrát wrote: > Because that test always fails right now. > > I wasn't aware of the #page convention. I'll be happy if the test was > fixed instead of course. Feel free to reject this patch.
The tests *have* to be fixed. So yeah if you don't plan on not removing the test from this I have to reject it. - Albert ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://git.reviewboard.kde.org/r/124026/#review82296 ----------------------------------------------------------- On jun. 8, 2015, 8:59 a.m., Jan Kundrát wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://git.reviewboard.kde.org/r/124026/ > ----------------------------------------------------------- > > (Updated jun. 8, 2015, 8:59 a.m.) > > > Review request for Okular. > > > Repository: okular > > > Description > ------- > > When this test was written, converting from user-supplied strings to > URLs with support for guessing whether the URL represents a local file > (but only if it exists) or a remote URL was something not provided by > Qt. That has changed in Qt 5.4, and what we have now is a unit test > which is too stupid because it blindly concatenates cwd with a > user-provided input. > > My understanding is that pre-5.4 Qt is supported on a best-effort basis, > and that there's little point it carrying unit tests which only > duplicate Qt's native features. Especially when these unit tests fail. > > > Diffs > ----- > > autotests/CMakeLists.txt PRE-CREATION > autotests/shelltest.cpp 0f750ae1f816e523f1bfaa8c19b9c947f0d0f8f3 > shell/okular_main.cpp b66b6a5f7cd30e6d123ba0b0a2d871e11807ea15 > shell/shellutils.h d5518c344c47dd81033a883cc43cca387764d362 > shell/shellutils.cpp 0dd737cf70ac888225a919799bd71905cfb89580 > > Diff: https://git.reviewboard.kde.org/r/124026/diff/ > > > Testing > ------- > > I have no idea whether this introduces any regressions. There's plenty of > test failures, both on the official KDE CI platform *and* on my local dev > box. How do I, for example, persuade the KService/whatever to ignore the > systemwide installation of Okular and to use, e.g., poppler generator from > the build directory? > > > Thanks, > > Jan Kundrát > >
_______________________________________________ Okular-devel mailing list Okular-devel@kde.org https://mail.kde.org/mailman/listinfo/okular-devel
