Closed musicinmybrain closed 2 years ago
Thanks for catching this error. In fact, the m_parameters.value(key)
will always have default values passed with the constructor and do not rely on Qt nonexistent key behavior.
@musicinmybrain Feel free to comment and reopen this pull request for the follow-up of https://bugzilla.redhat.com/show_bug.cgi?id=2045367. For me, it's seem to be an error with the FFmpeg backend of OpenCV not loading [ INFO:0@17.374] global /builddir/build/BUILD/opencv-4.5.5/modules/videoio/src/backend_plugin.cpp (383) getPluginCandidates Found 0 plugin(s) for FFMPEG
and GStreamer the secondary backend not able to load the test movie [ WARN:0@17.487] global /builddir/build/BUILD/opencv-4.5.5/modules/videoio/src/cap_gstreamer.cpp (862) isPipelinePlaying OpenCV | GStreamer warning: GStreamer: pipeline have not been created
.
According to the Qt documentation, I expect that
m_parameters.value("normParam").toDouble())
with nonexistent keynormParam
should be0.0
. This PR corrects the key tonormPerim
.See also this downstream bug in Fedora Linux where the expression
m_parameters.value("normParam").toDouble())
producedNaN
on theppc64le
architecture with a pre-release of GCC 12 (probably a separate Qt6/GCC12 bug), which caused the test to actually fail, resulting in the discovery of this typo. This PR is not, by itself, a fix for the downstream bug since it appears that even the correct keynormPerim
isn’t expected to be present the first time through thewhile
loop—som_parameters.value(key).toDouble())
with nonexistentkey
still needs to return0.0
as specified.