https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=245614
--- Comment #20 from John Hein <[email protected]> --- (In reply to Tobias C. Berner from comment #19) On the contrary, I think this approach is the best (or at least a very good) solution - thanks. If more ports explicitly enabled / disabled dependencies (features), "the world would be a better place". This allows users to specify global preferences (e.g., OPTIONS_UNSET=ALSA in make.conf) and override for individual ports as the situation dictates. Auto-detecting for enable/disable is probably the cause of many failures on bugs.freebsd.org (poudriere or not). Auto-detecting _location_ (of headers, libs - e.g., via pkgconf) is another story (and is a good thing generally). That said, it's still a qmake mystery to me why / how qmake detected that it was okay to use the "internal" (presumably that's another name for "bundled") version of alsa, but it failed at build time. I don't see any bundled alsa or flite/alsa in the qt5-speech distribution. Also, your patch has a problem with FLITE=on: . . Qt TextToSpeech: Flite .................................. yes Flite with ALSA ........................ no Speech Dispatcher ...................... yes . . ===> Checking for items in pkg-plist which are not in STAGEDIR Error: Missing: %%QT_CMAKEDIR%%/Qt5TextToSpeech/Qt5TextToSpeech_QTextToSpeechEngineFlite.cmake Error: Missing: %%QT_PLUGINDIR%%/texttospeech/libqttexttospeech_flite.so ===> Error: Plist issues found. *** Error code 1 Maybe somewhere in the qmake stuff there's something that confuses the flite and flite_alsa feature implementations? -- You are receiving this mail because: You are the assignee for the bug.
