-
-
Notifications
You must be signed in to change notification settings - Fork 11.4k
qscintilla2: Updated formula for Python3, Added QtDesigner Plugin #26601
Conversation
Hello? Is anyone there? |
ENV.append_path "CPLUS_INCLUDE_PATH", "#{include}" | ||
inreplace "designer.pro" do |s| | ||
s.gsub! "$$[QT_INSTALL_LIBS]", lib | ||
s.gsub! "$$[QT_INSTALL_PLUGINS]", "#{HOMEBREW_PREFIX}/Cellar/qt/4.8.5/plugins" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can't hardcode the version here. Needs to be a symlink rather than install to this path.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's the best way to determine the version of an installed formula?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
"#{Formula.factory("qt").opt_prefix}/plugins"
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Never mind. I worked it out.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So it's better to link to the opt prefix than to the Cellar prefix?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@dakcarto I think you and I are doing similar things with the include path.
I used ENV.append_path "CPLUS_INCLUDE_PATH", "#{include}"
, rather than adding it to the args.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should I completely remove the plugin code, or just comment it out?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@MikeMcQuaid How about moving the ln_sf
to post_install
, then in the qt
formula test for installation of formulae that install a plugin and run their post_install
if found? That would re-establish the symlinks.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@daviewales Remove for now.
@dakcarto Nah, we really need to move plugins to a shared location instead.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@MikeMcQuaid Yeah that makes more sense. Alternatively, only the third-party plugins could be installed in a shared location, and Qt may be able to be configured to append that directory, similar to the QT_PLUGIN_PATH env var. (Not sure that is possible on Qt's configure setup, though.)
I wonder if we could use qt.conf to set the plugin directory PATH. |
I think qt.conf is read only during runtime, per application/process. I don't think it can be read globally by Qt for all processes. (Can't find any data to verify this, though.) |
I have done some quick tests, and I think you might be right. Perhaps the easiest thing to do would be to install plugins to |
By setting Since we are (mostly) just referring to the installation of Qt Designer plugins, it is possible to add some code to the In those instances, and if those formulae even exist, the caveats could be updated with a notice about using QT_PLUGIN_PATH, much in the same manner as when using PYTHONPATH. The separate, shared plugins directory sounds best, though, if it can be correctly managed on uninstall of any formulae. |
|
I've removed the plugin code from this pull request now, by the way. |
Plugin removed, until qt formula is updated to use a separate plugin directory, possibly in var.
I've updated this with a |
If we can merge |
Closes Homebrew#26601. Signed-off-by: Mike McQuaid <mike@mikemcquaid.com>
Closes Homebrew#26601. Signed-off-by: Mike McQuaid <mike@mikemcquaid.com>
No description provided.