Hello Jeremy! Some first impressions from XQuartz 2.7.0 (xorg-server 1.10.99.901) on Snow Leopard. When I have one GNU Emacs window partially hiding another GNU Emacs window and I want to pick the top-most one with the mouse at the title bar, than this event is sent to the X client below, for example selecting there a line of text. The picking succeeds when I pick that part of the title bar which is outside of the buried X client. These malfunctions are quite often. The GNU Emacsen I use are of version 24.0.50 and compiled as a first try: In GNU Emacs 24.0.50.1 (x86_64-apple-darwin10.8.0, X toolkit, Xaw3d scroll bars) of 2011-07-14 on peter-dyballas-macbook-pro.fritz.box Windowing system distributor `The X.Org Foundation', version 11.0.11099901 configured using `configure '--without-sound' '--without-dbus' '--without-pop' '--without-gconf' '--without-gsettings' '--without-gpm' '--without-xpm' '--without-jpeg' '--without-tiff' '--without-gif' '--without-png' '--without-rsvg' '--with-x-toolkit=athena' '--enable-locallisppath=/Library/Application Support/Emacs/calendar24:/Library/Application Support/Emacs' 'CFLAGS=-g -H -pipe -fPIC -fno-common -Os' 'LDFLAGS=-Wl,-dead_strip_dylibs -Wl,-bind_at_load -Wl,-t' 'PKG_CONFIG_PATH=/opt/local/lib/pkgconfig:/opt/local/share/pkgconfig:/usr/lib/pkgconfig'' I also have GNU Emacs 23.3 as installed with MacPorts – with Motif. To have a zoo. Others are using GTK+2 or GTK+3. I think I'll have an eye on this and be making some tests on purpose. -- Greetings Pete Progress (n.): Process through which USENET evolved from smart people in front of dumb terminals to dumb people in front of smart terminals.
Hello! I think the problem is of a different nature. (I sometimes thought it could come from the fact that one GNU Emacs instance runs with super-user privileges, launched as 'sudo -H ...'.) It can *often* be reproduced when I leave GNU Emacs and switch to a different space to work with a native Mac OS X application (OmniWeb, Mail, ...). When I then return and choose a different GNU Emacs instance or other X client (gkrellm, Sunclock) this one does not get selected. I first have to remember and click the last selected X client to be able to select and use another X client. (I did not test with text input and X clients with more than one open window.) -- Greetings Pete "By filing this bug report you have challenged the honor of my family. Prepare to die!"
Am 30.07.2011 um 10:45 schrieb Peter Dyballa:
I think the problem is of a different nature. (I sometimes thought it could come from the fact that one GNU Emacs instance runs with super-user privileges, launched as 'sudo -H ...'.) It can *often* be reproduced when I leave GNU Emacs and switch to a different space to work with a native Mac OS X application (OmniWeb, Mail, ...). When I then return and choose a different GNU Emacs instance or other X client (gkrellm, Sunclock) this one does not get selected. I first have to remember and click the last selected X client to be able to select and use another X client. (I did not test with text input and X clients with more than one open window.)
Actually I have to click gkrellm or Sunclock to make GNU Emacsen behave sensibly again. This is the most reliable way. Sometimes it works to click into the window's bar when it is no obscuring another window. Then it can happen that the click event is received by the partially obscured X client. Sunclock and gkrellm are mostly only partially unobscured, so by clicking their window bars with no other X client below things become normal again. Another bug is with selecting text. When I do this somewhere near the top of an Emacs window and far below for example a compilation is producing text output, then I cannot copy the selected text with ⌘-c and insert it in another X client's (another GNU Emacs) window. What gets inserted here is a lot of text that was obviously produced during the period from start to end of text selection. Into a native Mac OS X application (the AppKit or NS/Cocoa variants of GNU Emacs) the selected and high-lighted text gets inserted. -- Greetings Pete Progress (n.): Process through which USENET evolved from smart people in front of dumb terminals to dumb people in front of smart terminals.
participants (2)
-
Peter Dyballa
-
Peter Dyballa