[Bug 1143055] New: <END>+<HOME> not working in screen anymore
http://bugzilla.suse.com/show_bug.cgi?id=1143055 Bug ID: 1143055 Summary: <END>+<HOME> not working in screen anymore Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.1 Hardware: x86-64 OS: Other Status: NEW Severity: Minor Priority: P5 - None Component: Basesystem Assignee: bnc-team-screening@forge.provo.novell.com Reporter: doerges@pre-sense.de QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- The problem is very similar to bug 1095661, but the regression occurs when going from openSUSE Leap 15.0 to 15.1. Everything is working as expected in Leap 15.0 (with or without screen). In Leap 15.1 only the setup without screen is working. Inside screen I get this output: --- snip --- user@box:~> lsb-release -d Description: openSUSE Leap 15.1 user@box:~> echo $TERM screen.xterm-256color --- snip --- How to reproduce ---------------- - Start screen from the command line (tested with Konsole) -> a new shell is spawned inside screen - try cursor movement using <HOME> or <END> Actual behavior --------------- -> nothing happens Expected behavior ----------------- -> Cursor moves either to beginning or the end of the line Workaround ---------- Adding these two lines to ~/.inputrc helps: "\e[1~": beginning-of-line "\e[4~": end-of-line N.B.: The following does *not* help: --- snip --- user@box:~> export TERM=xterm-256color --- snip --- -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
Alynx Zhou
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c1
Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c2
Till Dörges
So does screen interpret home/end correctly? I.e. if you enter 'scrollback' mode (C-a ESC), does the cursor move to the start/end of the window if you press the keys?
Indeed that works. After hitting C-a ESC, HOME brings me to the top of the buffer and END to the end.
Or do the applications started insider of screen don't work correctly? In that case, do they work if you use TERM='screen' instead of that screen.xterm-256color entry?
The application then would be bash. export TERM=screen doesn't change the behavior of HOME/END. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c3
--- Comment #3 from Till Dörges
Or do the applications started insider of screen don't work correctly? In that case, do they work if you use TERM='screen' instead of that screen.xterm-256color entry?
The application then would be bash. export TERM=screen doesn't change the behavior of HOME/END.
TERM=screen inside the screen session doesn't change it. But if I do 'export TERM=screen' before issuing 'screen' HOME and END work as expected in bash, too. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c4
--- Comment #4 from Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c5
--- Comment #5 from Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c6
--- Comment #6 from Till Dörges
Here's something else to test: does everything work if you start screen with the '-a' option (which forces it to set $TERM to screen instead of screen.XXX)?
Both your workarounds resolve the problem (HOME and END working as expected in bash) - screen -a - inside screen: export TERM=screen exec bash In both cases TERM is set to "screen". As for the terminfo files I have --- snip --- user@box:~> rpm -qf /usr/share/terminfo/s/screen-xterm-256color error: file /usr/share/terminfo/s/screen-xterm-256color: Datei oder Verzeichnis nicht gefunden user@box:~> ll /usr/share/terminfo/s/screen-*256* -rw-r--r-- 1 root root 1988 27. Apr 21:42 /usr/share/terminfo/s/screen-256color -rw-r--r-- 1 root root 2000 27. Apr 21:42 /usr/share/terminfo/s/screen-256color-bce -rw-r--r-- 1 root root 2032 27. Apr 21:42 /usr/share/terminfo/s/screen-256color-bce-s -rw-r--r-- 1 root root 2018 27. Apr 21:42 /usr/share/terminfo/s/screen-256color-s user@box:~> ll /usr/share/terminfo/s/screen-*xterm* -rw-r--r-- 1 root root 3481 27. Apr 21:42 /usr/share/terminfo/s/screen-bce.xterm-new user@box:~> rpm -qf /usr/share/terminfo/s/screen-*256* /usr/share/terminfo/s/screen-*xterm* terminfo-base-6.1-lp151.5.41.x86_64 terminfo-6.1-lp151.5.41.x86_64 terminfo-6.1-lp151.5.41.x86_64 terminfo-6.1-lp151.5.41.x86_64 terminfo-6.1-lp151.5.41.x86_64 user@box:~> rpm -qf /usr/share/terminfo/x/xterm-256color terminfo-base-6.1-lp151.5.41.x86_64 user@box:~> rpm -ql screen | grep termin user@box:~> --- snap --- -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c7
Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c16
--- Comment #16 from Dr. Werner Fink
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c17
--- Comment #17 from Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c18
--- Comment #18 from Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c20
Dr. Werner Fink
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c21
Dr. Werner Fink
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c22
Marcus Meissner
http://bugzilla.suse.com/show_bug.cgi?id=1143055
Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c26
--- Comment #26 from Michael Schröder
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c27
--- Comment #27 from Thomas Dickey
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c28
--- Comment #28 from Dr. Werner Fink
In any case, Werner's comments pointed to a problem in libreadline, which should be addressed.
Already done, submit request is running ... still waiting on QA team. Maybe I should think about a wildcard solution to submit this to bash upstream and hope that Chet will accept this -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c29
--- Comment #29 from Thomas Dickey
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c30
--- Comment #30 from Dr. Werner Fink
hmm - I missed a detail from Werner's comment #24:
Just an example from /etc/inputrc.keys (included several times in /etc/inputrc for the editor modes emacs, vi move, and vi command mode) # # Screen # $if term=screen "\e[1;2D": backward-word "\e[1;2C": forward-word "\e[1;2A": up-history "\e[1;2B": down-history "\e[1;2H": beginning-of-line "\e[1;2F": end-of-line $if mode=emacs # Note that this file is included for vi-command, vi-insert, and # emacs mapping, therefore do switch within the last $include as # otherwise the vi modes do miss the common mappings. set editing-mode vi set keymap vi-command "\e[2;2~": vi-editing-mode "\e[2;5~": vi-editing-mode "\e[2;3~": vi-editing-mode set keymap vi-insert "\e[2;2~": vi-replace "\e[2;5~": vi-replace "\e[2;3~": vi-replace set editing-mode emacs set keymap emacs "\e[2;2~": overwrite-mode "\e[2;5~": overwrite-mode "\e[2;3~": overwrite-mode $endif [...] $endif now this should apply to e.g. TERM=scren, TERM=screen-256color, TERM=screen.xterm, TERM=screen.xterm-256color, ... but upstream libreadline only match the first one. Hence this had missed e.g. TERM=screen.xterm-256color -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c31
--- Comment #31 from Thomas Dickey
http://bugzilla.suse.com/show_bug.cgi?id=1143055
Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c32
--- Comment #32 from Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c33
--- Comment #33 from Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c34
--- Comment #34 from Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
Swamp Workflow Management
http://bugzilla.suse.com/show_bug.cgi?id=1143055
http://bugzilla.suse.com/show_bug.cgi?id=1143055#c35
Dr. Werner Fink
participants (1)
-
bugzilla_noreply@novell.com