In vi-mode lineedit tries to detect some escape sequences.
After the ESC it reads the next character to check for certain
values. This read should have a timeout or a user-entered ESC to
switch to command mode doesn't properly handle the next character.
Signed-off-by: Ron Yorston <rmy@tigress.co.uk>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
In vi mode, the 'p' and 'P' commands caused a segfault when nothing had
been put in the buffer yet because the delptr was not initialized.
Signed-off-by: Shawn J. Goff <shawn7400@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
When CONFIG_FEATURE_EDITING_SAVE_ON_EXIT is set to y, the histfile
will get cleared if the total amount of history lines is less than MAX_HISTORY.
Only if the histfile is not empty _and_ the amount of lines currently
in memory are equal to or greater than MAX_HISTORY, history saving will
work as expected with this feature enabled.
Output from defconfig + CONFIG_FEATURE_EDITING_SAVE_ON_EXIT=y:
$ echo "foo" > ~/.ash_history
$ ./busybox ash
~/busybox/a $ echo "bar" > /dev/null
~/busybox/a $ exit
$ cat ~/.ash_history
$
Output with the patch applied and same config as above:
$ echo "foo" > ~/.ash_history
$ ./busybox ash
~/busybox/b $ echo "bar" > /dev/null
~/busybox/b $ exit
$ cat ~/.ash_history
foo
echo "bar" > /dev/null
exit
$
Signed-off-by: Dennis Groenen <tj.groenen at gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
First change fixes "read -n NUM". Apparently poll() won't report
data availability if cc[VMIN] > 1 until there are at least cc[VMIN] bytes.
function old new delta
read_line_input 3885 3877 -8
shell_builtin_read 1097 1087 -10
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-18) Total: -18 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Using ash in busybox git version dea28e1e, tab completion doesn't seem
to work properly for filenames that have special characters (such as
spaces) in them. For example, with filenames "foo bar" and "foo zap",
typing "ls fo<TAB>" correctly expands to "ls foo\ ", but then
continuing to type "b<TAB>" will produce "ls foo\ bbar", which is not
correct (the 'b' is duplicated).
Signed-off-by: Mike Shal <marfey@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
While at it, make ESC sequences more readable; and removed check for
isatty(stdin) in reset. Code shrink:
text data bss dec hex filename
884771 936 17216 902923 dc70b busybox_old
884723 936 17216 902875 dc6db busybox_unstripped
Signed-off-by: Marek Polacek <mmpolacek@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Now it is allocated temporarily only for the duretion of prefix generation,
and also we only allocate the needed size, not maximally possible.
function old new delta
build_match_prefix 579 590 +11
remove_chunk 43 28 -15
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/1 up/down: 11/-15) Total: -4 bytes
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
pos_buf is a strange hack, easy to do without it.
This also allows lines >32k long to be handled.
Also simplified match prefix generations and made behavior more like bash.
function old new delta
remove_chunk - 43 +43
collapse_pos 79 - -79
build_match_prefix 804 579 -225
------------------------------------------------------------------------------
(add/remove: 1/1 grow/shrink: 0/1 up/down: 43/-304) Total: -261 bytes
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Some logic changes and function renames. The fix will follow this patch,
to mkae it distinct from cleanup.
function old new delta
build_match_prefix - 892 +892
username_completion - 121 +121
read_line_input 4902 4966 +64
username_tab_completion 235 - -235
find_match 892 - -892
------------------------------------------------------------------------------
(add/remove: 2/2 grow/shrink: 1/0 up/down: 1077/-1127) Total: -50 bytes
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
