dc.collen is the length of dc.col, not the maximum index, hence if x is
equal to dc.collen, then it's an error.
With config.def.h, the last valid index is 259, so this correctly
reports "black":
$ printf '\033]4;259;?\e\\'
260 is an invalid index and this reports garbage instead of printing an
error:
$ printf '\033]4;260;?\e\\'
ref.
https://git.suckless.org/st/commit/a6bbc0c96b0a1db804061b0db79101c6b26aec57.html
This patch fixes the following sixel issues:
- The current sixel implementation cleared all cells from the left side
of the image when the image was drawn. The fix only clears the cells
where the image will be drawn.
- The deletion routine didn't work correctly. In certain situations,
it left the image or images undrawn. For example, if the first image
was marked for deletion, it didn't draw the second one.
- The drawing routine caused a high cpu usage, because XCopyArea()
triggered the X server to send the NoExpose event, which caused sixels
to be redrawn and the X server to send another NoExpose event and so
on. This loop caused constant redraw of sixels and high cpu usage.
The fix prevents the X server from sending GraphicsExpose and NoExpose
events.
The patch also adds a control sequence for removing sixels:
Because the sixels are implemented as overlay images, they cannot be
removed by clearing the underlaying cells. Therefore, we need a control
sequence to remove them. I opted to choose ESC[6J as the control
sequence because it is not used and the number refers to sixels. So when
the lf file manager supports sixels [1], you can use the following
minimal scripts to preview images in lf:
previewer:
#!/bin/sh
case "$(readlink -f "$1")" in
*.bmp|*.gif|*.jpg|*.jpeg|*.png|*.webp|*.six|*.svg|*.xpm)
chafa -s "$(($2-3))x$3" -f sixels "$1"
exit 1 ;;
*)
bat "$1" ;;
esac
cleaner:
#!/bin/sh
printf "\033[6J" >/dev/tty
[1] https://github.com/gokcehan/lf/pull/1211
fix buffer overflow when handling long composed input
To reproduce the issue:
"
If you already have the multi-key enabled on your system, then add this line
to your ~/.XCompose file:
[...]
<question> <T> <E> <S> <T> <question> :
"1234567890123456789012345678901234567890123456789012345678901234567890"
"
Reported by and an initial patch by Andy Gozas <andy@gozas.me>, thanks!
Adapted the patch, for now st (like dmenu) handles a fixed amount of composed
characters, or otherwise ignores it. This is done for simplicity sake.
Ref.
https://git.suckless.org/st/commit/e5e959835b195c023d1f685ef4dbbcfc3b5120b2.html
The openurlonclick and scrollback patches are now working together,
so links can be clicked in the scrollback buffer too. This update also
adds url underlining and other improvements to the openurlonclick patch.
The full list of changes in the openurlonclick patch:
- Adds scrollback support
- Adds modkey option
- Better url detection
- Underlines url when the mouse pointer is over a link
- Opens a browser as a background process, so it won't lock the terminal anymore
- Fixes a segmentation fault bug
This patch replaces the previous one I sent.
The following changes are made in this patch:
- Fix tracking of pressed buttons. Previously, pressing two buttons and
then releasing one would make st think no buttons are pressed, which
in particular broke MODE_MOUSEMOTION.
- Always send the lowest-numbered pressed button on motion events; when
no button is pressed for a motion event in MODE_MOUSEMANY, then send
a release. This matches the behaviour of xterm. (Previously, st sent
the most recently pressed button in the motion report.)
- Remove UB (?) access to potentially inactive struct member
e->xbutton.button of XEvent union.
- Fix (unlikely) possibility of overflow for large button numbers.
The one discrepancy I found between st and xterm is that xterm sometimes
encodes buttons with large numbers (>5) strangely. E.g., xterm reports
presses of buttons 8 and 9 as releases, whereas st properly (?) encodes
them as presses.
Ref.
- https://git.suckless.org/st/commit/ea7cd7b62fdfa6a1fbd882d1565d557577f2cf32.html
from the XmbTextListToTextProperty(3) man page:
"If insufficient memory is available for the new value string, the functions
return XNoMemory. If the current locale is not supported, the functions return
XLocaleNotSupported. In both of these error cases, the functions do not set
text_prop_return."
Reported by Steffen Nurpmeso <steffen@sdaoden.eu>, thanks!
Ref. https://git.suckless.org/st/commit/2f6e597ed871cff91c627850d03152cae5f45779.html
The bits of uint signal in an XKeyEvent which concern the key group (keyboard
layout) are bits 13 and 14, as documented here:
https://www.x.org/releases/X11R7.7/doc/libX11/XKB/xkblib.html#Groups_and_Shift_Levels
In the older version, only bit 13 was marked as part of XK_SWITCH_MOD, this
causes issues for users who have more than two keymaps. The 14th bit is not
in ignoremod, key sequences are not caught by match(), if they switch to a third
or fourth keyboard.
If the mouse cursor is changed to a bar or an underline then st will use that
when the terminal is first opened. When an application that changes the cursor
via escape sequences is executed, e.g. vim which uses a block cursor by default,
then that cursor will remain after exiting the program.
This change sets the cursor back to default when exiting alt mode.
These are typically mapped in X11 to the side-buttons (backward/forwards) on
the mouse. A comparison of the button numbers in SGR mode (first field):
st old:
0 1 2 64 65 66 67 68 69 70
st new (it is the same as xterm now):
0 1 2 64 65 66 67 128 129 130
A script to test and reproduce it, first argument is "h" (on) or "l" (off):
#!/bin/sh
printf '\x1b[?1000%s\x1b[?1006%s' "$1" "$1"
for n in 1 2 3 4 5 6 7 8 9 10; do
printf 'button %d\n' "$n"
xdotool click "$n"
printf '\n\n'
done
https://git.suckless.org/st/commit/9e68fdbcdb06dfa3d23fe3a7a7f7b59e40e1ea2f.html
