| Commit message (Collapse) | Author | Age | Lines |
| ... | |
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
As described in #883, creating a block selection (with ctrl+BTN_LEFT),
then *quickly* (within 300ms from) creating a new one, will, in fact,
_not_ create a new block selection, but a ‘select-word-whitespace’
selection (ctrl+BTN_LEFT-2).
A similar effect can be seen with plain selections (BTN_LEFT). Click
and drag to make a selection, but *release*, and make a new selection
within 300ms from the initial button press, and the new selection is
suddenly a word-based selection.
This happens because triggering a binding does *not* reset the button
click counter.
So, shouldn’t we just do that?
No, because we rely on this behavior to handle single-, double- and
triple-click actions. If we were to reset the button click handler,
then we’d have to instead delay triggering the first action with
300ms, which would make things appear laggy. If we don’t do this, it
would be impossible to double- and triple-click, since the
single-click action would keep resetting the click counter.
This patch takes a slightly different approach; we reset the click
counter when the mouse has moved “too much”. For now, “too much” is
when the cursor has moved to a different cell.
This way, single-, double- and triple-clicks continue to work as
before. But, creating actual selections, and then quickly releasing
and starting a new selection produces the expected results.
Closes #883
|
| |
|
|
|
| |
We’re already doing this for keyboard input, and regular clipboard
pastes.
|
| |\
| |
| |
| | |
Closes #876
|
| |/
|
|
|
|
|
|
|
|
|
| |
The compositor may offer version 4, while we’re still linked against
an old libwayland that does not implement version 4.
While we properly #ifdef the callbacks from version 4, we were still
requesting version 4 (if offered by the compositor), even when
compiled against a too old libwayland.
Closes #876
|
| |\ |
|
| | | |
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
| |
Our getopt_long() based argument parser would (correctly) break out of
the while loop immediately. However, we would then proceed and
decrement ‘argc’ by 1, resulting in it being -1.
This was then passed to the terminal constructor, which passed it on
to the client application fork+exec logic. That finally resulted in an
exec(3) failure:
err: slave.c:339: SHELL=/bin/zsh: failed to execute: No such file or directory
err: fdm.c:215: no such FD: 7
info: main.c:604: goodbye
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\
| |
| |
| | |
Closes #762
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Ensure the outline width doesn’t exceed the cell width/height.
|
| |\ |
|
| |/ |
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Before this patch, foot only applied [scrollback].multiplier on the
normal screen, never the alt screen.
However, scrolling can be done in two ways on the alt screen:
If the application has enabled mouse support, we simply pass on the
mouse scroll events to the application. Here, it makes sense to not
apply the multiplier, and instead let the application choose how much
to scroll for each scroll event.
But, if the application has not enabled mouse support, we can still
scroll by simulating the arrow keys being pressed - alternate
scrolling (private mode 1007).
This is enabled by default in foot (but can be disabled in foot.ini
with the [mouse].alternate-scroll-mode setting).
In this mode, it makes more sense to apply the multiplier. And that’s
what this patch changes - the multiplier is now applied, on the alt
screen, when the application has not enabled mouse support, and
alternate scrolling has been enabled in foot.
Closes #859
|
| |/ |
|
| | |
|
| |\ |
|
| |/ |
|
| |
|
|
|
|
|
|
|
|
| |
If we have an “active” OSC-8 URI, term_print() would correctly replace
an existing URI with the new one.
But, if we don’t have an active URI, an existing URI was not
erased. This can be reproduced with e.g
echo -e '\e]8;;http://foo.bar\e\\foobar\e]8;;\e\\\b\b\b\b\b😀\n'
|
| | |
|
| |\ |
|
| |/ |
|
| |\ |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When using indexed colors (i.e. SGR 30/40/90/100), store the index
into the cell’s fg/bg attributes, not the actual color value.
This has a couple of consequences:
Color table lookup is now done when rendering. This means a rendered
cell will always reflect the *current* color table, not the color
table that was in use when the cell was printed to.
This simplifies the OSC-4/104 logic, since we no longer need to update
the grid - we just have to damage it to trigger rendering.
Furthermore, this change simplifies the VT parsing, since we no longer
need to do any memory loads (except loading the SGR parameter values),
only writes.
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
The missing return meant failures were not reported correctly.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
This fixes a memory leak when we failed to parse a command line (for
example, because it contained an unclosed quote).
|
| | |
|
| | |
|
| |\
| |
| |
| | |
Closes #857
|
| | | |
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| |\
| |
| |
| | |
Closes #851
|
| | | |
|
| |/
|
|
| |
Closes #851
|
| |\
| |
| |
| | |
Closes #844
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When going through the cached buffers, we only set buffer->busy on
the *first* re-usable buffer we found.
In some cases, we will find more than one re-usable buffer. In this
case, we select the “youngest” one (i.e the one most recently used, in
the hopes that we can use damage tracking instead of re-rendering the
entire buffer).
If the “current” buffer is younger than the previously detected,
re-usable, buffer, then we unref:ed the previously selected buffer,
and replaced it with the current one.
But, we did not sanitize it. That is, we did not:
* set buffer->busy
* clear its dirty region
* clear its scroll damage
That buffer would eventually get rendered to, and committed to the
compositor. Later, the compositor would free it. And there, in our
buffer_release() callback, we’d assert that buffer->busy was
set. And fail.
Closes #844
|
| |\ |
|
| |/ |
|
| |\ |
|
| |/ |
|
| |\ |
|
| |/
|
|
|
|
|
|
|
|
|
| |
When handling “generic” keys (i.e. keys not in the Kitty keymap), we
use the pressed key’s Unicode codepoint as “key” in the kitty CSI.
If we failed to convert the XKB symbol to a Unicode codepoint, we used
to (before this patch), fallback to using the XKB symbol as is.
This can never be correct... and it caused us to emit a meaningless
CSI for XKB_KEY_ISO_Next_Group, which confused e.g. Kakoune.
|
| |\
| |
| |
| | |
Closes #842
|
| |/
|
|
| |
Closes #842
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| | |
These are new in version 4 of the wl_output interface (first included
in wayland-1.20).
This allows us to get the name and description of the outputs, also on
compositors without the XDG output interface.
|
| |/
|
|
|
|
|
|
|
|
|
| |
This makes us slightly more resilient against a missing XDG output
interface.
We use the “real” (the physical) dimensions, combined with the scaling
factor, to estimate the logical dimensions.
This works out correctly with non-fractional scaling, but not
otherwise.
|
| |
|
|
| |
If term_to_slave() fails, then the key wasn’t ‘handled’...
|
| |\ |
|
| |/ |
|
| |\ |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
With this patch, key- and mouse-bindings structs (the non-layout
specific ones) are unified into a single struct.
The logic that parses, and manages, the key- and mouse binding lists
are almost identical. The *only* difference between a key- and a mouse
binding is that key bindings have an XKB symbol, and mouse bindings a
button and click-count.
The new, unified, struct uses a union around these, and all functions
that need to know which members to use/operate on now takes a ‘type’
parameter.
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We want to handle SIGPIPEs without crashing... One way to trigger this
was to use e.g.
pipe-visible=[cat /foo/bar] Control+Shift+q
That is, pipe output to something that did not consume it. This led
to a SIGPIPE when we tried to write the terminal contents to the pipe,
and crashed the whole foot instance.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This allows the user to write, to swap two key bindings:
[key-bindings]
show-urls-launch=Control+Shift+r
search-start=Control+Shift+u
instead of:
[key-bindings]
search-start=none
show-urls-launch=Control+Shift+r
search-start=Control+Shift+u
This should simplify the configuration for people who replace a lot of
the default key bindings.
This also simplifies the parsing somewhat, since we no longer has to
parse a key-binding into a temporary list, ensure it doesn’t have any
collisions, and then copy it into the actual key binding list.
_While_ parsing a key-binding, we still need a temporary array (at
least for the time being), but that temporary array is no longer
visible outside the parsing function that allocates it.
|
| |/ |
|
| | |
|
| |\
| |
| |
| | |
Part of #813
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| | |
This adds tests for all [key-bindings] options, as well as an invalid
“action”.
The tests also verify key combo collisions are detected correctly.
|
| | | |
|