<feed xmlns='http://www.w3.org/2005/Atom'>
<title>mirrors/foot.git, branch spawn-terminal-cmdline-arguments</title>
<subtitle>A fast, lightweight and minimalistic Wayland terminal emulator</subtitle>
<id>https://git.0x0f.dev/mirrors/foot.git/atom?h=spawn-terminal-cmdline-arguments</id>
<link rel='self' href='https://git.0x0f.dev/mirrors/foot.git/atom?h=spawn-terminal-cmdline-arguments'/>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/'/>
<updated>2026-11-12T11:01:00Z</updated>
<entry>
<title>spawn-terminal: spawn with all the original command line options</title>
<updated>2026-11-12T11:01:00Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-11-12T10:50:55Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=f51f18d33557ea9078f7781f4a4235706228f5c9'/>
<id>urn:sha1:f51f18d33557ea9078f7781f4a4235706228f5c9</id>
<content type='text'>
Before this, spawn-terminal would spawn a new foot or footclient,
but *without* any of the command line options the user started the
original foot/footclient with.

We did have a hack for --config, but all other options were forgotten
in the new instance.

This patch fixes this.

* term_init() stores the raw argc/argv in the terminal struct.
* term_spawn_new() calls spawn() with the stored argc/argv.
* for single instances, main() calls term_init() with the unparsed
  argc/argv.
* the client/server protocol is extended to always send the raw
  argc/argv to the server.
* the server calls term_init() with the argc/argv from the client.
* bonus: on linux, we now use /proc/self/exe to determine which binary
  to launch. This means if you run e.g. a foot from your build tree,
  spawn-terminal will exec _that_ foot, not the system foot.

A couple of behavioral changes, and/or side-effects, caused by the
above (not exhaustive):

* footclient -E: spawn-terminal will not use the _exact_ environment
  from the original footclient, but will simply pass -E to the new
  footclient. This means the new instance will inherit the *current*
  environment of the original instance.
* foot htop: spawn-instance will _also_ launch a new foot+htop. Before
  this patch, spawn-instance would launch a new foot+shell.

Closes #1622
</content>
</entry>
<entry>
<title>ime: reset view &amp; cancel selection on text commit</title>
<updated>2026-11-12T08:54:23Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-27T04:56:05Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=8768826c887c27721fddbf4670e27215b201bce2'/>
<id>urn:sha1:8768826c887c27721fddbf4670e27215b201bce2</id>
<content type='text'>
IMEs may commit text directly without updating preedit text
(seen on Squeekboard), thus 5c8d520807 is not enough.
</content>
</entry>
<entry>
<title>selection: fix multi-cell chars selection update</title>
<updated>2026-10-28T08:39:41Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-28T08:39:41Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=61e88e44fc0f8ad8aa2ab2f69968d7a8ceaa8a77'/>
<id>urn:sha1:61e88e44fc0f8ad8aa2ab2f69968d7a8ceaa8a77</id>
<content type='text'>
As in 0e7e943062, absolute row numbers cannot be used for comparison.
Here it may break selection update with multi-cell characters.

Note that there's no need to compare the coordinates again.
A better fix is to check the selection direction instead.
</content>
</entry>
<entry>
<title>wayland: update_terms_on_monitor(): skip terminals in the process of shutting down</title>
<updated>2026-10-25T11:08:15Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-10-25T11:08:15Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=152487e494993ad7fb5633a27c7d7ad0f8810ea5'/>
<id>urn:sha1:152487e494993ad7fb5633a27c7d7ad0f8810ea5</id>
<content type='text'>
In rare cases, xdg_output_handle_done() may be called while a terminal
is shutting down. This triggered a crash in
update_terms_for_monitor(), which de-references term-&gt;window (which is
NULL for terminals shutting down).

Closes #2487
</content>
</entry>
<entry>
<title>ime: reset view and cancel selection on input</title>
<updated>2026-10-25T10:48:42Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-05T07:52:06Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=5c8d52080750b81752595ac8f00fbacece17978b'/>
<id>urn:sha1:5c8d52080750b81752595ac8f00fbacece17978b</id>
<content type='text'>
The text input area should be visible when using input methods.

This also keeps the behavior consistent with the wl_keyboard.key
event handler.
</content>
</entry>
<entry>
<title>selection: fix row comparison in selection update</title>
<updated>2026-10-25T10:47:06Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-10T23:29:39Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=0e7e943062cc002b07d107f408eea7208f2d836c'/>
<id>urn:sha1:0e7e943062cc002b07d107f408eea7208f2d836c</id>
<content type='text'>
Absolute row numbers may wrap around, thus not suitable for comparison.
This may incorrectly update the selection direction and break word-wise
selection.

Compare scrollback-relative row numbers instead.

Fixes #2483.
</content>
</entry>
<entry>
<title>changelog: fix links</title>
<updated>2026-10-25T10:44:19Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-10T23:42:51Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=5f01c3d70479252561df71ddd64908bf912a509d'/>
<id>urn:sha1:5f01c3d70479252561df71ddd64908bf912a509d</id>
<content type='text'>
</content>
</entry>
<entry>
<title>input: no need to strip trailing NULs from keymap</title>
<updated>2026-10-14T14:00:31Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-10-14T14:00:31Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=faa4b09b9fbbe0080d4db587119056e52831d37f'/>
<id>urn:sha1:faa4b09b9fbbe0080d4db587119056e52831d37f</id>
<content type='text'>
Since libxkbcommon 1.6.0, xkb_keymap_new_from_buffer() allows a
NUL-terminated string as keymap text [1]. Also the keymap text cannot
have multiple trailing NULs according to the Wayland protocol spec.

This also avoids length mismatch for munmap(), which may leak resource
when 'size % sysconf(_SC_PAGESIZE) == 1'.

Also see the original patch [2] which reported this issue.

[1]: https://github.com/xkbcommon/libxkbcommon/commit/b06aedb8ba
[2]: https://codeberg.org/dnkl/foot/pulls/2482
</content>
</entry>
<entry>
<title>grid: reflow: fix two issues when truncating wide characters</title>
<updated>2026-09-18T11:32:53Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-18T08:46:33Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=cb2771788998f6a6288e2de8273c55a8214551e0'/>
<id>urn:sha1:cb2771788998f6a6288e2de8273c55a8214551e0</id>
<content type='text'>
When resizing to a very small grid, it may be impossible to fit a wide
character (for example, a 6-cell character won't ever fit in a 5-cell
grid).

What we do is replace with wide character with a space. There were
however two issues:

1. We never set the attributes of the space cell, only the wc member
2. We breaked out of the loop, causing any tracking points associated
   with any of the SPACER cells to never be matched.

The last point is very real, since e.g. OSC-8 links cover the spacer
cells.

The first point is easy to fix; we first copy the entire cell from the
old grid, then replace wc with 0.

The second point is somewhat more complex. We remove the break, to
ensure we run all the tracking point logic. Then, we only write the
empty cell in the first iteration. Subsequent iterations do
nothing (to the new grid).

Then, to ensure tracking points associated with the spacer cells point
to a valid cell, we keep the new_col_idx value at the first (now
empty) cell. Only when we've processed the last SPACER cell do we
increment new_col_idx once. This ensures consistency (the wide
character was replaced with a single-cell character, thus new_col_idx
should only be incremented by one).
</content>
</entry>
<entry>
<title>shm: clang 23 complains max_pool_size is unused on x86 (which it is)</title>
<updated>2026-09-18T11:24:00Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-18T11:24:00Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=9f5c70ea6eed64706471ea3ca349181a77f21c66'/>
<id>urn:sha1:9f5c70ea6eed64706471ea3ca349181a77f21c66</id>
<content type='text'>
</content>
</entry>
<entry>
<title>ime: ignore enter events with a NULL surface</title>
<updated>2026-09-18T05:51:33Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-18T05:50:39Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=f3226bc5eba53ce4b0306546e99f5928198526c3'/>
<id>urn:sha1:f3226bc5eba53ce4b0306546e99f5928198526c3</id>
<content type='text'>
Closes #2467
</content>
</entry>
<entry>
<title>input: add log message when ignoring a keyboard_enter() with NULL surface</title>
<updated>2026-09-18T05:50:18Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-18T05:50:18Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=72418933884ced81355b91395c98a5fa47aaaa7e'/>
<id>urn:sha1:72418933884ced81355b91395c98a5fa47aaaa7e</id>
<content type='text'>
</content>
</entry>
<entry>
<title>selection: treat spacer pad cells as empty cells when marking cells for highlighting</title>
<updated>2026-09-10T06:37:01Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-02T12:25:49Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=2705e36f0ecf3ef50c13b41165de852f134859d6'/>
<id>urn:sha1:2705e36f0ecf3ef50c13b41165de852f134859d6</id>
<content type='text'>
</content>
</entry>
<entry>
<title>selection: don't allow the start/end-points to be inside spacer pad cells</title>
<updated>2026-09-10T06:37:01Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-02T12:23:56Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=92c61ec9b1cef3bf6d1da01db3310daff6bba522'/>
<id>urn:sha1:92c61ec9b1cef3bf6d1da01db3310daff6bba522</id>
<content type='text'>
Having the start/end-points in the pad area made the spacer cells
tracking points when reflowing the grid. This caused the spacer cells
to be kept, instead of removed.
</content>
</entry>
<entry>
<title>selection: don't include spacer pad cells when extending the selection to cover all of a multi-cell character</title>
<updated>2026-09-10T06:37:01Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-02T12:22:01Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=e6b286403f08163e914eb8ff5af06999dec52214'/>
<id>urn:sha1:e6b286403f08163e914eb8ff5af06999dec52214</id>
<content type='text'>
When the selection starts or ends in the middle of a multi-cell
character, we stretch the selection so that all of the cells belonging
to the character is selected.

When we do this, we do *not* want trailing pad cells to be included in
the selection.
</content>
</entry>
<entry>
<title>grid: reflow: Preserve cells with background color</title>
<updated>2026-09-10T06:35:29Z</updated>
<author>
<name>Celina Sophie Kalus</name>
<email>hello@celinakalus.de</email>
</author>
<published>2026-09-08T20:19:05Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=d4896e386b3757264775e3ac2e1d1abc9d13a0c2'/>
<id>urn:sha1:d4896e386b3757264775e3ac2e1d1abc9d13a0c2</id>
<content type='text'>
A cell was previously considered empty if its content is empty.
When an empty cell has a non-default background color, truncating such
cells from the end of the line when reflowing is unexpected. Instead,
treat such cells as non-empty to preserve them.

Closes #2462

Signed-off-by: Celina Sophie Kalus &lt;hello@celinakalus.de&gt;
</content>
</entry>
<entry>
<title>server: fix use-after-free during shutdown</title>
<updated>2026-09-10T06:28:00Z</updated>
<author>
<name>CismonX</name>
<email>admin@cismon.net</email>
</author>
<published>2026-09-05T22:17:29Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=96b717b7480314239736b19ae26db245385407b6'/>
<id>urn:sha1:96b717b7480314239736b19ae26db245385407b6</id>
<content type='text'>
When terminal shutdown is in progress, calling client_destroy()
results in subsequent fdm_shutdown() (registered by term_shutdown())
always called with a dangling 'struct terminal' pointer.

This may happen when footclient and the client application exit
simultaneously in a short window. See #2432 for more information.

Fixed by not calling client_destroy() when terminal shutdown is
in progress. The client will eventually be destroyed in
term_shutdown_handler() (invoked from shutdown_maybe_done()),
so there's no resource leak.
</content>
</entry>
<entry>
<title>foot.ini: `color-theme-switch-1`, `color-theme-switch-2` -&gt; `color-theme-switch-dark`, `color-theme-</title>
<updated>2026-09-06T07:56:46Z</updated>
<author>
<name>Arnaudv6</name>
<email>arnaudv6@noreply.codeberg.org</email>
</author>
<published>2026-09-06T07:56:46Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=95643e03c5996e8bc6131b50833ea7ed51121d35'/>
<id>urn:sha1:95643e03c5996e8bc6131b50833ea7ed51121d35</id>
<content type='text'>
foot.ini: `color-theme-switch-1`, `color-theme-switch-2` -&gt; `color-theme-switch-dark`, `color-theme-switch-light`
</content>
</entry>
<entry>
<title>doc: foot.ini: colors.cursor: tweak wording of the default value</title>
<updated>2026-09-05T13:00:33Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-05T13:00:33Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=180df200a35ce94cb2b38ae84d40b53c63467bd6'/>
<id>urn:sha1:180df200a35ce94cb2b38ae84d40b53c63467bd6</id>
<content type='text'>
Try to make it clearer that it is the *current* fg/bg inversed, not
the *default* fg/bg.

Related: #2455
</content>
</entry>
<entry>
<title>changelog: add new 'unreleased' section</title>
<updated>2026-09-02T11:35:07Z</updated>
<author>
<name>Daniel Eklöf</name>
<email>daniel@ekloef.se</email>
</author>
<published>2026-09-02T11:35:07Z</published>
<link rel='alternate' type='text/html' href='https://git.0x0f.dev/mirrors/foot.git/commit/?id=4697829af56b9fee24c1ef2901fe871c5b569cab'/>
<id>urn:sha1:4697829af56b9fee24c1ef2901fe871c5b569cab</id>
<content type='text'>
</content>
</entry>
</feed>
