Requests on a registered stream skip authentication, so restrict that to
genuine continuations of the authorized request that opened the stream:
- A continuation may differ from the opening request only in its data,
so it can no longer, for example, supply a different cmdline to run.
- Streams are bound to the socket peer or window that opened them, and
only requests from that origin can continue or cancel them.
- Streams are aborted, discarding any buffered data, when the peer
disconnects, the window is closed or the request is cancelled, so
abandoned streams no longer stay authorized.
The hard-coded 0x5402/0x5403/0x5404 are the x86 (generic) values. On
powerpc TCSETS, TCSETSW and TCSETSF are 0x802c7414..6, so Tcsetattr
failed with ENOTTY and every kitten that sets terminal modes exited
at once: on ppc64le the paste confirmation (kitten ask) only flickered
and the paste was dropped. x/sys/unix defines the right value per
architecture, as already used for TCGETS in Tcgetattr and for
TIOCSETA* in tty_bsd.go. Drop the rest of the constant block too:
copied from pkg/term, unused, and also x86-only.
Verified on ppc64le: kitten ask under script(1) failed with
"inappropriate ioctl for device" (strace: ioctl 0x5402 = -ENOTTY);
with this change the TCSETS ioctl succeeds and the prompt runs.
Cancellations of async requests are allowed without authentication, and
requests on a registered stream are allowed without authentication too.
handle_cmd() registered the stream before returning for a cancellation, so
an unauthenticated cancellation could register a stream id that a later
request then used to skip authentication. Register the stream only after
cancellations have returned.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Keep edit_loop free of side effects: it now reports whether the editor
was started and edit_and_upload decides whether to keep the local copy.
Print the recovery path verbatim instead of Go-quoted.
The child was reaped by the first waitpid() but since WIFEXITED was false the
loop continued and the next waitpid() raised ChildProcessError, hiding the
actual reason for the failure. Now report the signal and screen contents.
Strip the optional leading = from values attached to short options, matching
rg semantics (-e=foo means the pattern foo). Replace the rg execution based
test with unit tests of the argument rewriting.
Store per color scheme auto theme state in a typed dataclass instead of
string built attribute names. Track which theme files were seen during
the directory scan rather than checking for existence afterwards, so a
file deleted mid-scan cannot invalidate unscanned themes. Reset
applied_theme when the applied theme file is deleted.
Always re-read theme files on config reload and apply the theme for the
current system color scheme. Previously, once the applied theme was
invalidated, reloads stopped refreshing theme files, so re-creating the
file had no effect until the next system color scheme change. This also
means theme files created after startup are now applied on reload.
Move the D-Bus service Exec line generation into a helper with a fast unit
test instead of copying the test binary and re-executing it. Refuse paths
containing a carriage return, since D-Bus decodes \r as a newline and
rejects .service files containing a raw carriage return.
enable_portal checked eerr instead of err after filepath.Abs, so a failing
os.Executable always aborted with a nil error and the Which fallback was
never used, while an Abs error was ignored. Check err, and report the
os.Executable error if kitten is not in PATH either.
Track move source identities in drop_status rather than parsed_uri, and
also compare birth times where available, so that a replacement reusing
the inode number of a deleted source is not removed. Clarify the
changelog entry.
libbase64's stream decoder fails every call made after it has seen
padding, even an empty one, so the empty message that terminates a
stream was rejected. Also size the decode buffers for pre-sent and
drag item data correctly, so that very short padded chunks such as
QQ== are not rejected.
Fixes#10777
When the color scheme is queried at startup, wait for the reply to the
pending ReadAll request instead of making a second blocking ReadOne call,
only falling back to ReadOne if ReadAll failed or did not reply in time.
Session bus dispatching is now guarded against re-entrancy so this wait is
safe even when called from inside a DBus callback.
Coalesce titlebar redraws: settings changes only mark the titlebar dirty,
redrawing once after the initial ReadAll reply and once per burst of
SettingChanged signals via a short one-shot timer.
Fix the legacy Read color-scheme handler checking a stale type for the
nested variant and reading the value without verifying it is a uint32.
wl_shm buffers must have non-zero size and at scales below 1/24 the
shadow margin rounds to zero, which led to out of bounds reads when
compositing the shadow into the titlebar corners.
Previously, every time the compositor released a decoration buffer it was
destroyed and all decoration buffers, for all nine surfaces, were
recreated on the next update. In addition, the titlebar could be
rendered into a buffer the compositor had not yet released. Now buffers
are marked busy while attached and kept when released. The titlebar is
never rendered into a busy buffer, instead the render is done when the
buffer is released. Busy buffers are still not destroyed when freeing
buffers, to avoid flicker, they are destroyed when released.