2461 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
cat	c538df7daa	internal/pipewire: expose connection props ... Unused in hakurei but could be useful when the package is moved out of internal. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-09 06:51:12 +09:00
cat	44e5aa1a36	internal/pipewire: include remaining size in recvmsg wrapper ... This otherwise truncates the received data by len(remaining) bytes. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-09 06:36:46 +09:00
cat	cf0e7d8c27	internal/pipewire: reset per-roundtrip state once per call ... This was left in consume when relocating per-roundtrip code out of the per-receive consume method. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-09 06:28:33 +09:00
cat	130add21e5	internal/pipewire: increment remote sequence after establishing bounds ... This avoids incrementing it twice proceeding from a partial message. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-09 06:21:41 +09:00
cat	5ec4045e24	internal/pipewire: do not clobber error parsing SCMs ... The error is handled later, clobbering it here breaks error handling when SCMs are present. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-09 06:02:15 +09:00
cat	be2075f169	Revert "internal/pipewire: work around remote sequence quirk" ... This reverts commit 564db6863b.	2025-12-09 05:25:41 +09:00
cat	e9fb1d7be5	container/initdaemon: copy wstatus from wait4 loop ... Due to the special nature of the init process, direct use of wait outside the wait4 loop is racy. This change copies the wstatus from wait4 loop state instead. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:58:42 +09:00
cat	dafe9f8efc	container: spin instead of block on wait4 ECHILD ... Blocking prevents further wait4 processing causing ops to never receive their signals. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:56:13 +09:00
cat	96dd7abd80	container: improve error message fallback ... This now falls back to message.Error if no other concrete type is matched. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:45:54 +09:00
cat	d5fb179012	cmd/hakurei: exec instead of fork/exec from shell ... There is no reason to keep the shell process around. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:29:41 +09:00
cat	462863e290	container: friendlier error message for op timing out ... This includes the string for the failing op which helps with troubleshooting. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:19:03 +09:00
cat	2786611b88	test/interactive: add app with bad daemon ... This is useful for testing daemon error handling behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 22:12:07 +09:00
cat	791a1dfa55	container: make wait4 loop available to ops ... Due to the special nature of the init process, regular wait calls are unavailable. This change provides infrastructure to access wait4 loop state from Op. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 21:43:49 +09:00
cat	564db6863b	internal/pipewire: work around remote sequence quirk ... Remote sequence sometimes start with a non-zero value, and keeps the same value for a while before going back to zero. Conditions for reproducing this behaviour is not yet known. This change works around this behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 20:10:01 +09:00
cat	87781c7658	treewide: include PipeWire op and enforce PulseAudio check ... This fully replaces PulseAudio with PipeWire and enforces the PulseAudio check and error message. The pipewire-pulse daemon is handled in the NixOS module. Closes #26. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 08:53:04 +09:00
cat	0c38fb7b6a	hst: expose daemon as fs entry ... This is slightly counterintuitive, but it turned out well under this framework since the daemon backs its target file. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 07:38:47 +09:00
cat	357cfcddee	container: start daemons within container ... This is useful for daemons internal to the container. The only current use case is pipewire-pulse. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 07:21:04 +09:00
cat	6bf245cf1b	container: pass context as setup state ... This is useful currently for daemon Op, but could be used for many other things. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 06:06:19 +09:00
cat	c8eeb4a4d1	internal/outcome: integrate pipewire server ... This is very simple and takes almost no inputs. This is not yet hooked up to anything to prevent breaking any existing behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 05:03:16 +09:00
cat	5785714b64	container: call op method right before initial process ... This is at a point considered to be already "within" the container. Daemons internal to the container can be started here. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 04:57:24 +09:00
cat	422efcf258	hst: check for insecure PulseAudio enablement ... This is currently still a noop, but required for #26. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 03:13:02 +09:00
cat	104eeecf65	cmd/hakurei: add pipewire flag ... This is for "run" command, formerly permissive defaults behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-08 02:39:55 +09:00
cat	bf856f06e5	internal/pipewire: constant for PIPEWIRE_REMOTE ... This is not defined anywhere in upstream PipeWire, no idea why. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 23:35:29 +09:00
cat	1931b54600	hst: add pipewire flag ... These are for #26. None of them are implemented yet. This fixes up test cases for the change to happen. Existing source code and JSON configuration continue to have the same effect. Existing flags get its EPulse bit replaced by EPipeWire. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 22:34:40 +09:00
cat	093e30c788	internal/system: integrate PipeWire SecurityContext ... Tests for this Op happens to be the best out of everything due to the robust infrastructure offered by internal/pipewire. This is now ready to use in internal/outcome for implementing #26. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 17:39:34 +09:00
cat	1b17ccda91	internal/system: optional op check parallelism ... The PipeWire Op check cannot be made parallel due to the OS interaction. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 17:31:10 +09:00
cat	7c6fc1128b	internal/pipewire: set finalizer on scc ... This prevents leaking the socket and pipe fds. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 17:29:22 +09:00
cat	8cdd659239	internal/pipewire: seq access method for consume ... This improves readability as the offset is not immediately obvious. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 17:07:30 +09:00
cat	15c2839a09	internal/pipewire: respond to Core::Ping ... There is currently no known message that will get the PipeWire server to emit this event. It should be handled regardless. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 14:55:56 +09:00
cat	b9b9705b52	internal/pipewire: specify opcode and file count with message ... This adds checking of FileCount while writing a message. Message encoding is relocated to an exported method to be used externally, probably for test stubbing. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 13:54:11 +09:00
cat	246e04214a	internal/system: pass syscall error message ... This makes wrapped syscall errors produce a cleaner error message. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 02:08:05 +09:00
cat	503bfc6468	internal/system: port connect by name ... This behaviour is a bit messy and checks what appears to be a windows-specific environment variable for some reason. Keeping everything intact regardless to match upstream behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 02:00:19 +09:00
cat	d837628b4c	internal/system: remove ineffectual join reverting wayland ... Removing the pathname socket used to be handled separately, now it is done during the Close call. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-07 00:43:50 +09:00
cat	3cb58b4b72	internal/pipewire: high level SecurityContext helper ... This sets up close pipe and socket internally, and exposes the resulting pathname socket and close_fd cleanup as an io.Closer. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 21:16:27 +09:00
cat	bb1fc4c7bc	internal/pipewire: check pending ids after done ... This is not guaranteed to have completed after a roundtrip. This is leftover from when Roundtrip also sent and waited for sync. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 21:09:55 +09:00
cat	f44923da29	internal/pipewire: post-sync cleanup functions ... This makes it easier to handle resources who only needs to stay alive before the next sync. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 21:03:21 +09:00
cat	5e7861bb00	internal/pipewire: handle dangling files in roundtrip ... This should not be handled on every receive as it could cause valid (though impossible in current upstream implementation) messages to be rejected and raise a protocol error. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 19:21:57 +09:00
cat	7cb3308a53	internal/pipewire: store proxy errors in context ... This change fixes handling of non-fatal errors during a roundtrip as there can be multiple receive calls per roundtrip. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 19:13:46 +09:00
cat	490093a659	internal/pipewire: set errno on an empty message ... This matches upstream behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 16:50:49 +09:00
cat	2b22efcdf1	internal/pipewire: rename context consume method ... This name is more correct since it does not roundtrip, but receives messages. This is also more consistent with the method on event proxies. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 16:44:39 +09:00
cat	8a2f9edcf9	internal/pipewire: use sendmsg/recvmsg directly ... The PipeWire protocol does not work with Go abstractions. This change makes relevant methods call sendmsg/recvmsg directly. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-06 02:12:47 +09:00
cat	0d3f332d45	internal/pipewire: do not send ancillary msg without files ... This is unnecessary and does not match upstream behaviour. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-05 04:20:04 +09:00
cat	d5509cc6e5	internal/pipewire: constants from pipewire/keys.h ... These are PipeWire spa_dict keys. Interestingly many keys in the sample are undefined and appear as magic strings in upstream source code. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-05 03:27:30 +09:00
cat	0d3ae6cb23	internal/pipewire: improve protocol error messages ... These are mostly small formatting changes, with the biggest change being to UnexpectedEOFError where its kind is now described as part of the error type. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-04 03:15:58 +09:00
cat	69b1131d66	internal/pipewire: use type name in error strings ... This provides more useful messages for protocol errors. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-04 02:29:27 +09:00
cat	2c0b92771a	internal/pipewire: relocate constants ... This should make things easier to navigate, and possible to fully automatically generate the constants in the future. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-03 03:52:17 +09:00
cat	054c91879f	internal/pipewire: finalizers for dangling files ... This makes their handling much less error-prone. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-03 03:36:02 +09:00
cat	c34439fc5f	internal/pipewire: collect non-protocol errors ... These errors are recoverable and should not terminate event handling. Only terminate event handling for protocol errors or inconsistent state that makes further event handling impossible. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-03 01:35:43 +09:00
cat	32fb137bb2	internal/pipewire: fail on unacknowledged proxies ... These proxies (with special cases documented in the implementation) are only safe for use after acknowledgement from the server. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-02 21:17:46 +09:00
cat	e7a665e043	internal/pipewire: handle Core::Error ... This event is not encountered in the pw-container sample, but already has existing sample from an excerpt. Signed-off-by: Ophestra <cat@gensokyo.uk>	2025-12-02 06:16:18 +09:00