API Reference
Channels
A two-way message link between a Python process and a helper process that it spawns.
- Channel
-
A shared-memory channel handle (process-private).
Task pools
A set of worker processes that divide submitted tasks among themselves.
- Pool
-
A shared-memory work-stealing task pool handle (process-private).
- Task
-
A task handle from Pool.submit(). It is collected exactly once, and an uncollected handle releases its slot when garbage collected.
- Map
-
The handle for one submitted map run.
- PreparedMap
-
A reusable map staged in shared memory.
- PoolExecutor
-
A
concurrent.futures.Executorbacked by a pymizu pool.
Cross-language interop
R peers and workers (the mizu R package), neutral task specifications, and the data.frame’s Python home.
- call
-
A task specification for a pool of another language’s workers.
- launcher()
-
Return a launcher spawning Python processes running pymizu.
- r_launcher()
-
Return a launcher spawning R processes running the
mizupackage. - Frame
-
A data.frame’s Python home: named columns with a row count.
- DimArray
-
An
ndarraycarrying R labels.
Outcomes and errors
A full ring, a timeout, a close, or a dead peer returns as a sentinel singleton — FULL / TIMEOUT / CLOSED / PEER_GONE — tested by identity, never raised. Real failures raise exceptions.
- Sentinel
-
A terminal-state sentinel returned by pymizu verbs. Test by identity against pymizu.FULL / TIMEOUT / CLOSED / PEER_GONE.
- is_sentinel()
-
is_sentinel(x) -> bool
- is_remote_error()
-
Test whether a received channel value is a remote error.
- MizuError
-
Base class for all pymizu errors.
- ShmError
-
A shared-memory region operation failed. Also an OSError: the failure is at the OS layer. Carries a ‘bytes’ attribute: the requested size of a region that was not created, None when a region was not opened.
- StartupError
-
A channel peer or pool worker failed to attach within the startup timeout.
- SubmitTimeoutError
-
Pool.submit() timed out waiting for injection-ring space. Also a TimeoutError.
- SlotsExhaustedError
-
Pool.submit() found no free result slot: too many outstanding (uncollected) tasks.
- StoppedError
-
The pool is stopped; no further submission is possible.
- CancelledError
-
The task was cancelled before it ran.
- TaskError
-
The task callable raised. Carries ‘remote_type’ and ‘detail’ attributes describing the worker-side exception.
- WorkerDiedError
-
The executing worker died mid-task. Carries ‘slot’ and ‘pid’ attributes identifying the worker.
- TaskLostError
-
The task’s payload region vanished or could not be read before any worker ran it. No worker died.
- DeclinedError
-
A send on a foreign-language channel of a value outside the portable interchange subset. Carries ‘path’ and ‘reason’ attributes (‘reason’ includes the one-line rewrite guidance for the value).
- ForeignPayloadError
-
A received payload in a language-private or unrecognized format: the peer sent a value outside the portable interchange subset. The message is consumed, so the ring is never blocked, and under recv_async() only the Future waiting on that message rejects.
- ConcurrentUseError
-
A verb ran concurrently with another verb in the same handle role: each handle role is confined to one thread at a time (libmizu ‘Handle threading’).
Housekeeping
The process-wide default pool, the current worker’s own pool handle, and cleaning up shared memory left by dead processes.
- default_pool()
-
The process-wide default pool, or None when none is set.
- set_default_pool()
-
Set the process-wide default pool; None clears it.
- using_pool()
-
Use
poolas the default for the with block, then restore. - current_pool()
-
The pool handle of the worker running the task.
- current_rng()
-
Return the current element’s numpy Generator inside a seeded map.
- prune()
-
Remove orphaned shared memory regions.