8000 load command fails in noflo shell due to a typo in normalizePath function declaration by DavidDurman · Pull Request #9 · noflo/noflo · GitHub
[go: up one dir, main page]

Skip to content

load command fails in noflo shell due to a typo in normalizePath function declaration - #9

Merged
bergie merged 1 commit into
noflo:masterfrom
DavidDurman:master
Oct 10, 2011
Merged

bergie merged 1 commit into
noflo:masterfrom
DavidDurman:master

Conversation

@DavidDurman
Copy link
Copy Markdown

No description provided.

bergie added a commit that referenced this pull request Oct 10, 2011
load command fails in noflo shell due to a typo in normalizePath function declaration
@bergie
bergie merged commit c1c577a into noflo:master Oct 10, 2011
bergie added a commit that referenced this pull request Oct 9, 2026
SPEC §6 (work document #4 updates #9/#11) declares manifest entries as
carrying the component kind, but the codec still encoded name → sig_hash
strings and referenced an encodeManifestEntry helper that never existed.
Manifest entries are now positional [sig_hash, kind] tuples on the wire,
with the kind drawn from the shared elementary | subgraph | stub
vocabulary and rejected when not declared data. Malformed legacy wire
entries fail decoding instead of being tolerated.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 2: the process listing must carry the
implementation kind explicitly, so clients never infer stub-ness by
joining component names against their own registries — the runtime knows
isStub(), and its answer is authoritative.

- 0x44 CMD_PROCESS_LIST_REQ: [0x44], a bare request
- 0x45 CMD_PROCESS_LIST: [0x45, epoch_id, {node: [component, kind, state]}]
  — the component the node resolves to, its declared kind from the shared
  elementary | subgraph | stub vocabulary, and its execution state
  (0x01 enabled, 0x02 disabled), mirroring 0x43 CMD_PROCESS_CTRL so the
  disabled state that block sets is observable
- epoch_id correlates with the CRDT sync handshake (0x10)

The listing is introspection and requires GRAPH_READ, not LIFECYCLE_CTRL:
it observes what the control commands do. SPEC §8's 'separate open
surface' note is replaced by the command pair; AsyncAPI gets both channels
with verified wire examples.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 3: stub-raised errors must be classifiable
as unimplemented rather than indistinguishable from genuine component
failures. The classification is its own signal — 0x09 STUB_ERROR — instead
of a convention on the 0x04 ERROR payload string: the payload is the
exception string exactly as in ERROR, and the event type itself carries the
classification, with no string parsing. Additive event enum, so the
protocol version does not bump (update #7).

Runtimes may additionally record stub errors in the LXMF [errors] snapshot
field. SPEC §7 gains the event table row and prose; the AsyncAPI
flowtraceChunk enum grows the value and its wire example demonstrates a
DATA-into-stub followed by the STUB_ERROR, codec-verified.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 1, observation half: per-edge capacity as
a 0x32 flowtrace event. 0x0a EDGE_CAPACITY samples the backpressure state —
one event per edge per flush, payload [edge_id, in_flight, desired_size]:
the number of admitted in-flight packets and the live remaining capacity
(desiredSize()), nil when the edge is unbounded.

Sampling rides the subscription's flush cadence — the runtime's
physical-policy decision — and the sample's delta places it at flush time.
Backpressure is normal operation, not a fault: the event lives on the data
channel, never an error channel. The consumer is the noflo-ui
high-water-mark visual vocabulary (route color/age).

Additive event enum, so the protocol version does not bump (update #7).
SPEC §7's last observation Open item is replaced by the decided encoding;
the AsyncAPI flowtraceChunk enum grows the value and its wire example
demonstrates bounded and unbounded samples in the same flush,
codec-verified.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 1, configuration half — the last open
wire decision before freeze:

- 0x46 CMD_HWM_SET: [0x46, high_water_mark] — a non-negative integer
  (0 synchronous, n up to n admitted in-flight packets) or nil
  (unbounded). It is the global default in the backpressure hierarchy:
  edge metadata overrides it, port defaults sit between.
- Runtime configuration, not graph state: it does not travel the CRDT
  op log. No ack frame — clients observe the effective per-edge outcome
  through 0x0a EDGE_CAPACITY samples.
- Edge-level high-water marks need no new wire surface: they travel as
  edge metadata inside 0x14 payloads, and changes project to
  tombstone + insert exactly as update #8 decided — modify stays a
  wire-level encoding detail, never the semantic unit. SPEC §5's Open
  bullet is resolved accordingly.

SPEC §8 gains the command; the AsyncAPI document gets the channel with a
verified wire example.
bergie added a commit that referenced this pull request Oct 9, 2026
Phase 4 of work document #28: the 0x30/0x32 streaming telemetry surface,
fed by a real NoFlo network.

NetworkHost owns the network lifecycle: builds the NoFlo network from the
runtime's graph with an injected component loader (platform-neutral, per
work documents #6/#16), re-emits the network's EventTarget events so
protocol handlers observe a network before, after, and across network
recreations, and supports rebuild-on-discard. The engine gap it surfaces
is documented: the 2.x engine consumes the graph at connect() and does
not follow live graph changes, so CRDT operations applied while a network
runs update the wire state but not the running network until the engine
learns to follow graph deltas.

TelemetryProtocol maps host events onto the 0x32 vocabulary:
- data packets become DATA events with raw payloads; brackets become
  BEGIN_GROUP/END_GROUP with the group name
- network start/stop become LIFECYCLE START/STOP codes
- process errors become ERROR events with the exception string — except
  errors in the stub error class (default StubNotImplementedError), which
  become STUB_ERROR: the event type itself is the classification (#4
  update #9)
- payloads the wire cannot carry are frugalized to type-named
  placeholders so serialization can never take down the flush loop

Flush policy is the runtime's physical decision: per-subscription buffers
flush on timers at the effective cadence (the client's requested
interval, bounded below by a configurable floor), with explicit flushing
available for transports that prefer it. The protocol has no unsubscribe
command; subscriptions live with their link, and dropContext tears them
down on link close.

The integration test streams a real network run — a real ComponentLoader,
asComponent factory, graph with IIP — through the full pipeline to codec
round-trips. @noflo/noflo re-exports the Network class, restoring 1.x
surface parity that the host needed to type against.
bergie added a commit that referenced this pull request Oct 9, 2026
SPEC §6 (work document #4 updates #9/#11) declares manifest entries as
carrying the component kind, but the codec still encoded name → sig_hash
strings and referenced an encodeManifestEntry helper that never existed.
Manifest entries are now positional [sig_hash, kind] tuples on the wire,
with the kind drawn from the shared elementary | subgraph | stub
vocabulary and rejected when not declared data. Malformed legacy wire
entries fail decoding instead of being tolerated.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 2: the process listing must carry the
implementation kind explicitly, so clients never infer stub-ness by
joining component names against their own registries — the runtime knows
isStub(), and its answer is authoritative.

- 0x44 CMD_PROCESS_LIST_REQ: [0x44], a bare request
- 0x45 CMD_PROCESS_LIST: [0x45, epoch_id, {node: [component, kind, state]}]
  — the component the node resolves to, its declared kind from the shared
  elementary | subgraph | stub vocabulary, and its execution state
  (0x01 enabled, 0x02 disabled), mirroring 0x43 CMD_PROCESS_CTRL so the
  disabled state that block sets is observable
- epoch_id correlates with the CRDT sync handshake (0x10)

The listing is introspection and requires GRAPH_READ, not LIFECYCLE_CTRL:
it observes what the control commands do. SPEC §8's 'separate open
surface' note is replaced by the command pair; AsyncAPI gets both channels
with verified wire examples.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 3: stub-raised errors must be classifiable
as unimplemented rather than indistinguishable from genuine component
failures. The classification is its own signal — 0x09 STUB_ERROR — instead
of a convention on the 0x04 ERROR payload string: the payload is the
exception string exactly as in ERROR, and the event type itself carries the
classification, with no string parsing. Additive event enum, so the
protocol version does not bump (update #7).

Runtimes may additionally record stub errors in the LXMF [errors] snapshot
field. SPEC §7 gains the event table row and prose; the AsyncAPI
flowtraceChunk enum grows the value and its wire example demonstrates a
DATA-into-stub followed by the STUB_ERROR, codec-verified.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 1, observation half: per-edge capacity as
a 0x32 flowtrace event. 0x0a EDGE_CAPACITY samples the backpressure state —
one event per edge per flush, payload [edge_id, in_flight, desired_size]:
the number of admitted in-flight packets and the live remaining capacity
(desiredSize()), nil when the edge is unbounded.

Sampling rides the subscription's flush cadence — the runtime's
physical-policy decision — and the sample's delta places it at flush time.
Backpressure is normal operation, not a fault: the event lives on the data
channel, never an error channel. The consumer is the noflo-ui
high-water-mark visual vocabulary (route color/age).

Additive event enum, so the protocol version does not bump (update #7).
SPEC §7's last observation Open item is replaced by the decided encoding;
the AsyncAPI flowtraceChunk enum grows the value and its wire example
demonstrates bounded and unbounded samples in the same flush,
codec-verified.
bergie added a commit that referenced this pull request Oct 9, 2026
Work document #4 update #9 item 1, configuration half — the last open
wire decision before freeze:

- 0x46 CMD_HWM_SET: [0x46, high_water_mark] — a non-negative integer
  (0 synchronous, n up to n admitted in-flight packets) or nil
  (unbounded). It is the global default in the backpressure hierarchy:
  edge metadata overrides it, port defaults sit between.
- Runtime configuration, not graph state: it does not travel the CRDT
  op log. No ack frame — clients observe the effective per-edge outcome
  through 0x0a EDGE_CAPACITY samples.
- Edge-level high-water marks need no new wire surface: they travel as
  edge metadata inside 0x14 payloads, and changes project to
  tombstone + insert exactly as update #8 decided — modify stays a
  wire-level encoding detail, never the semantic unit. SPEC §5's Open
  bullet is resolved accordingly.

SPEC §8 gains the command; the AsyncAPI document gets the channel with a
verified wire example.
bergie added a commit that referenced this pull request Oct 9, 2026
Phase 4 of work document #28: the 0x30/0x32 streaming telemetry surface,
fed by a real NoFlo network.

NetworkHost owns the network lifecycle: builds the NoFlo network from the
runtime's graph with an injected component loader (platform-neutral, per
work documents #6/#16), re-emits the network's EventTarget events so
protocol handlers observe a network before, after, and across network
recreations, and supports rebuild-on-discard. The engine gap it surfaces
is documented: the 2.x engine consumes the graph at connect() and does
not follow live graph changes, so CRDT operations applied while a network
runs update the wire state but not the running network until the engine
learns to follow graph deltas.

TelemetryProtocol maps host events onto the 0x32 vocabulary:
- data packets become DATA events with raw payloads; brackets become
  BEGIN_GROUP/END_GROUP with the group name
- network start/stop become LIFECYCLE START/STOP codes
- process errors become ERROR events with the exception string — except
  errors in the stub error class (default StubNotImplementedError), which
  become STUB_ERROR: the event type itself is the classification (#4
  update #9)
- payloads the wire cannot carry are frugalized to type-named
  placeholders so serialization can never take down the flush loop

Flush policy is the runtime's physical decision: per-subscription buffers
flush on timers at the effective cadence (the client's requested
interval, bounded below by a configurable floor), with explicit flushing
available for transports that prefer it. The protocol has no unsubscribe
command; subscriptions live with their link, and dropContext tears them
down on link close.

The integration test streams a real network run — a real ComponentLoader,
asComponent factory, graph with IIP — through the full pipeline to codec
round-trips. @noflo/noflo re-exports the Network class, restoring 1.x
surface parity that the host needed to type against.
AsherBond pushed a commit to AsherBond/noflo that referenced this pull request Oct 12, 2026
add ability to start the runtime using a default graph
AsherBond pushed a commit to AsherBond/noflo that referenced this pull request Oct 12, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

0