Code review finding: build.initial_plan was left in _WAITING with zero
_ACTIONS_FOR entries but a handler still gated on find_resources. Since a
waiting state settles the runner's advance loop (is_waiting(next_state) ->
settled), continue_adjusting from build.review landed a session there with
no button the FE could render and no action the handler would accept --
the loop-back could never advance.
Reclassify build.initial_plan as working (state.py _WORKING, not _WAITING):
it is now a genuine pass-through step, matching build.test_and_repair /
build.publish / build.governance_feedback. handle_initial_plan drops its
find_resources gate and runs resource discovery unconditionally on entry;
the runner drives it with the action already consumed and falls straight
through to build.resource_recommendation within the same advance() call.
Removed the now-stale BUILD_INITIAL_PLAN entry from service.py's
_BACKEND_ACTIONS_FOR (no action to gate on any more) and the compensating
exemption it required in test_every_backend_waiting_action_has_a_ui_path
(the invariant now holds with no special case).
Added a state-classification test (test_build_initial_plan_is_working_not_
waiting) and an end-to-end regression driving the Runner from build.review
through continue_adjusting with no find_resources action ever sent
(test_continue_adjusting_reaches_resources_without_find_resources_action),
which is the scenario this fix exists to unwedge. Fixed the two existing
idempotency flow tests that asserted the old (broken) two-step loop-back.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
submit_requirements now tail-calls a shared _discover_and_offer_resources
helper straight into build.resource_recommendation, instead of stopping at
build.initial_plan to show a plan card the user must click through with no
real decision to make. The resource-discovery body (progress reporting,
discover_resources, ResourceSelectCard, empty-options notice, assistant
turn) is extracted verbatim into that helper so the straight-through path
and the continue_adjusting/retry_after_revert loop-back (which still
re-enters at build.initial_plan and still requires find_resources) share
one implementation instead of drifting apart.
build.initial_plan survives as the loop-back target but now offers no UI
action (_ACTIONS_FOR entry removed); its handler still accepts
find_resources at the backend for that loop-back.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The resource-recommendation step made the user pick a "conflict
policy" that never decided anything. Remove it end to end: the
ConflictPolicyOption card option, ResourceSelectCard's
conflict_policy_options field, the third bind_resources() parameter
across the port/placeholder/LLM-agent implementations, and the dead
conflict_policy field on the (unused) ResourcesPayload contract type.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
discover_resources initialised its selection to the entire tenant inventory and
only replaced it when the model returned a non-empty pick, so three paths -- no
model, a model that picked nothing, a failed call -- recommended everything. On
a workspace with 57 resources that is not merely noisy: the client pre-checks
every recommendation and submits them all by default, and bind_resources then
appends all 57 labels to the last plan item, which is the text fed to the graph
generator.
Recommending nothing is the honest answer when there is no signal. The card is
still emitted so the confirm gate keeps its active interaction and a later pass
cannot inherit a stale one, and a notice says why rather than leaving a blank
card. The conflict-policy control no longer renders over an empty list.
The test that should have caught this used a one-resource catalog, where "all"
and "none" are indistinguishable; it now uses 57.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The testdata gate is the one place a human types for tens of seconds, and it
opens right after Builder has written the canvas -- long enough for the editor's
own autosave to move the app revision underneath them. Both revision gates then
refused the submit, and because a conflict leaves the session with only
check_recovery as an action, active_interaction drops to None and the client
freezes every input: the form stayed on screen but could no longer be typed into.
A testdata submit carries run inputs, not graph edits, so neither gate applies
to it. Exempting only the draft-changed check would not have been enough --
the stale-app-revision check above it fires first, on the base_app_revision the
client has held since the form was issued.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Spec N4 has the build-complete card headline read "<App name> is ready", with
the annotation 名字只在回复句里念一次;记录卡上不出现 -- said once, in that
sentence, and never repeated on the completion receipt. We said it nowhere: the
card was titled "Build plan" and the engine had no app name to say.
DifyBuilderContext now carries app_name, seeded at session creation through a
get_app_name_fn seam alongside the existing revision/authorize ones. Env.rename_app
returns the name the database actually took rather than None, so the refine
records what landed instead of what it proposed -- the service normalizes after
the engine hands off, and the two can differ.
The headline is interpolated, so it gets a strings.py template or it would
survive localization untranslated. That pattern matches almost anything ending
in "is ready", so it sits last and every more specific template gets first
refusal.
The receipt already omitted the name; there is now a test pinning that.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two App Builder spec changes that share the Builder session contract.
Naming (spec N1-N3): an app created from the homepage prompt is no longer named
by the user. core/dify_builder/naming.py cuts the prompt's opening clause at
create time -- instant and infallible, since it runs inside POST /apps -- and
CreateAppPayload.name becomes optional alongside a new prompt field. The Builder
then refines that name once during goal analysis, where a model and the detected
reply language are already in hand, so the judgement calls that are
English-shaped in pure Python (dropping an imperative lead-in, spotting a
contentless prompt) work in any language. Plan approval freezes the name.
Card options: SessionView gains a Decision carrying the gate's choice as options
inside the card plus a fixed confirm/cancel pair, replacing a button row whose
shape changed at every state. Option ids are the old action ids untouched, so
resolve_action_kind and the per-state legality check are unchanged, and
SessionView.actions keeps being populated -- the deployed web client goes on
working until it reads decision instead.
The regenerated console OpenAPI doc also picks up ErrorCard.diagnostics, missed
when cc9ca8b1dd landed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
api pods lose their logs on restart, so a failed build's real cause
("Workflow generator: structural validation failed: ...") could only be read
from a pod log that may already be gone. Session 01a08fd0 hit exactly that.
Carry the generator's own diagnostics back with the result instead:
BuildNodesResult and ErrorCard gain a ``diagnostics`` list, and build_nodes
records one entry per failed generation attempt -- initial AND the corrective
retry, so an error that changes between attempts stays visible -- plus the
all-rejected and exception paths.
Each entry holds a server-side UTC ``at`` (so the surviving pod-log window can
be searched around it), ``source``, ``attempt``, the generator's ``message``
(the same text the server logs), its ``codes``, and the full structured
``errors`` including ``node_id``. That is strictly more than the log line
carries: for an UNRESOLVED_REFERENCE the export now names the offending node.
The list rides in the ErrorCard's item payload, which is what the streamed
conversation item -- and therefore the exported debug log -- preserves. It is
not rendered in the UI. Retries that eventually succeed keep their breadcrumbs.
Deliberately not mirroring the "structural validation failed" prefix: build.py
cannot tell structural failures from the rest without duplicating the
generator's code taxonomy, which would go stale.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
When build-generate-graph failed at plan-approval, the error card always
showed one hardcoded generic message regardless of the actual cause
(UNRESOLVED_REFERENCE, non-object JSON, credit_balance_exhausted, ...).
Root cause: build_nodes() returned only list[MutationIntent]; on failure it
returned [], so the reason was logged internally but never reached the
handler, which had nothing to show but a fallback string.
Thread the reason out: build_nodes now returns BuildNodesResult{intents,
error}. _generation_error_text() extracts the best reason (top-level error
-> joined errors[].detail/message/code -> generic fallback), and every
failure path populates .error (generator error, all-intents-rejected, and
the except path via str(exc) so provider errors surface verbatim).
handle_plan_approval builds the ErrorCard body from build_result.error when
present, keeping the generic sentence only as a genuine fallback.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Build wrote the Builder's session/cognition model (e.g. deepseek) into every
generated node's model block, ignoring the model the user picked at resource
confirmation (e.g. gpt). _ground always used the generation model, and the
selected model resource only reached the plan as label text (bind_resources) —
so a workflow got a model with no runtime credentials -> ModelNotExistError ->
instant test failure.
Thread the selected resource_ids from resource_selection through
agent.build_nodes into build.build_nodes, resolve the selected MODEL resource
(id is `{provider}/{name}`) via _selected_workflow_model, and ground node model
blocks to it — falling back to the session/generator model when the user
selected no model. The generator still runs on the session model (cognition);
only the built workflow's runtime model follows the user's choice.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Harden the Localizer's batch translation against models that format JSON
differently (the Builder model is user-changeable):
- correlate by numeric INDEX instead of echoing source strings as keys, so a
model that normalizes/truncates long keys no longer misses every lookup
- accept the reply as an index-keyed object OR a positional array, optionally
wrapped one level (_translations_by_index)
- cache only real translations; a failed/partial reply is left UNCACHED so the
next turn retries (self-heals) instead of sticking on the English original
TDD: added parser shape-variation tests + a no-cache-failure/self-heal test;
the fake now speaks the index protocol.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Live e2e revealed M2 never translated the static catalog strings: the model
returns the batch-translate mapping wrapped under a "strings" key (echoing the
request envelope) — {"strings": {src: translation}} — so _fill_cache's
top-level table.get(src) missed every source and cached the English original.
Unwrap one level when the sources aren't top-level keys, and tighten the prompt
to ask for a flat object. The unit-test fake now mirrors the real wrapped shape
so this can't regress (a flat fake hid the bug).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Follow-up to the reply-language feature. The update_model action commits its
"Model changed to <name>" notice directly via compare_and_advance, bypassing
Runner._commit and thus the Env.localize_items hook, so that one notice stayed
English. Localize it in the service (when the session has a detected
reply_language) with a "Model changed to {name}" catalog template that
translates the frame and re-inserts the model name verbatim.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Follow-up to the reply-language feature. `_model_or_none` was private to
LlmBuilderAgent, so the advance-task wiring reached it via a fragile
`getattr(agent, "_model_or_none", lambda: None)`. Make it part of the
DifyBuilderAgent protocol (typed `object | None` to keep the pure port free of
any model-runtime import) and implement it on every conforming agent
(LlmBuilderAgent renames the private method public; PlaceholderAgent and the
test doubles return None). The Localizer wiring now uses `agent.model_or_none`
directly. Also drops the useless-return ruff nit in the wiring test.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Final-review fix wave for the "reply in the user's input language" feature:
- serde.py: context_from_dict omitted reply_language from the explicit field
allowlist, so it round-tripped to "" on every read despite context_to_dict
(asdict) serializing it -- the localization feature silently no-op'd after
the first persisted write. Added it to the constructor call; TDD'd with a
failing-then-passing round-trip test in test_serde.py.
- localize.py: Localizer._cache was per-instance, but a fresh Localizer is
built per advance-task invocation, so the static string catalog was
re-translated every turn instead of once per language. Moved the cache to a
module-level _TRANSLATION_CACHE dict (safe to share: keys are
tenant/session-independent catalog-string+language pairs); added an autouse
fixture in test_localize.py to clear it between tests for isolation.
- localize.py: detect_language and _fill_cache called self._model_provider()
outside their try/except, violating the Env callback contract that these
must never raise. Moved both calls inside their try/except so a raising
provider degrades to the existing safe fallback instead of propagating into
advance/_commit.
- strings.py: documented that Template.translate_fields is advisory only --
the Localizer re-inserts all captured groups verbatim and does not consult
it. No behavior change.
uv run --project api pytest tests/unit_tests/core/dify_builder tests/unit_tests/services/dify_builder -q
-> 722 passed.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The build/edit test handlers wrapped run_draft in a bare, non-logging
`except Exception` that turned any launch-time failure into a blank failed
run (dify_run_id="", per_node=[], no error). Diagnose then had nothing to
act on and the repair loop ran blind — the prod symptom where a stale test
input (ValueError "<var> is required in input form") produced an
undiagnosable await_repair loop.
Now the launch error is logged, captured, and persisted so diagnose/routing
can act on it:
- Run.error field + dify_builder_runs.error column (migration a9cc8d552db4)
- handlers_build/edit: logger.exception + launch_error_text(exc) -> Run.error
- is_input_failure() also inspects Run.error (+ "in input form" signal), so a
stale-input launch failure routes to the testdata gate, not blind repair
- diagnose/_degraded_diagnosis surface the launch error when no node executed
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Render change sets directly in plan-style cards with affected node names,
IDs, and change details. Capture node metadata for proposed and applied
changes and persist it with conversation history.
Remove the view_changes action and expand/collapse state. Update generated
contracts, translations, and regression coverage.
Validation: 112 frontend tests and 742 backend unit tests passed.
Integrates the FE team's live-progress + conversation-streaming rework
(durable progress events, repository layer, snapshot-consistent reads,
idempotent message retries, EDIT_PUBLISH->EDIT_COMPLETE) and reconciles its
interrupted-step recovery model with the reconciler already on
feat/dify-builder.
Combined recovery model:
- interrupted (crash mid working-step, lock released, nothing durably
recorded) -> Retry (recovery_continue, re-runs the handler) + Start over
- FAILED (handler raised, recorded by runner.fail) -> Restart only
- a Retry that itself raises lands in FAILED -> Restart (no wedge)
Kept from the PR: EDIT_PUBLISH working / EDIT_COMPLETE terminal, snapshot
refresh in apply_recovery_action, paused reset. Kept from the reconciler:
Retry on interrupted, state-aware access tier (EDIT_PUBLISH added to the
publish tier), apply_repair idempotency filter.
680 dify_builder unit tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A recovery_continue (Retry) re-runs the interrupted working handler, which
re-invokes the state's privileged op -- publish (RELEASE) at build.publish/
fix.publish, a draft test run (TEST_AND_RUN) at the test/verify states. Before
this, recovery_continue was classified as base EDIT, so a Retry re-ran the
privileged op below its own authorization tier without re-checking. Make
_app_access_for_action state-aware for recovery_continue so it enforces the
same tier the state's normal action requires. recovery_restart only resets +
re-enters the entry state, so it stays EDIT.
Also: an interrupted working step now always surfaces the Retry/Start-over
offer, overriding a stale drift recovery_class a user may have left set at an
earlier waiting gate (drift recovery only applies at a waiting gate).
Addresses the two findings from the final reconciliation review.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When a session crashes mid working-step (e.g. build.publish; the async advance
raised and released the lock), it was a locked dead-end: working states surface
no UI actions and the composer locked with no way forward.
- Surface a RecoveryRef offer (can_continue=Retry, can_restart=Start over,
recovery_class="") on an interrupted working-state view -- built directly so a
real drift class still wins.
- Allow recovery_continue / recovery_restart at an interrupted working state in
_internal_action_allowed. recovery_continue returns the working state
(non-settled) so the runner re-runs the stalled handler; recovery_restart
resets working fields and re-enters the entry state -- both reuse existing
recovery machinery, no new action kinds / no runner change.
Adds unit tests (offer surfaced, allowance) + regression tests (recovery_continue
re-runs handle_publish and advances; recovery_restart resets to entry state;
repeat-failure re-surfaces the offer with no wedge).
This reconciler was reviewed as a standalone plan on the pre-merge base; here it
is re-applied onto feat/dify-builder after PR #41677 merged in the reworked
service.py + _internal_action_allowed gate (which makes the allowance necessary).
The apply_repair idempotency + delete-guard commits (safe fix.apply Retry) land
alongside it.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
apply_repair's already-present filter (added for interrupted-step Retry
idempotency) compared create_node/connect intents against before_graph
alone, with no awareness of delete_node intents in the same batch. A
from-scratch build sends delete_node(placeholder_start) + create_node(same
id) in one apply_repair call to replace the canvas's default placeholder
start with the generator's own start node -- the filter dropped the
create_node as "already present" while the delete_node still ran, deleting
the start node with no re-create, so the following connect raised
ValueError: node not found. Mirror handlers_build.py's existing M2 guard:
subtract in-batch delete_node targets from the present-node/edge sets
before filtering, so a delete+recreate of the same id in one batch survives.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Drops create_node intents whose id already exists and connect intents whose
edge already exists (referenced against the original draft graph), so a
fix.apply Retry re-applying the same intents is a clean no-op instead of a
duplicate/ValueError. graph_ops validation (raise on a genuinely fresh
duplicate id) is preserved.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
progress_bus.subscribe() eagerly enters the pub/sub subscription (opening a
Redis connection + starting a listener thread) BEFORE the SSE generator
exists, so the subscribe completes before the Celery advance can publish.
But the only close() lived in the generator's finally -- which Python does
not run if the generator is never started. A streaming Response dropped
before its first frame (client aborted, or an after_request/teardown error)
therefore leaked the connection + thread until process restart.
stream_advance_frames now returns a small _ClosingFrameStream wrapper whose
close() closes the subscription directly (idempotent); the WSGI server calls
close() on the response iterable even on early disconnect, so cleanup fires
regardless of iteration. The inner generator keeps its own finally for the
iterated/disconnect paths, so this is strictly additive -- no happy-path
change. All seven call sites get it unchanged.
Adds a regression test (never-iterated stream -> close() -> subscription closed).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two paths ended a streaming advance without a terminal `state`/`error`
frame, leaving the SSE client stalled to the deadline with a stale held
version -> its next action 409s:
1. advance_session published the terminal `state` frame outside the
try/finally and unguarded. If the post-advance view projection
(get_session_view -> read_graph) or the publish itself raised after a
SUCCESSFUL advance, the exception escaped and no terminal frame was
sent. Guard it and fall back to a terminal `error` frame.
2. submit_action_stream emitted a terminal `state` frame on settle only
for `message`. `update_model` also settles synchronously while bumping
the version, so it now emits one too -- otherwise the FE keeps the old
version and 409s on its next action.
Adds regression tests for both (projection-failure -> error frame;
update_model -> state frame).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
WorkflowService.publish_workflow returns a single Workflow (-> Workflow), but
dify_port.publish still unpacked it as (workflow, retirement_candidates) -- a
stale signature. Every other caller was updated when the return type changed
(via the main merge); only the Dify Builder port was missed, so every publish
crashed with "TypeError: cannot unpack non-iterable Workflow object", surfaced
to the user as a generic "step failed" SSE frame.
Assign the single return value (retirement_candidates was unused anyway). The
existing test_dify_port publish test mocked publish_workflow as a 2-tuple, which
masked the bug -- corrected it to the real -> Workflow signature so it now guards
the regression.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A build for a workflow app whose goal reads like a chatbot produced an empty
canvas while the assistant reported "Workflow built on the canvas.": the plan
was chatbot-shaped, build_nodes fed it to the generator (mode="workflow"), the
LLM emitted a chatflow graph (answer node / no end node), the generator
rejected it as MISSING_TERMINAL -- a class it does NOT retry -- and build_nodes
swallowed the error into []. handle_plan_approval then treated [] as a
successful build: it deleted the placeholder start and advanced to
build.execution, leaving nothing on the canvas.
Two parts:
1. build_nodes now prepends an explicit workflow-topology directive (one
start + >=1 end node, no answer nodes) to the generator instruction, and
retries ONCE on an error/empty result with the specific error fed back as
a corrective nudge -- covering the terminal-node failure the shared
generator skips.
2. handle_plan_approval refuses to fake success: when build_nodes yields zero
create_node intents it surfaces an honest error and stays in
build.plan_approval (retryable) without deleting the placeholder, instead
of emptying the canvas and claiming a build. A loop-back re-approve still
returns the fixed create intents (only filtered as already-present), so
zero create intents unambiguously means a genuine generation failure.
Adds regression tests for both. Note: part 1 improves LLM reliability
(verifiable live); part 2 is the deterministic guarantee against the
false-success empty canvas.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
It was stubbed to [] in LlmBuilderAgent, canned in PlaceholderAgent, and never
called by any handler — the Build failure path reuses Fix cognition (diagnose +
propose_repair). Removed from the DifyBuilderAgent Protocol, both implementations,
and all test doubles.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clear fc.test_input_ref and fc.verify_run_id on the four re-plan/revert
escape branches in Build and Edit (handle_review's continue_adjusting
and handle_reverted's retry_after_revert) so a rebuild+retest
regenerates fresh schema-shaped mock inputs instead of reusing the
first build's stale ones -- the approve_repair retest path is left
untouched since it must fairly retest against the same failing case.
Also corrects three doc/comment mismatches (llm_agent's 12->13 method
count, an await_repair docstring describing a branch that doesn't
exist, and state.py's Build state-count banner), and adds tests for
the run_draft-raises degrade path and Build/Edit pass-path canvas
event symmetry.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>