MCPcopy Create free account

hub / github.com/PerpetualSoftware/pad / functions

Functions28,840 in github.com/PerpetualSoftware/pad

FunctionTestDashboardSuggestedNextOrphan_RanksBelowActivePlan
TestDashboardSuggestedNextOrphan_RanksBelowActivePlan pins the rank order: when both an active-plan candidate AND an orphan candidate exist at the sam
internal/server/handlers_dashboard_test.go:932
FunctionTestDashboardSuggestedNext_FiltersBlockedItems
TestDashboardSuggestedNext_FiltersBlockedItems is the regression test for BUG-990's blocked-item filter. Items with at least one active "blocks" link
internal/server/handlers_dashboard_test.go:773
FunctionTestDashboardSummary
(t *testing.T)
internal/server/handlers_dashboard_test.go:265
FunctionTestDashboard_AdminBearer_RestrictedMemberIsScoped
TestDashboard_AdminBearer_RestrictedMemberIsScoped pins BUG-1917: a platform admin who is only a restricted member (collection_access "specific") of a
internal/server/handlers_dashboard_test.go:1298
FunctionTestDashboard_NoTextDerivedBlocked
TASK-3915: "blocked" on the dashboard is the dependency graph's alone. The runner here still ASKS the retired text-derived question and its answer for
internal/server/dashboard_attention_decisions_test.go:139
FunctionTestDateBucket
(t *testing.T)
internal/store/dialect_test.go:50
FunctionTestDeadVerdictStillReaps
Controls: the other two verdicts behave exactly as before.
internal/cli/session_arm_unverifiable_bug2771_test.go:95
FunctionTestDeadlineEvalArchivedRows
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:378
FunctionTestDeadlineEvalBadRowDoesNotStopPass
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:439
FunctionTestDeadlineEvalConcurrentPassesRecordOnce
Two evaluators racing over the same rows (two ticks, or two instances on one Postgres) record each miss and send each notice once.
internal/store/schedule_deadline_eval_test.go:167
FunctionTestDeadlineEvalConversionGuard
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:321
FunctionTestDeadlineEvalDateOnlyLocalEndOfDay
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:269
FunctionTestDeadlineEvalFloatingRecipientZone
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:292
FunctionTestDeadlineEvalFractionalSecondOverride
A deadline override stored with fractional seconds still reads as an instant on both dialects, so the row is scanned in order and recorded once.
internal/store/schedule_deadline_eval_test.go:495
FunctionTestDeadlineEvalLegacyMissedStateDoesNotSuppress
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:426
FunctionTestDeadlineEvalMalformedInstantDoesNotStopPass
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:467
FunctionTestDeadlineEvalNoNotificationsAndItemEvent
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:396
FunctionTestDeadlineEvalOverrideDefersStoredDeadline
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:238
FunctionTestDeadlineEvalPagesPastNotYetDueRows
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:251
FunctionTestDeadlineEvalRecordedRowsDoNotOccupyTheCap
Rows already missed do not occupy the cap: with the cap at 1, a row past its deadline behind many recorded ones is still reached on the next pass.
internal/store/schedule_deadline_eval_test.go:148
FunctionTestDeadlineEvalRecordsAMissOnceAcrossTicks
A deadline that passed records one miss and one notice, and no later tick, for any behavior, adds another.
internal/store/schedule_deadline_eval_test.go:65
FunctionTestDeadlineEvalRenegotiationAndCanonicalEvidence
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:210
FunctionTestDeadlineEvalResolvedBeforeDeadline
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:361
FunctionTestDeadlineEvalScanCapDoesNotStarve
More overdue rows than the scan cap: every pass takes at most the cap, reports that it was cut short, and a later pass reaches the rest, so no hard de
internal/store/schedule_deadline_eval_test.go:113
FunctionTestDeadlineEvalSeedDisplacementAcrossDST
(t *testing.T)
internal/store/schedule_deadline_eval_test.go:451
FunctionTestDeadlineEvalSkipsARowPastItsDeadlineAtMigration
The first tick after the migration records nothing for a row already past its deadline: the backfilled roles are not native, so no miss or notice is d
internal/store/schedule_deadline_eval_test.go:97
FunctionTestDeadlineEvalSkipsTerminalItemsAndRecordsOnReopen
An item in a terminal status never shows overdue work: no miss and no event while it is done, and a reopened item past its stays-owed deadline records
internal/store/schedule_deadline_eval_test.go:548
FunctionTestDeadlineEvalTerminalIsPerWorkspace
Terminal is judged by each workspace's own terminal options: the same status word closes an item in one workspace and leaves it open in another.
internal/store/schedule_deadline_eval_test.go:575
FunctionTestDebounceRowUnchanged
TestDebounceRowUnchanged covers the classifier that tells the merge UPDATE's two refusal arms apart. Its answer decides between two OPPOSITE dispositi
internal/store/activities_debounce_race_test.go:260
FunctionTestDecideSkillWrite
(t *testing.T)
internal/cli/skillstamp_test.go:16
FunctionTestDecideSkillWriteLegacyCopies
(t *testing.T)
internal/cli/skillstamp_test.go:70
FunctionTestDecisionJobs_ClaimIsExclusiveUntilLeaseLapses
A claim is exclusive until its lease lapses, and then any runner may take it.
internal/store/decisions_test.go:130
FunctionTestDecisionJobs_CommentEditAndDeleteEnqueue
Comment edit and delete change the trail, so they are doors too (decision 3's rule applied). An edit that leaves the body unchanged is not.
internal/store/decisions_test.go:316
FunctionTestDecisionJobs_DebounceCoalescesRacingComments
Racing writes coalesce: the second enqueue keeps the first not_before (COALESCE(old, new)) rather than pushing it back, so a stream of comments cannot
internal/store/decisions_debounce_test.go:130
FunctionTestDecisionJobs_DebounceDefersReask
A write inside the window after an evaluation owes a job that is NOT claimable until latest evaluation + window. The control legs: the same job IS cla
internal/store/decisions_debounce_test.go:80
FunctionTestDecisionJobs_DebounceUsesLatestEvaluation
The window runs from the LATEST evaluation, not the first: an item asked two hours ago and again ten minutes ago is inside its window.
internal/store/decisions_debounce_test.go:236
FunctionTestDecisionJobs_DropPreservesANewerGeneration
Dropping a claim (retry limit, gone item) must not take a NEWER generation with it — a write that landed mid-evaluation is still owed.
internal/store/decisions_test.go:348
FunctionTestDecisionJobs_EachDoorEnqueuesAndCoalesces
Each of the three doors enqueues, and repeats coalesce into ONE row whose generation counts the writes.
internal/store/decisions_test.go:75
FunctionTestDecisionJobs_FailureBacksOff
(t *testing.T)
internal/store/decisions_test.go:201
FunctionTestDecisionJobs_FirstEvaluationNotDebounced
A first evaluation is never delayed: an item with no answers for the set is claimable at once, with not_before NULL.
internal/store/decisions_debounce_test.go:64
FunctionTestDecisionJobs_MoveEnqueuesForTargetCollection
Move is a door (TASK-3117 decision 3), keyed on the TARGET collection's sets: a move INTO a scoped collection makes an evaluation owed where the item'
internal/store/decisions_test.go:292
FunctionTestDecisionJobs_NoResolverEnqueuesNothing
The nil-provider case: no resolver installed means every door writes nothing. The doors are exercised for real, so a pass is not a door that never ran
internal/store/decisions_test.go:47
FunctionTestDecisionJobs_PerCommentSetNotDebounced
A set with no window (a per-comment set) is never debounced, beside one that is: the window is per set, not global.
internal/store/decisions_debounce_test.go:258
FunctionTestDecisionJobs_ReenqueueDuringEvaluationSurvivesCompletion
A write that lands mid-evaluation leaves the job owed: completion of the older generation releases rather than deletes.
internal/store/decisions_test.go:165
FunctionTestDecisionJobs_ResolverScopesByCollection
(t *testing.T)
internal/store/decisions_test.go:109
FunctionTestDecisionJobs_RestoreEnqueues
Restore is a door. A job owed at delete time is dropped as gone; without an enqueue on restore the item would come back with stale answers and nothing
internal/store/decisions_test.go:410
FunctionTestDecisionJobs_SupersededCompletionSetsNotBefore
A write that lands DURING an evaluation leaves the row behind (the generation rule). Its enqueue read the latest evaluation BEFORE the run's answers l
internal/store/decisions_debounce_test.go:171
FunctionTestDecisionJobs_SupersededNoCallKeepsTheOlderWindow
A superseded run that STORED nothing (its state was already answered) recomputes from the older evaluation and invents no delay (ruling ae2e0a90): not
internal/store/decisions_debounce_test.go:203
FunctionTestDecisionProviderBucketDefaults
(t *testing.T)
internal/server/handlers_playbook_match_ratelimit_test.go:39
FunctionTestDecisionSQLState
The fallback may carry a SQLSTATE, but only one of the fixed five-character shape: anything else read off the failing error is driver text (BUG-4022,
internal/store/decisions_failopen_test.go:156
FunctionTestDecisionSettings_AdminOnlyAndInputValidation
(t *testing.T)
internal/server/decision_settings_test.go:343
FunctionTestDecisionSettings_ChangeTakesEffectWithoutRestart
Takes effect without a restart, in BOTH directions, observed at the write doors — whether an item write owes a decision job — which is the half the sw
internal/server/decision_settings_test.go:159
FunctionTestDecisionSettings_CloudIsEnvOnlyAndReadOnly
Pad Cloud: env is the only source, the section is read-only, and a stored setting is ignored even when present.
internal/server/decision_settings_test.go:293
FunctionTestDecisionSettings_DisableCancelsTheInFlightPass
Disabling the provider must end a pass already in flight, not let the old runner keep sending item data for the rest of it. The job must also stay owe
internal/server/decision_settings_test.go:403
FunctionTestDecisionSettings_EnabledWithoutKeyReportsTheError
Enabled with no key saves, reports why decisions are off, and runs none.
internal/server/decision_settings_test.go:331
FunctionTestDecisionSettings_EnvOverridesAndRefusesTheWrite
Env overrides the stored setting: GET says which fields the environment sets, the effective value is the environment's, and a write to such a field is
internal/server/decision_settings_test.go:226
FunctionTestDecisionSettings_GenericSettingsPatchCannotWriteTheKey
The generic settings PATCH must not become a second, unencrypted door.
internal/server/decision_settings_test.go:143
FunctionTestDecisionSettings_KeyNeverReturnedAndEncryptedAtRest
Pin (TASK-3121): the key is stored through the PAD_ENCRYPTION_KEY path and never returned by ANY read — not this endpoint's, not the generic settings
internal/server/decision_settings_test.go:96
FunctionTestDecisionSettings_NoTickStartsAfterStop
A settings write racing Stop() must not start a tick after Stop() drained the background goroutines.
internal/server/decision_settings_test.go:457
FunctionTestDecisionSettings_PlaintextKeyRowIsNotUsed
A decision_api_key row that is not ciphertext — from a restore, a migration, or a direct write — is refused, not used as a live key.
internal/server/decision_settings_test.go:480
FunctionTestDecisionSettings_Precedence
Effective enablement follows env over admin over file, per field.
internal/server/decision_settings_test.go:260
FunctionTestDecisionSettings_TickLoopSurvivesDisableAndReenable
It might be assumed that a disabled runner makes the next tick panic and kill the loop. It does not — RunOnce is nil-safe — and this pins that: the lo
internal/server/decision_settings_test.go:523
FunctionTestDecisions_NoProviderIsEmpty
No provider: the endpoint answers an empty list, the item has no `decisions` member, and no write left a job row.
internal/server/handlers_decisions_test.go:88
FunctionTestDecisions_TickEvaluatesAndReadReportsCurrency
Through the HTTP doors end to end, with the tick as the evaluator — the binding under test is the wiring (write door -> queue -> tick -> read), not th
internal/server/handlers_decisions_test.go:108
FunctionTestDecisions_UnknownItemIs404
(t *testing.T)
internal/server/handlers_decisions_test.go:183
FunctionTestDecodeBodyPreserved
(t *testing.T)
internal/artifact/decode_test.go:70
FunctionTestDecodeCRLFTolerant
(t *testing.T)
internal/artifact/decode_test.go:87
FunctionTestDecodeFieldsJSON_AcceptsWhatUnmarshalAccepts
(t *testing.T)
internal/models/fields_json_test.go:36
FunctionTestDecodeFieldsJSON_RoundTripsNumbersExactly
BUG-3202. DecodeFieldsJSON must change only how an accepted number is represented: it accepts and refuses the same blobs json.Unmarshal does, and a de
internal/models/fields_json_test.go:11
FunctionTestDecodeJSONKeepsEmptyBodyEOF
TestDecodeJSONKeepsEmptyBodyEOF pins a contract a caller depends on: handlers_playbooks.go treats errors.Is(err, io.EOF) as "no arguments supplied" an
internal/server/decode_json_nul_test.go:206
FunctionTestDecodeJSONRefusesARepeatedMember
TestDecodeJSONRefusesARepeatedMember drives the refusal through the real router, on a member that no per-struct check guards, so the 400 can only come
internal/server/decode_json_nul_test.go:1114
FunctionTestDecodeJSONRefusesNULThroughTheHandler
TestDecodeJSONRefusesNULThroughTheHandler exercises the BINDING, not the predicate: a real request through the real router, so the test has an opinion
internal/server/decode_json_nul_test.go:149
FunctionTestDecodeJSONRefusesNestedNULThroughTheHandler
TestDecodeJSONRefusesNestedNULThroughTheHandler is the wiring leg for the nested case, on SQLite for the same reason as the single-layer one: there th
internal/server/decode_json_nul_test.go:429
FunctionTestDecodeJSONTrimsOnlyJSONWhitespace
TestDecodeJSONTrimsOnlyJSONWhitespace pins that the empty-body shortcut uses JSON's whitespace set, not Go's. bytes.TrimSpace uses unicode.IsSpace, w
internal/server/decode_json_nul_test.go:1201
FunctionTestDecodeJSONWithLimit_CustomCap
TestDecodeJSONWithLimit_CustomCap verifies callers can opt in to a larger cap (for bulk-import style endpoints).
internal/server/decode_json_test.go:54
FunctionTestDecodeJSON_AcceptsWithinLimit
TestDecodeJSON_AcceptsWithinLimit confirms the happy path still works.
internal/server/decode_json_test.go:36
FunctionTestDecodeJSON_RejectsOversizeBody
TestDecodeJSON_RejectsOversizeBody ensures the default 2 MiB cap is enforced — without http.MaxBytesReader a multi-GB POST would stream into a single
internal/server/decode_json_test.go:13
FunctionTestDecodeMissingFrontmatter
(t *testing.T)
internal/artifact/decode_test.go:40
FunctionTestDecodePayloadAcceptsBothWireForms
--- the wire form ------------------------------------------------------
internal/events/redis_idspace_test.go:62
FunctionTestDecodeUnknownKind
(t *testing.T)
internal/artifact/decode_test.go:62
FunctionTestDecodeUnsupportedVersion
(t *testing.T)
internal/artifact/decode_test.go:54
FunctionTestDecryptPlaintext
(t *testing.T)
internal/store/encryption_test.go:43
FunctionTestDedupeKeepsTheCollectionTheUserWroteIn
TestDedupeKeepsTheCollectionTheUserWroteIn is the fixture that MUST be decisive: the renamed collection holds a convention the user wrote, the acciden
internal/store/trait_dedupe_test.go:109
FunctionTestDedupeReportsAnArbitraryTieAsArbitrary
TestDedupeReportsAnArbitraryTieAsArbitrary is the bare reproduction: nobody wrote anything, so both collections hold four identical template rows and
internal/store/trait_dedupe_test.go:149
FunctionTestDedupeStripsEveryDeclarationALoserHolds
TestDedupeStripsEveryDeclarationALoserHolds covers the case a collection loses BOTH declarations, which the single-declaration fixtures above cannot r
internal/store/trait_dedupe_test.go:239
FunctionTestDeepNestFrameShallow
The generator is well-formed: a shallow nest materializes to what the editor makes of it.
internal/materialize/adversarial_test.go:75
FunctionTestDefaultExportPath_BUG3185
BUG-3185: the CLI writes the export file itself, so it applies the device rule to the name it will write, whether that name came from the server (an o
cmd/pad/item_export_device_test.go:8
FunctionTestDefaultNotificationSettings
(t *testing.T)
internal/store/notification_settings_test.go:100
FunctionTestDefaultPreservesHistoricalNames
TestDefaultPreservesHistoricalNames is the compatibility promise made executable: an upgrade with no namespace configured must address the SAME keys a
internal/redisns/redisns_test.go:13
FunctionTestDefaultTracksWork
(t *testing.T)
internal/models/tracks_work_test.go:26
FunctionTestDeferredAppendSeesTriggerInstalledByTheTransactionItWaitedOn
TestDeferredAppendSeesTriggerInstalledByTheTransactionItWaitedOn is proving test (a) for BUG-3072, and its name is the claim: an append that is ALREAD
internal/store/migrated_test.go:80
FunctionTestDeleteAccountAtomicReport_IssuedGrantWorkspaces
TASK-3365: account deletion reports the workspaces of the grants it deleted, from the deleted rows themselves.
internal/store/task3365_account_delete_report_test.go:12
FunctionTestDeleteAccountAtomic_ConcurrentMutualDeletions
Two account deletions at the same moment, by users whose rows reach into each other (BUG-3286). Each deletion writes some rows through a predicate on
internal/store/delete_account_lockorder_test.go:18
FunctionTestDeleteAccountAtomic_ConcurrentWriters
An account deletion racing an ordinary writer that touches rows the deletion also writes (BUG-3288). The advisory lock in DeleteAccountAtomic serialis
internal/store/delete_account_writer_lockorder_test.go:24
FunctionTestDeleteAccountAtomic_DetachesAuthoredComments
TestDeleteAccountAtomic_DetachesAuthoredComments is a regression test for the FK that became live when TASK-1663 started populating comments.user_id:
internal/store/account_deletion_comments_test.go:16
FunctionTestDeleteAccountAtomic_DoesNotWaitOnAMintBeforeItsLock
The mint has inserted its row but not yet taken the lock. The deletion must NOT wait on it (the mint holds nothing on users: no deadlock), and the min
internal/store/bug3099_delete_account_lock_test.go:89
FunctionTestDeleteAccountAtomic_FullyPopulatedUser
TestDeleteAccountAtomic_FullyPopulatedUser is the TASK-1959 acceptance test: deleting a user who has rows in every table that references users(id) — a
internal/store/account_deletion_test.go:25
FunctionTestDeleteAccountAtomic_ScrubsOutbox
TestDeleteAccountAtomic_ScrubsOutbox is the WIRING test (CONVE-19): the scrub is exercised through the public deletion entry point, not by calling the
internal/store/event_outbox_scrub_test.go:509
FunctionTestDeleteAccountAtomic_WaitsForAMintHoldingTheOwnerLock
The mint holds the owner lock. The deletion must WAIT for it, and then see the committed workspace and delete it. Without the lock the deletion reads
internal/store/bug3099_delete_account_lock_test.go:45
FunctionTestDeleteAccountCtx_CancelStopsBlockedStatement
(t *testing.T)
internal/store/task4278_ctx_test.go:191
← previousnext →19,101–19,200 of 28,840, ranked by callers