MCPcopy Create free account

hub / github.com/PerpetualSoftware/pad / functions

Functions28,840 in github.com/PerpetualSoftware/pad

FunctionTestNeedsSnooze_NewCommentResurfaces
(t *testing.T)
internal/store/needs_snooze_test.go:94
FunctionTestNeedsSnooze_PerUser
(t *testing.T)
internal/store/needs_snooze_test.go:292
FunctionTestNeedsSnooze_SameSecondBoundary
(t *testing.T)
internal/store/needs_snooze_test.go:152
FunctionTestNeedsSnooze_StatusRoundTripResurfaces
(t *testing.T)
internal/store/needs_snooze_test.go:172
FunctionTestNeedsSnoozesForItems_DismissMatchesSingle
The batch dismissal state must agree with GetNeedsSnooze item by item.
internal/store/needs_read_test.go:298
FunctionTestNeedsSnoozesForItems_JoinsItemsWorkspace
(t *testing.T)
internal/store/needs_read_test.go:262
FunctionTestNeedsYouSignals_BindingThroughServeHTTP
Through a real server.New and the router (team CONVE-19): the scorers New registers fill both scores, and sort=importance and simplicity order by them
internal/server/needs_scorers_test.go:633
FunctionTestNeedsYouSignals_CursorPageScoresAtAsOf
Through the router: a request with a cursor scores at the cursor's as_of, not the wall clock, so every page measures age and deadlines from one moment
internal/server/needs_scorers_test.go:683
FunctionTestNeedsYouSignals_DeadlineOverdueMatchesDashboard
The scorer's overdue verdict is the dashboard's, for every stored shape.
internal/server/needs_scorers_test.go:152
FunctionTestNeedsYouSignals_DeadlineUsesServerLocalDay
D agrees with the dashboard's server-local day (overdueToday), not the UTC day.
internal/server/needs_scorers_test.go:132
FunctionTestNeedsYouSignals_FanoutCycleTerminates
(t *testing.T)
internal/server/needs_scorers_test.go:41
FunctionTestNeedsYouSignals_FanoutFailsClosed
Every form of scope that does not grant anything fails closed: nothing is counted, whatever the graph holds.
internal/server/needs_scorers_test.go:102
FunctionTestNeedsYouSignals_FanoutSkipsHidden
(t *testing.T)
internal/server/needs_scorers_test.go:74
FunctionTestNeedsYouSignals_FanoutUnknownCollectionFailsClosed
A target whose collection is not known is neither counted nor expanded.
internal/server/needs_scorers_test.go:123
FunctionTestNeedsYouSignals_HeuristicWithoutRunner
(t *testing.T)
internal/server/needs_scorers_test.go:323
FunctionTestNeedsYouSignals_HintTable
(t *testing.T)
internal/server/needs_scorers_test.go:371
FunctionTestNeedsYouSignals_ImportancePartsPinned
Every input's exact contribution (hand-computed in the comments). The parts are the unweighted inputs; the values are round(100 * weighted sum), chose
internal/server/needs_scorers_test.go:182
FunctionTestNeedsYouSignals_ImportanceRoundingSnapped
A weighted sum that is exactly .325 rounds up whatever order it was summed in.
internal/server/needs_scorers_test.go:271
FunctionTestNeedsYouSignals_MemoReadsOncePerRequest
A request does each of a workspace's batch reads once, though two scorers want them, and the next request does them again: the memo is request-scoped.
internal/server/needs_scorers_test.go:504
FunctionTestNeedsYouSignals_NeedsHumanNeedsFullCurrency
H and "judged to need a human" read needs_human through the inbox's own flag rule, which wants the answer current by item state as well as question: a
internal/server/needs_scorers_test.go:854
FunctionTestNeedsYouSignals_NeedsHumanProbabilityClamped
H is a probability: a stored value outside [0,1] cannot push importance past its bounds.
internal/server/needs_scorers_test.go:289
FunctionTestNeedsYouSignals_NeedsHumanWhyAtThreshold
"judged to need a human" is given from the attention threshold up, not below it.
internal/server/needs_scorers_test.go:307
FunctionTestNeedsYouSignals_ReadFailureDegrades
A batch read that fails leaves the rows unscored and says so: each real scorer reports degraded with the error, and sets no score from partial data.
internal/server/needs_scorers_test.go:827
FunctionTestNeedsYouSignals_RestrictedViewerScoreHasNoHiddenBlocks
Through the router (team CONVE-19): a restricted member's score is built from what that member may see. The row blocks an item they cannot see; the hi
internal/server/needs_scorers_test.go:743
FunctionTestNeedsYouSignals_VocabularyIsDocumented
docs/api.md publishes the Why vocabulary; the code's list and the doc agree.
internal/server/needs_scorers_test.go:727
FunctionTestNeedsYouSignals_WhyOnlyFromTheVocabulary
Why is a fixed vocabulary: whatever the inputs, a scorer emits only strings from needsWhyVocabulary (the list docs/api.md publishes), never the title.
internal/server/needs_scorers_test.go:461
FunctionTestNeedsYouSort_CursorFieldsAreChecked
Every decoded cursor field is checked: a row per field, each a 400 invalid_cursor. as_of's far-future case is deliberately absent: it is clamped, not
internal/server/needs_scorers_test.go:915
FunctionTestNeedsYouSort_CursorStableAcrossPages
(t *testing.T)
internal/server/needs_scorers_test.go:553
FunctionTestNeedsYouSort_FutureAsOfIsClamped
A cursor's as_of in the far future is clamped to now, not trusted: the page scores as of now, never later (docs/api.md: "an as_of later than the serve
internal/server/needs_scorers_test.go:992
FunctionTestNeedsYouSort_OddBoundaryTimesPageThrough
Every cursor the server mints is one it follows (TASK-3959 part 2 r2): a boundary row whose created_at, and so its asked waiting_since, is pre-1970, u
internal/server/needs_scorers_test.go:1028
FunctionTestNegativeTerminalsMatchTheWebCopy
TestNegativeTerminalsMatchTheWebCopy pins the web's NEGATIVE_TERMINALS (web/src/lib/types/index.ts) to NegativeTerminals. The web copy decides which c
internal/models/negative_terminals_parity_test.go:16
FunctionTestNeverAPartialAnswer
0 is the default TestNeverAPartialAnswer is the instrument for the "CapStall, never a partial answer" contract (TASK-4100). Earlier cases each exposed
internal/schedule/neverpartial_test.go:37
FunctionTestNew
(t *testing.T)
internal/metrics/metrics_test.go:18
FunctionTestNewClientFromURL_EnvTokenLeavesStoreUntouched
The override must not touch the store: reads stay side-effect-free (the invariant the v2 store documents), so a process running under PAD_TOKEN never
internal/cli/env_token_test.go:88
FunctionTestNewClientFromURL_EnvTokenOverridesStore
PAD_TOKEN must beat the stored credential: the env var is the caller's explicit, per-process choice; the file is ambient machine state.
internal/cli/env_token_test.go:65
FunctionTestNewClientFromURL_NoEnvFallsBackToStore
(t *testing.T)
internal/cli/env_token_test.go:75
FunctionTestNewErrorResult_Envelope
TestNewErrorResult_Envelope confirms the wire shape every error path produces. The test inspects: - IsError flag set on the result. - StructuredConte
internal/mcp/errors_test.go:23
FunctionTestNewErrorsOnConfiguredButUnsatisfiable
(t *testing.T)
internal/decision/config_test.go:34
FunctionTestNewFSStore_EmptyBaseDirRejected
(t *testing.T)
internal/attachments/fs_store_test.go:282
FunctionTestNewIPRateLimiter_RaisesShortRetention
TestNewIPRateLimiter_RaisesShortRetention (BUG-3638): a retention shorter than the refill window is raised to it, whether set explicitly or left to th
internal/server/middleware_ratelimit_bug3638_test.go:54
FunctionTestNewItemCopyDeps_ServerHonorsKeysFollowsTheCapability
TASK-4131 (CONVE-19): the real binding reports the server's idempotency_keys capability, so the unknown-outcome advice follows the server rather than
cmd/pad/item_copy_test.go:2250
FunctionTestNewItemTop_ChildJoinsTheEnd
A new child JOINS THE END of its siblings: ship and decompose rely on it, including for children created in the same second.
internal/store/items_new_item_top_test.go:150
FunctionTestNewItemTop_ClampsAtTheIntegerFloor
The reorder API accepts the INTEGER floor; MIN - 1 would overflow Postgres.
internal/store/items_new_item_top_test.go:314
FunctionTestNewItemTop_ConcurrentCreates
(t *testing.T)
internal/store/items_new_item_top_test.go:249
FunctionTestNewItemTop_CrossWorkspaceCopy
(t *testing.T)
internal/store/items_new_item_top_test.go:229
FunctionTestNewItemTop_DraggedLane
(t *testing.T)
internal/store/items_new_item_top_test.go:78
FunctionTestNewItemTop_FencedCreate
(t *testing.T)
internal/store/items_new_item_top_test.go:286
FunctionTestNewItemTop_FencedCreateClampsAtTheIntegerFloor
The fenced create path clamps at the INTEGER floor too.
internal/store/items_new_item_top_test.go:331
FunctionTestNewItemTop_NeverDraggedLane
(t *testing.T)
internal/store/items_new_item_top_test.go:51
FunctionTestNewItemTop_ParentlessMinIgnoresChildren
The MIN covers PARENTLESS live items only: children with very negative sort_order do not push a parentless create further down.
internal/store/items_new_item_top_test.go:214
FunctionTestNewItemTop_ProvisionedArtifactsKeepManifestOrder
An app's artifacts append, so they list in manifest order (BUG-4005 r7-H1).
internal/store/items_new_item_top_test.go:385
FunctionTestNewItemTop_ShipOrderAfterDecompose
The ship playbook expands a plan into its children "ordered by sort_order" and works them in that order; decompose creates them in sequence.
internal/store/items_new_item_top_test.go:178
FunctionTestNewItemTop_TemplateSeedsKeepTheirOrder
Template seeds append, so a fresh workspace lists them in the order the template defines: first seeded, first listed (BUG-4005 r7-H1). Seeds are creat
internal/store/items_new_item_top_test.go:361
FunctionTestNewPinsTheModel
(t *testing.T)
internal/decision/config_test.go:61
FunctionTestNewPostgres_CloneCarriesTheMigratedSchema
TestNewPostgres_CloneCarriesTheMigratedSchema is the counterpart to the isolation test above, and it fails in the opposite direction: a clone that arr
internal/store/storetest/postgres_test.go:123
FunctionTestNewPostgres_ClonesAreIsolatedAndFunctional
TestNewPostgres_ClonesAreIsolatedAndFunctional proves each call returns its own writable, working database rather than a shared one: a real end-to-end
internal/store/storetest/postgres_test.go:43
FunctionTestNewPostgres_DatabasesAreClonedNotMigrated
TestNewPostgres_DatabasesAreClonedNotMigrated is the test that actually pins the mechanism, and it exists because the build-once counter above does NO
internal/store/storetest/postgres_test.go:90
FunctionTestNewPostgres_TemplateBuildsOnce
TASK-2900. These mirror the SQLite template's tests, and they exist for a reason worth stating plainly: **a Postgres template that silently stopped be
internal/store/storetest/postgres_test.go:23
FunctionTestNewPostgres_TraceGate
(t *testing.T)
internal/store/storetest/pgtrace_test.go:49
FunctionTestNewReturnsNilWhenUnconfigured
(t *testing.T)
internal/decision/config_test.go:8
FunctionTestNewSQLite_CopiesAreIsolatedAndFunctional
TestNewSQLite_CopiesAreIsolatedAndFunctional proves each NewSQLite call returns its own writable, working copy: writes to one copy (a real end-to-end
internal/store/storetest/storetest_test.go:70
FunctionTestNewSQLite_NoWALSidecars
TestNewSQLite_NoWALSidecars asserts the template directory contains only the template file itself, beside testtmp's owner lock (BUG-4088) — the checkp
internal/store/storetest/storetest_test.go:45
FunctionTestNewSQLite_TemplateBuildsOnce
TestNewSQLite_TemplateBuildsOnce asserts the migration chain (store.New on the template path) runs at most once per test binary regardless of how many
internal/store/storetest/storetest_test.go:31
FunctionTestNewServer_Construction
TestNewServer_Construction is the smoke test: NewServer with a zero-value Options must not panic, must return a usable Server, and must expose the und
internal/mcp/server_test.go:21
FunctionTestNewServer_FillsDefaultLifespans
(t *testing.T)
internal/oauth/server_test.go:101
FunctionTestNewServer_RefreshTokenScopesIsEmpty
TestNewServer_RefreshTokenScopesIsEmpty pins #371: fosite defaults Config.RefreshTokenScopes to ["offline", "offline_access"] — without overriding, re
internal/oauth/server_test.go:152
FunctionTestNewServer_RejectsMissingAudience
(t *testing.T)
internal/oauth/server_test.go:87
FunctionTestNewServer_RejectsMissingStore
===================================================================== NewServer — argument validation ================================================
internal/oauth/server_test.go:59
FunctionTestNewServer_RejectsShortHMACSecret
(t *testing.T)
internal/oauth/server_test.go:72
FunctionTestNewStore
(t *testing.T)
internal/store/store_test.go:524
FunctionTestNewStorePostgres
(t *testing.T)
internal/store/store_test.go:543
FunctionTestNewWatchEventsStreamRequest_AnnouncesIdentity
TestNewWatchEventsStreamRequest_AnnouncesIdentity covers PLAN-2558 S2's client half: the session identity has to reach the server as request headers,
internal/cli/client_stream_identity_test.go:15
FunctionTestNewWatchEventsStreamRequest_ArmedSendsQueryParam
TestNewWatchEventsStreamRequest_ArmedSendsQueryParam is PLAN-2613 S2's client half: an armed identity must reach the server as ?armed=true (a query pa
internal/cli/client_stream_identity_test.go:123
FunctionTestNewWatchEventsStreamRequest_OmitsUnsetIdentity
TestNewWatchEventsStreamRequest_OmitsUnsetIdentity pins absence rather than emptiness. Sending `X-Pad-Session-Label: ""` would make every unannounced
internal/cli/client_stream_identity_test.go:45
FunctionTestNewWatchEventsStreamRequest_UnarmedOmitsQueryParam
TestNewWatchEventsStreamRequest_UnarmedOmitsQueryParam pins absence, not armed=false. Sending armed=false would be indistinguishable at the server (on
internal/cli/client_stream_identity_test.go:157
FunctionTestNewWatchEventsStreamRequest_UnsendableLabelStillConnects
TestNewWatchEventsStreamRequest_UnsendableLabelStillConnects is the regression for a past failure, and the reason it is asserted by DOING THE ROUND TR
internal/cli/client_stream_identity_test.go:77
FunctionTestNew_RejectsBadEntries
(t *testing.T)
internal/appfetch/appfetch_test.go:135
FunctionTestNextAfterAcrossDST
(t *testing.T)
internal/schedule/nextbetween_test.go:109
FunctionTestNextAfterBeforeStartReturnsSyncedStart
(t *testing.T)
internal/schedule/nextbetween_test.go:65
FunctionTestNextAfterExhaustedCount
(t *testing.T)
internal/schedule/nextbetween_test.go:95
FunctionTestNextAfterExhaustedUntil
(t *testing.T)
internal/schedule/nextbetween_test.go:102
FunctionTestNextAfterHonoursExDates
(t *testing.T)
internal/schedule/nextbetween_test.go:120
FunctionTestNextAfterStrictlyAfter
(t *testing.T)
internal/schedule/nextbetween_test.go:57
FunctionTestNextAfterUnsyncedStartSkipsStart
TestNextAfterUnsyncedStartSkipsStart pins dateutil semantics: a "every Friday" schedule made on a Tuesday does not fire on the Tuesday.
internal/schedule/nextbetween_test.go:78
FunctionTestNilMultiRelation_ControlANilSCALARRelationIsUntouched
(t *testing.T)
internal/items/multi_relation_normalise_test.go:181
FunctionTestNilMultiRelation_FullWriteBecomesAnAbsentKey
The NIL spelling. On a full write the traversal treats a present-but-nil key as absent for required-ness and defaults, but does not REMOVE it — so `{"
internal/items/multi_relation_normalise_test.go:158
FunctionTestNilMultiRelation_PartialWriteKeepsTheDeletionSentinel
(t *testing.T)
internal/items/multi_relation_normalise_test.go:168
FunctionTestNo500DiscardsStoreError
(t *testing.T)
internal/server/task4278_error_mapping_test.go:45
FunctionTestNoAgentFacingTextLoadsAnItemWithMarkdown
The guard: no agent-facing instruction loads an item with --format markdown to READ it. A line may name the format only to describe the edit round tri
cmd/pad/task3451_markdown_load_test.go:104
FunctionTestNoCoverageIsDroppedAtStartup
THE ENFORCEMENT FOR A CLAIM THE RECEIVE LOOP MAKES ABOUT ANOTHER FUNCTION. receiveMessages has no "skip the first confirmation" flag — unlike interna
internal/watchevents/redis_reconnect_test.go:429
FunctionTestNoGoroutineOutlivesCall
TestNoGoroutineOutlivesCall is the instrument for "all work runs in the caller" (TASK-4100). MaxDuration is tiny, so a restored abandon path returns C
internal/schedule/budget_test.go:88
FunctionTestNoInterruptLeaksIntoNextJob
A context ending AFTER the job's JavaScript returned, but before the watcher is stopped, makes the watcher Interrupt a VM that is no longer running. g
internal/materialize/materialize_test.go:242
FunctionTestNoLongerNeededOfAnItemWithoutAnOutstandingOccurrenceIsFalse
(t *testing.T)
internal/store/commitment_cancel_test.go:278
FunctionTestNoPadTomlRetryInterval_IsAnHour
(t *testing.T)
cmd/pad/cmd_watch_test.go:55
FunctionTestNoProtectedTableShadowsRowid
The rowid merge (and the rowid address item_wiki_links already uses) reads SQLite's hidden rowid. A table that DECLARED a column named rowid, _rowid_
internal/store/nulutf8_test.go:184
FunctionTestNoRemoteEquivalentDoesNotAdvertiseTheBrokenPRWorkaround
TestNoRemoteEquivalentDoesNotAdvertiseTheBrokenPRWorkaround. noRemoteEquivalent's text IS the message a remote agent receives when it calls `github l
internal/mcp/stored_state_unreadable_test.go:142
FunctionTestNoRemoteEquivalent_RationaleIsPerEntry
--- noRemoteEquivalent rationale clause is per-entry ---
internal/mcp/dispatch_http_attachments_test.go:440
FunctionTestNoWarningsWhenEveryFieldIsDeclared
A write with nothing to report carries NO warnings element, so the response stays byte-identical to before for every well-formed write (BUG-2850).
internal/server/handlers_items_field_coercion_test.go:404
FunctionTestNonUpgradeCollabRequestPaysTheAPIBucket
Only an upgrade request is a dial. A plain request under the prefix is an ordinary API request and pays the API bucket, so it cannot be used to spend
internal/server/handlers_collab_bug1308_test.go:120
FunctionTestNormalizeAgentNameDelegatesToModels
(t *testing.T)
internal/server/task4186_chat_agent_guard_test.go:35
← previousnext →21,801–21,900 of 28,840, ranked by callers