MCPcopy Create free account

hub / github.com/alibaba/open-code-review / functions

Functions7,169 in github.com/alibaba/open-code-review

FunctionTestExecuteToolCall_CodeCommentSuspectRepairReportsTheError
The far side of the accept/decline decision: a refused batch must reach the model as a failure, since that error is what makes it resend. No warning e
internal/reviewtask/loop_test.go:1109
FunctionTestExecuteToolCall_CodeCommentThinkingBackfill
TestExecuteToolCall_CodeCommentThinkingBackfill covers the thinking backfill: when the current turn carries reasoning content, comments without an exp
internal/reviewtask/loop_execute_more_test.go:336
FunctionTestExecuteToolCall_CodeCommentUsesPerItemPath
(t *testing.T)
internal/reviewtask/loop_test.go:614
FunctionTestExecuteToolCall_CodeCommentValidationErrorPrintsTerminalFailure
(t *testing.T)
internal/reviewtask/loop_execute_test.go:191
FunctionTestExecuteToolCall_CodeCommentWellFormedArgsDoesNotWarn
The contrast that keeps the tests above honest: a conformant batch must warn nothing, so comment_args_repaired counts real violations only.
internal/reviewtask/loop_test.go:1142
FunctionTestExecuteToolCall_CodeCommentWithNoUsableEntryIsReported
(t *testing.T)
internal/reviewtask/loop_test.go:658
FunctionTestExecuteToolCall_DynamicExecuteError
TestExecuteToolCall_DynamicExecuteError verifies newly registered tools use the common failure-observation path without tool-specific instrumentation.
internal/reviewtask/loop_execute_test.go:50
FunctionTestExecuteToolCall_DynamicNotRegistered
TestExecuteToolCall_DynamicNotRegistered covers the path where the LLM calls a name that is neither a built-in tool nor present in the registry.
internal/reviewtask/loop_execute_test.go:34
FunctionTestExecuteToolCall_DynamicParseError
TestExecuteToolCall_DynamicParseError verifies malformed calls to registered tools are captured by the same failure-observation path.
internal/reviewtask/loop_execute_test.go:157
FunctionTestExecuteToolCall_DynamicSuccessRecordsResult
TestExecuteToolCall_DynamicSuccessRecordsResult covers the dynamic-tool success path with a non-nil TaskRecord so AddToolResult runs.
internal/reviewtask/loop_execute_test.go:85
FunctionTestExecuteToolCall_FailureStreakScopedPerTaskAndTool
TestExecuteToolCall_FailureStreakScopedPerTaskAndTool confirms the counter is keyed by (taskKey, toolName): failures against a different taskKey or a
internal/reviewtask/tool_failure_streak_test.go:70
FunctionTestExecuteToolCall_KnownToolNotRegistered
TestExecuteToolCall_KnownToolNotRegistered covers the lookupTool-nil branch: a built-in tool the model may call but which is absent from the registry.
internal/reviewtask/loop_execute_test.go:113
FunctionTestExecuteToolCall_RepeatedFailureBackoff
TestExecuteToolCall_RepeatedFailureBackoff drives the same (taskKey, toolName) pair to failure three times in a row and checks the response escalates
internal/reviewtask/tool_failure_streak_test.go:21
FunctionTestExecuteToolCall_SuccessResetsFailureStreak
TestExecuteToolCall_SuccessResetsFailureStreak confirms a success for a (taskKey, toolName) pair resets its streak, so a later failure starts over at
internal/reviewtask/tool_failure_streak_test.go:113
FunctionTestExecuteToolCall_TaskDone
TestExecuteToolCall_TaskDone covers every branch of the task_done handling: argument parse error, missing state (implicit completion), non-string stat
internal/reviewtask/loop_execute_more_test.go:187
FunctionTestExecute_CommitMode
(t *testing.T)
internal/tool/file_read_test.go:401
FunctionTestExecute_DefaultStartLine
(t *testing.T)
internal/tool/file_read_test.go:420
FunctionTestExecute_EmptyFilePath
(t *testing.T)
internal/tool/file_read_test.go:344
FunctionTestExecute_InvalidLineRange
(t *testing.T)
internal/tool/file_read_test.go:357
FunctionTestExecute_LineCapIsUtf8Safe
The mid-line cut must land on a UTF-8 rune boundary so the rendered line stays valid UTF-8 instead of ending in a split multibyte sequence.
internal/tool/file_read_test.go:633
FunctionTestExecute_MissingFile
(t *testing.T)
internal/tool/file_read_test.go:391
FunctionTestExecute_ResultBudgetCountsFormattedLine
The budget must be counted over the fully formatted lines: the line-number prefix and the trailing newline are body bytes too. Counting only the line
internal/tool/file_read_test.go:587
FunctionTestExecute_SingleOversizeLineIsCapped
A minified or generated file can pack its whole content into one line, which the 500-line cap counts as a single line. The tool result must still boun
internal/tool/file_read_test.go:479
FunctionTestExecute_StartBeyondTotalLines
(t *testing.T)
internal/tool/file_read_test.go:374
FunctionTestExecute_TotalResultBytesCapped
The per-line cap alone is not enough: 500 capped lines can still render far more bytes than one request should carry, so the total rendered body needs
internal/tool/file_read_test.go:547
FunctionTestExecute_Truncation
(t *testing.T)
internal/tool/file_read_test.go:304
FunctionTestExecute_UnrenderedOversizeLineDoesNotReportLineTruncation
(test *testing.T)
internal/tool/file_read_test.go:507
FunctionTestExecute_WithEndLine
setupTestRepo is already defined in code_search_test.go (same package) getHeadCommit is already defined in code_search_test.go (same package)
internal/tool/file_read_test.go:442
FunctionTestExecutionCommandQuotesArguments
(t *testing.T)
acp/internal/adapter/presentation/progress_format_test.go:40
FunctionTestExecutionDetailsSurviveUpdatesWithoutCommandInChat
(t *testing.T)
acp/internal/adapter/progress_test.go:31
FunctionTestExecutionStatsFromCLIJSON
(t *testing.T)
acp/internal/adapter/presentation/progress_format_test.go:70
FunctionTestExpandBraces_MultipleOptions
(t *testing.T)
internal/config/rules/system_rules_test.go:39
FunctionTestExpandBraces_NoBraces
(t *testing.T)
internal/config/rules/system_rules_test.go:19
FunctionTestExpandBraces_SingleGroup
(t *testing.T)
internal/config/rules/system_rules_test.go:26
FunctionTestExpandBraces_UnclosedBrace
(t *testing.T)
internal/config/rules/system_rules_test.go:52
FunctionTestExpandSessionKeyInBody
(t *testing.T)
internal/llm/sessionkey_test.go:170
FunctionTestExpandSessionKeyInHeaders
(t *testing.T)
internal/llm/sessionkey_test.go:144
FunctionTestExplainErrorClassifiesBedrockFailures
TestExplainErrorClassifiesBedrockFailures covers the diagnosis Bedrock's own wording does not give. Two of these are actively misleading: the API-key
internal/llm/bedrock_test.go:379
FunctionTestExplainErrorLeavesNonBedrockErrorsAlone
TestExplainErrorLeavesNonBedrockErrorsAlone keeps the diagnosis scoped: every other protocol shares this client type.
internal/llm/bedrock_test.go:492
FunctionTestExplainErrorNamesTheBearerTokenVariable
TestExplainErrorNamesTheBearerTokenVariable separates the two ways the same 403 arrives: an SSO token the SDK attached on its own, versus a token the
internal/llm/bedrock_test.go:481
FunctionTestExplicitPlanConsent
(t *testing.T)
internal/chatgptauth/recovery_test.go:94
FunctionTestExportSession_Errors
TestExportSession_Errors pins that a failure writes nothing, so a caller pointing at a file cannot end up with a half-rendered page.
internal/viewer/export_test.go:138
FunctionTestExportSession_SelfContained
TestExportSession_SelfContained is the whole point of the feature: the page must render offline, so all /static/ assets have to arrive inlined and cor
internal/viewer/export_test.go:41
FunctionTestExportSession_Variants
(t *testing.T)
internal/viewer/export_test.go:80
FunctionTestExtFromPath
(t *testing.T)
internal/scan/coverage_test.go:302
FunctionTestExtFromPath
(t *testing.T)
internal/agent/agent_test.go:379
FunctionTestExtraEnumCaseCompatibility
(t *testing.T)
acp/internal/contract/intent_test.go:369
FunctionTestExtraHeadersAndBody
(t *testing.T)
acp/internal/intent/llm/client_test.go:305
FunctionTestExtractCodeBlock
(t *testing.T)
internal/diff/relocation_test.go:275
FunctionTestExtractSideLines_DivergentStartLines
(t *testing.T)
internal/diff/resolver_test.go:388
FunctionTestExtractSideLines_NewSide
--------------------------------------------------------------------------- extractSideLines unit tests ----------------------------------------------
internal/diff/resolver_test.go:326
FunctionTestExtractSideLines_OldSide
(t *testing.T)
internal/diff/resolver_test.go:357
FunctionTestExtractSideLines_OnlyAdded
(t *testing.T)
internal/diff/resolver_test.go:424
FunctionTestExtractSideLines_OnlyDeleted
(t *testing.T)
internal/diff/resolver_test.go:445
FunctionTestExtractText_ContentBlocks
(t *testing.T)
internal/llm/message_test.go:88
FunctionTestExtractText_Default
(t *testing.T)
internal/llm/message_test.go:115
FunctionTestExtractText_NestedContentBlocks
(t *testing.T)
internal/llm/message_test.go:98
FunctionTestExtractText_NilContent
(t *testing.T)
internal/llm/message_test.go:122
FunctionTestExtractText_String
(t *testing.T)
internal/llm/message_test.go:81
FunctionTestExtractTopLevelJSON
TestExtractTopLevelJSON covers extractTopLevelJSON directly: quoted braces inside strings must not perturb the depth count, and escaped quotes must no
internal/reviewtask/tool_args_json_test.go:61
FunctionTestFail
(t *testing.T)
internal/tool/response_message_test.go:21
FunctionTestFailedDifferentClassErrorsAndKeepsFirst
Re-marking a failed item with a different classification is a conflicting transition: it errors and the first classification is kept (never silently o
internal/session/manifest_test.go:368
FunctionTestFailedPlusWaivedIsPartial
A failed + waived + completed mix still has an unresolved failure -> partial.
internal/session/manifest_test.go:286
FunctionTestFailedSameClassIsIdempotent
Re-marking a failed item with the same classification is idempotent; reason is free text and does not participate, so the first reason is kept.
internal/session/manifest_test.go:345
FunctionTestFileFilter_CaseInsensitive
(t *testing.T)
internal/config/rules/system_rules_test.go:778
FunctionTestFileFilter_IsUserExcluded
(t *testing.T)
internal/config/rules/system_rules_test.go:722
FunctionTestFileFilter_IsUserIncluded
(t *testing.T)
internal/config/rules/system_rules_test.go:747
FunctionTestFileFilter_IsUserIncluded_EmptyInclude
(t *testing.T)
internal/config/rules/system_rules_test.go:771
FunctionTestFileFindProvider_Tool
(t *testing.T)
internal/tool/file_find_test.go:74
FunctionTestFileFind_BasenamePrecisionAndFallback
(t *testing.T)
internal/tool/file_find_test.go:283
FunctionTestFileFind_BlankQuery
(t *testing.T)
internal/tool/file_find_test.go:81
FunctionTestFileFind_CaseSensitive
(t *testing.T)
internal/tool/file_find_test.go:213
FunctionTestFileFind_GitRepo_CommitMode
(t *testing.T)
internal/tool/file_find_test.go:145
FunctionTestFileFind_GitRepo_StarlarkBuildFiles
(t *testing.T)
internal/tool/file_find_test.go:166
FunctionTestFileFind_GitRepo_WithRunner
(t *testing.T)
internal/tool/file_find_test.go:199
FunctionTestFileFind_GitRepo_WorkspaceMode
(t *testing.T)
internal/tool/file_find_test.go:132
FunctionTestFileFind_NonASCIIPath
TestFileFind_NonASCIIPath verifies that file_find reports non-ASCII paths literally instead of as quoted octal escapes under git's default core.quotep
internal/tool/file_find_test.go:347
FunctionTestFileFind_NonGitDirectoryFallback
TestFileFind_NonGitDirectoryFallback verifies file_find works in a plain (non-git) directory by falling back to a filesystem walk instead of failing w
internal/tool/file_find_test.go:20
FunctionTestFileFind_NonGitDirectoryNoMatch
TestFileFind_NonGitDirectoryNoMatch verifies the not-found path in a non-git dir returns the sentinel rather than an error.
internal/tool/file_find_test.go:58
FunctionTestFileFind_PathWithSubdirectory
(t *testing.T)
internal/tool/file_find_test.go:240
FunctionTestFileGroupKey_Multiple
(t *testing.T)
internal/agent/grouping_test.go:234
FunctionTestFileGroupKey_Single
(t *testing.T)
internal/agent/grouping_test.go:227
FunctionTestFileReadDiffAllPathsFoundHasNoError
Every requested path has a diff: no part of the missing-path message may appear at all.
internal/tool/file_read_diff_missing_test.go:80
FunctionTestFileReadDiffCapsAvailablePathList
The available-path list exists to be read in one glance, so it is capped and the remainder is counted rather than listed.
internal/tool/file_read_diff_missing_test.go:102
FunctionTestFileReadDiffEmptyRequestUnchanged
An empty request stays a malformed call with its own message.
internal/tool/file_read_diff_missing_test.go:124
FunctionTestFileReadDiffEmptySnapshotClaimsNoMode
An empty snapshot has no paths to offer. Both review and scan reach this state only through their own initialization (scan injects full file contents,
internal/tool/file_read_diff_missing_test.go:140
FunctionTestFileReadDiffNamesMissingPathAndAvailablePaths
During a review the snapshot holds the change set, and the model is asked to confirm cross-file concerns by reading other files' diffs. When it names
internal/tool/file_read_diff_missing_test.go:24
FunctionTestFileReadDiffProvider_Execute
(t *testing.T)
internal/tool/file_read_diff_test.go:35
FunctionTestFileReadDiffProvider_SetDiffMap
(t *testing.T)
internal/tool/file_read_diff_test.go:94
FunctionTestFileReadDiffProvider_Tool
(t *testing.T)
internal/tool/file_read_diff_test.go:104
FunctionTestFileReadDiffReportsPartialHits
Several paths in one call: the message must not pretend every path was missing, or the model stops looking for the one that was. The error line names
internal/tool/file_read_diff_missing_test.go:56
FunctionTestFileReadProvider_Tool
(t *testing.T)
internal/tool/file_read_test.go:337
FunctionTestFileReadRef
TestFileReadRef pins the other half of sealing the input. The diff is pinned to the commit admission resolved, so file_read has to read at that same c
cmd/opencodereview/review_resume_more_test.go:430
FunctionTestFileReader_ReadLines_CommitMode_MissingFile
(t *testing.T)
internal/tool/filereader_read_test.go:192
FunctionTestFileReader_ReadLines_CommitMode_WithRunner
(t *testing.T)
internal/tool/filereader_read_test.go:174
FunctionTestFileReader_ReadLines_Workspace
(t *testing.T)
internal/tool/filereader_read_test.go:73
FunctionTestFileReader_Read_CommitMode
(t *testing.T)
internal/tool/filereader_read_test.go:119
FunctionTestFileReader_Read_CommitMode_MissingFile
(t *testing.T)
internal/tool/filereader_read_test.go:136
FunctionTestFileReader_Read_CommitMode_MonorepoSubdirPath
TestFileReader_Read_CommitMode_MonorepoSubdirPath reproduces #287 at the git-show layer: in a monorepo, git reports paths relative to the repo root (e
internal/tool/filereader_read_test.go:228
FunctionTestFileReader_Read_CommitMode_WithRunner
(t *testing.T)
internal/tool/filereader_read_test.go:147
← previousnext →3,901–4,000 of 7,169, ranked by callers