newQuietHandle routes [ocr] progress lines away from stdout so they cannot corrupt a structured output document. What it does depends on why stdout ne

retryCodesMiddleware returns an HTTP middleware that forces the SDK to retry responses whose status code is in the given set, by injecting the x-shoul

runBrowserCmd starts one opener and reports whether it worked. Checking only Start() is not enough: it returns as soon as fork/exec succeeds, while xd

sanitizeReason is the single, best-effort redaction+truncation floor applied to every reason the builder stores, so no caller path can write an unreda

BuildReLocationMessages renders the re-location prompt for cm against d. Returns nil when the task template is absent or empty, which the caller treat

DiscoverRepos walks the sessions root and returns one entry per subdirectory.

ErrorEvent emits an error event with error status.

Finalize performs the hard-validated close of the run and returns the immutable manifest. In order it: sweeps any item still selected into failed (col

IsPathExcluded returns true when relPath matches any of the supplied gitignore-style patterns or any default excluded directory prefix (see ExcludedDi

LoadSummary loads a single session's Summary. Errors when the session file is missing or unreadable.

Native extracts the replay state of the first choice, if any.

NewProvider creates a Provider for range mode: from..to (via merge-base).

ResolveIdentity computes the identity a review with these args would record, without creating anything: no session, manifest, runner or LLM call. Resu

RunManifest returns the frozen v1 coverage result for review runs. Scan remains legacy and returns nil.

SessionID returns the current scan's persisted session ID. It returns an empty string when no session file exists.

Swap replaces the stdout writer with replacement and returns a restore function. It lets tests capture output written through Writer(), and lets calle

VisibleContent extracts visible text only, never falling back to ReasoningContent. History builders must use this to avoid duplicating reasoning that

buildOpenAIParams converts the shared ChatRequest into OpenAI SDK parameters.

emitFailureUsage writes a best-effort structured usage record to stderr when a review fails, so the outer caller still sees the cost of the failed att

filterLargeDiffs drops diffs whose diff content alone consumes more than 80% of MaxTokens.

initRepoWithChange creates a real git repository with one committed file and an uncommitted working-tree modification, returning the repo dir. There i

loadLLMRuntime loads tool defs from toolConfigPath, reads the app config from the user's default config path (applying the configured language to tpl

metaFactory builds the same closure internal/agent injects, so these tests exercise the review wiring rather than a test-only shape.

peekSession reads the first record, all review_item records (for comment count), and the last record of a JSONL file.

End-to-end coverage of the #368 P5 wiring: a real review run against the fake Anthropic server, driven through runReview so review_cmd.go's own Freeze

setTestHome redirects the user home resolved by os.UserHomeDir() to dir for the duration of the test. Setting HOME alone is NOT enough on Windows: os.

--- helpers ---

validateReviewRefs rejects ref-option injection (#112): any --from/--to/ --commit value must be a real commit ref and must not start with '-'.

CountMessagesTokens returns the rough token count of msgs by summing the per-message token count. Exported because both review and scan top layers may

IsUserExcluded reports whether the given path matches any user exclude pattern. The check is case-insensitive: both path and pattern are lowercased.

LoadDetail returns the summary plus per-file item records for one session.

MarkReused records that the item was reused from a parent session checkpoint.

MarkWaived records that the user explicitly waived this selected item, which must still be in the pre-terminal (selected) state. reason is required an

ParseEffort validates a user-supplied effort value.

ParseHunks parses raw unified diff text for a single file into a slice of Hunks. Lines before the first @@ header (file-level headers like "diff --git

PrintTraceSummary prints a one-line summary of the review to stdout. If SessionID is non-empty, an "[ocr] Session: <id>" line follows the summary.

ReLocateComment calls the LLM to regenerate a precise existing_code snippet when text-based matching fails, then retries ResolveComment with the new s

ReasoningContent extracts the reasoning content of the first choice, if any.

RecordReviewItemReused records that this run reused a checkpoint from another session.

ResolveComment attempts to resolve StartLine/EndLine for a single comment by matching ExistingCode against the diff. Returns true on success.

ReviseLastAttempt rewrites the last attempt of m as an error. It is the client-boundary correction for what the observer cannot see: an error surfaci

Session returns the session history associated with this Agent.

SessionsDir returns the on-disk directory that holds JSONL session files for a given repository. It does not create the directory.

StartLLMSpan creates a span for an LLM request with standard attributes.

StripMarkdownFences removes ```json and ``` wrappers some models add around structured outputs. Exposed so callers (e.g. agent's review-filter post-st

WithRequestMeta attaches m to ctx so the retry observer can associate every HTTP attempt made under that context with one logical request. It is the o

applyCLIExcludes appends user-supplied --exclude patterns (already split into a []string) onto cc.FileFilter.Exclude. Creates the FileFilter if none w

extractSideLines extracts one side of the diff from a hunk. When newSide is true, returns context+added lines with new-file line numbers. When newSide

fileGroupKey returns a deterministic key for a file group: sorted paths joined by comma.

groupBatches partitions items according to strategy, then slices each natural group into BatchSize-sized chunks (when size > 0). Within a batch the in

initBareRepo creates an empty (unborn) git repository with identity configured but no commits, returning the repo dir.

initRepo creates a git repo with one commit containing the given files and returns the repo dir and the commit SHA.

looksLikeFilePath returns true when s is likely a file path (not inline content). Heuristic: multi-line text is always inline; single-line text withou

newCustomModelTUI builds a modelTUIModel backed by a custom provider with a writable config path, so add/delete/persist operations exercise the save p

newOpenAITestServer returns an httptest server that captures each request's headers and JSON body and responds with a minimal valid chat completion.