ContextWithSessionKey returns a context carrying the given session key. Review and scan runs bind their session history's SessionID at the top of Run

LoadSession fully parses a JSONL file into a ViewSession.

MarkFailed records that the item failed. class must be a valid, specific failure class — an empty or unknown-string class is rejected rather than down

executeToolCall dispatches a single tool call from the LLM response and records the result in session history. code_comment handling includes optional

parseScanFlags provides test compatibility: parses args through a fresh cobra command instance and returns the resulting scanOptions.

runGitTest runs a git command in dir and fails the test on error.

IsEnabled returns true when telemetry has been initialized with exporters.

SessionFilePath returns the JSONL path for a persisted session.

Build an HTTPError suitable for mocking urlopen.

NewCommentWorkerPool creates a pool with the given concurrency limit. workerCount <= 0 defaults to 8.

RegisterSelected records an item in the selected set (the coverage denominator). It must be called once per item during the pre-dispatch pass, after f

ResolveEndpointWithOptions resolves an endpoint by first honoring an explicit provider selection. Otherwise it processes config, environment, then she

SetError records an error for this task record, writes an llm_error entry to the JSONL stream, and increments the session-level LLM failure counter.

newModelTUI builds a model-only TUI for tests. It has no config path or existing config, so add/delete/persist operations are unavailable — use newMod

LoadResumeState replays a previous session JSONL into a fingerprint index. A record that cannot be parsed fails the load: with nothing to arbitrate co

Manifest returns the run's coverage builder, or nil for legacy/scan sessions that did not opt in via SessionOptions.Operation. Callers must be nil-saf

ValidateOptions verifies that this session can be resumed in the requested review mode at all. It deliberately does not compare the ref text the user

cloneProviderEntry deep-copies a ProviderEntry so callers (rollback paths, map cloning) can safely mutate the returned value without aliasing the orig

GetDiff returns all changes as parsed model.Diff structs.

NormalizeAuthHeader normalizes an auth header value to a canonical form. It returns an error for unrecognized values.

NormalizeProtocol canonicalizes protocol names. It is case-insensitive and trims whitespace. Empty string is returned as-is (the caller decides the de

Output executes a git command and returns stdout only.

Read returns the full content of a file path (relative to RepoDir), resolved according to the active review mode. - Workspace: reads directly from the

mustUnmarshal is a test helper that unmarshals SARIF output and fails the test on error, so malformed output produces a clear failure instead of a nil

Await blocks until all submitted work has completed and returns aggregated results from every Submit call so far. A panic in submitted work is recove

CompletionTokenLimit returns the output cap for LLM requests. Runtime prompt-limit overrides must not silently expand the model's output budget.

ResolveEndpointWithModelOverride is a compatibility wrapper around ResolveEndpointWithOptions for callers that only override the model.

SetAttr sets a single attribute on a span.

SetResponse records the LLM response in the most recent TaskRecord of the given type. It uses actual token usage from the API response when available,

applyResumeLine folds one record into the index. It reports an unparseable line to the caller, which decides whether that is fatal.

captureStderr captures everything written to os.Stderr during fn. Mirrors captureStdout; used to assert structured usage emitted on the failure path.

--- helpers ---

outputSARIF writes a SARIF v2.1.0 document to stdout. The document always contains a single run with the 8 built-in category rules, one result per Llm

resolveKeyCmd runs a credential-fetching shell command and returns its trimmed, single-line stdout. label names the source (e.g. `api_key_cmd for prov

retryReportFixture is the report used by most rendering assertions: one request recovered after two errors, one request that never succeeded, and a to

setColor pins the resolved color decision for one test and restores the previous state afterwards. Tests capture stdout through a pipe, which is not a

Event emits a structured event as a span with immediate end.

LoadDefault parses the embedded system_rules.json and resolves rule file references.

--- Factory --- NewLLMClient creates the appropriate client based on the resolved endpoint protocol. protocol dispatch (canonical names from protocol.

NewManifestBuilder creates a builder for a run identified by runID (the canonical session ID) and operation ("review" or "scan").

NewProvider creates a Provider that enumerates the repository at repoDir. If paths is non-empty each element must be a repo-relative path (file or dir

NewWorkspaceProvider creates a Provider for workspace mode (current uncommitted changes).

ParseDiffText splits the unified diff text into per-file Diff structs. ref, if non-empty, is a git ref used to read new-file content via git show inst

RecordReviewItemFailed persists an incomplete file-level checkpoint.

SessionTaskKey derives the prompt-cache affinity key for one task conversation within a session. Prompt caches match on prefixes, and OCR's task types

SetRunFailure records why the whole run stopped. It rejects an invalid class (so an unclassifiable stop is never silently downgraded) and a change of

beginAPIKeyReplace switches from the fixed-length mask placeholder to edit mode so the next keystroke or paste fully replaces the saved key. While edi

Default config file location: ~/.opencodereview/config.json

expandBraces turns "{a,b,c}" style patterns into individual strings. e.g. "*.go.{java,kotlin}" → ["*.go.java", "*.go.kotlin"]. If no braces exist, ret

gitCommitFile writes a file and commits it, returning after the commit lands.

loadCommonContext validates the working directory, loads the embedded template, raises MaxToolRequestTimes when maxTools exceeds the default, resolves

matchConsecutive scans sideLines for a consecutive run matching all targetLines.

newFakeLLM returns a server that succeeds on the first attempt for every file. rateLimitOnce and hardFail are set by the caller before use.

resolveUsage parses raw JSON bytes into a map and extracts token usage by probing configured paths sequentially. Returns nil if no total_tokens found.

retryTestRepo builds a two-file repo with one reviewable commit range (HEAD~1..HEAD), each file carrying its own marker in the second commit only.

silenceStdout redirects os.Stdout to /dev/null for the duration of fn so the delegate commands' Printf output does not clutter test logs.

startFakeLLM starts srv and points the OCR_LLM_* endpoint resolution at it, with HOME/XDG_CONFIG_HOME redirected so the developer's real config and se