Skip to content
AI usage dashboard

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs

Comment 1, line 1

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:1

//! Codex per-version regression tests (M2-D, V17/V30).

Comment 2, line 2

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:2

//! Redacted local samples cover 0.153.0, 0.154.0-alpha.6.1, and 0.154.0-alpha.6.2.

Comment 3, line 3

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:3

//! Check registry KnownVersion selection and individual usage against independent jq sums,

Comment 4, line 4

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:4

//! known_version parse basis, matched snapshot comparisons, and repeat-scan deduplication.

Comment 5, line 5

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:5

//! Expected values: each sample's _expectations.md, checked manually against these tests.

Comment 6, line 6

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:6

//! Older 0.139-0.151 token_count/last_token_usage formats dispatch to rollout_legacy.

Comment 7, line 7

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:7

//! Samples 0.139.0/0.142.5/0.146.0-alpha.3 check repeated reports, carried compaction

Comment 8, line 8

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:8

//! responses, and source counter regressions with visible comparison mismatches.

Comment 9, line 26

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:26

/// Final token_count snapshot total; matched comparisons equal the individual usage sum.

Comment 10, line 74

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:74

// Detection and scanning use the same registry to select checked versions.

Comment 11, line 94

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:94

// Compare values with independent jq calculations in _expectations.md.

Comment 12, line 122

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:122

// Checked versions carry no fallback marker; preserve the original version string.

Comment 13, line 133

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:133

// Query reasoning directly from storage for the expected subset sum.

Comment 14, line 149

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:149

// Matched reconciliation: individual usage sum equals the final snapshot.

Comment 15, line 162

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:162

// Repeat scanning adds no events.

Comment 16, line 174

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:174

// Older 0.139-0.151 files lack token_usage_record; individual usage comes from

Comment 17, line 175

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:175

// token_count.last_token_usage under the cumulative-total change rules.

Comment 18, line 176

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:176

// Expectations use independent JavaScript calculations and manual review.

Comment 19, line 190

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:190

/// Snapshot comparison differences produce diagnostics without degrading source_files.status.

Comment 20, line 191

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:191

/// Mismatches remain active; actual individual-usage parse errors degrade the file.

Comment 21, line 244

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:244

// Snapshot mismatch is diagnostic reconciliation only; source health stays active.

Comment 22, line 256

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:256

// Registered older versions use KnownVersion in detection and scanning.

Comment 23, line 274

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:274

// Calls = regular + carried; 0.146.0-alpha.3 includes one carried compaction response.

Comment 24, line 277

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:277

// Compare values with independent JavaScript calculations in _expectations.md.

Comment 25, line 305

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:305

// Preserve known_version, rollout_legacy parser, and sequence identity without response_id.

Comment 26, line 323

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:323

// Sum reasoning subsets; the carried response contributes a known zero in this sample.

Comment 27, line 333

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:333

// Model comes from turn_context; subagent category from parent_thread_id/source.subagent.

Comment 28, line 350

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:350

// Compare regular+carried sums against the final snapshot plus carried usage.

Comment 29, line 374

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:374

// Source regressions and carried responses retain visible diagnostics.

Comment 30, line 396

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:396

// Repeat scanning adds no events.

Comment 31, line 408

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:408

// Unregistered older version 0.141.0 remains a LatestFallback compatibility attempt.

Comment 32, line 409

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:409

// rollout_v1 finds no individual usage records in this older shape, so imports zero events.

Comment 33, line 410

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:410

// Nonempty snapshots differ from zero details and mark the attempted format incompatible.

Comment 34, line 411

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:411

// Preserve diagnostics without fabricating usage or advancing the cursor.

Comment 35, line 439

desktop/src-tauri/crates/core/tests/codex_versions_v17.rs:439

// Incompatible attempts commit no events, cursors, or aggregates.