Mirror the worker's LRU-eviction ACTIVE_RUNS liveness check in
_evict_session_agent so truncate/clear/model-switch can't close the SessionDB
out from under an in-flight turn on the same session (#5096 Bug D follow-up).
The cache handle still drops (harmless — worker holds a local ref); only the
lifecycle commit + _session_db.close() are skipped while a run is active.
Co-authored-by: claw-io <claw-io@users.noreply.github.com>
The streaming path already honored a max_tokens cap from the active profile's
config.yaml, but there was no UI to manage it. Settings > Preferences now has a
"Max output tokens" field that writes a root-level max_tokens override through
the profile config (not settings.json, which the streaming worker never reads
for this), preserving unrelated YAML keys under _cfg_lock. Blank clears the
override so agent.max_tokens fallback resumes; get_max_tokens_status() surfaces
root / effective / fallback so the field shows what a new turn would use.
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
Two gate-found CORE/SILENT edges fixed on-branch:
- api/config.py: switching the main default to a different custom:* provider kept
the OLD provider's base_url, routing New Chat to the wrong endpoint. The
base_url-clear on provider-change now covers custom:* too (was exempted).
- static/sessions.js: newSession() skipped the configured default when S.session
was null but the picker held stale state (e.g. after deleting the last session),
inheriting the stale model. Default now wins over legacy picker regardless of
loaded-session state (explicit empty-composer override still takes priority).
- Updated 2 stale source-grep precedence assertions + widened a brittle 6000-char
newSession() slice window to 7000 (assignment block grew; intent unchanged).
- 2 new non-vacuous regression tests (custom base_url clear; precedence).
Co-authored-by: Rod Boev <rod.boev@gmail.com>
The JSON/YAML tree viewer (#484) hardcoded its default: valid json/yaml
fenced blocks opened in Tree view at 10+ lines (`const showTree=lineCount>=10;`).
Make that default user-configurable while preserving current behavior.
- New settings (persisted via /api/settings -> settings.json):
- structured_code_default_view: auto | on | off (default auto)
- structured_code_auto_tree_lines: int 1..1000 (default 10)
- ui.js: replace the hardcode with a sanitized, unit-testable decision
helper _structuredCodeShowTree(mode, threshold, lineCount); Tree/Raw
toggle, Raw <pre><code>, parse-failure fallback, and YAML lazy-load
are unchanged.
- Appearance settings: 'JSON/YAML code blocks' select + 'Auto threshold
lines' input (disabled outside auto mode); wired through the existing
appearance-autosave and explicit-save paths; live re-render on change.
- boot.js seeds window globals; i18n labels added to all 13 locales.
- Default (auto + 10) reproduces the original >=10 behavior exactly.
Two greptile findings on #4741:
- api/routes.py: removed the redundant 'from api.config import load_settings'
inside _session_list_payload_to_response — load_settings is already imported at
module scope, and this function only runs at request time (after module load).
- api/config.py: _load_yaml_config_file_raw now deep-copies the cached raw dict by
default so a caller can never mutate the shared _yaml_file_cache entry. The
internal _load_yaml_config_file passes _copy=False (it immediately runs
_expand_env_vars, which returns a fresh structure and never mutates its input)
so the /api/reasoning hot path keeps its cheap no-copy fast path; reload_config
uses the safe default copy.
The 'and loaded' guard skipped the _cfg_mtime stamp for a config that parses to
an empty {} dict, so get_config()'s current_mtime != _cfg_mtime stale check fired
on every call and re-entered reload_config() under _cfg_lock forever — a hot-path
spin for a freshly created/reset profile's {} config. Split the guard: the no-op
_cfg_cache.update is still skipped for {}, but _cfg_mtime is always stamped when
the file parsed to a dict, matching master's pre-#4662 behavior. Incidentally
fixes the pre-existing empty/None-config spin. Adds a regression test.
Phase 2 of #4662. reload_config() did a raw yaml.safe_load of config.yaml on
every call (~125ms+ on a large file), including on the profile-switch hot path
(~446ms load_settings stage in /api/sessions diagnostics). It now parses through
the mtime+size-keyed cache added in #4652, via a shared _load_yaml_config_file_raw
core (DRY: _load_yaml_config_file = raw + env-expand). reload_config still runs
its OWN env expansion pinned to the unscoped process-env view (block_process_env_fallback
reset + _thread_ctx.env={}) so the #798 TLS profile-isolation invariant is byte-for-byte
preserved; only the disk parse is now cached. An unchanged config.yaml is no longer
reparsed; an mtime change still busts the cache.
Opus final gate reproduced two issues in the config-expansion hardening:
MUST-FIX — import-order forward reference. _expand_env_vars (config.py top) calls
_thread_local_env_value, but that fn + _thread_ctx were defined ~6900 lines later,
AFTER the import-time reload_config(). A ${VAR} in a startup config.yaml raised a
(swallowed) NameError, leaving import-time config consumers with empty config.
Fix: relocate _thread_ctx + _thread_local_env_value above the config-file section
(they depend only on os + threading). Verified: startup ${VAR} now expands, no
NameError.
SHOULD-FIX — reload_config baked block-scoped expansion into the process-global
_cfg_cache. If a reload fired while a readonly/worker scope was active (profile
alternation resolves _get_config_path to the named profile), under-expanded
literal ${VAR}s were written to the shared cache, starving concurrent readers of
the `cfg` alias. Fix: reload_config now expands the global cache with the
block flag forced off (process-env view), since the cache is process-global;
per-request scoped reads still go through _thread_local_env_value at their own
sites. Verified: global cache holds the process-expanded value, not the literal.
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
Codex re-gate found a config-template leak vector distinct from the env-name
scrub: _expand_env_vars() expanded ${VAR} references in config values via raw
os.environ, so a profile config.yaml of e.g. `api_key: ${ANTHROPIC_TOKEN}` let
_get_provider_api_key reconstruct the server-process token and pass it through as
the API key argument — even when the child-env scrub stripped the env var. This
bypassed the thread-local channel entirely.
_expand_env_vars now substitutes via _thread_local_env_value (the same
thread-local-first accessor with the block_process_env_fallback guard the PR
introduced): a ${VAR} resolves to the profile's own value, and refuses the
process-env fallback when a profile-scoped readonly/background scope set the block
flag (leaving the reference unexpanded). Cache integrity preserved — only the RAW
unexpanded dict is cached; expansion re-runs per read against the current thread's
scope. Normal (no-scope) behavior unchanged. + regression that fails without the fix.
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
Two-layer perf fix for the /api/reasoning storm dropping streaming to ~2 tok/s:
- ui.js: syncReasoningChip() serves the cached chip on routine topbar syncs, only
re-fetching when the model/provider key (_lastReasoningFetchKey) changes — restores
pre-a9ce2889 behavior while keeping its model-switch-refresh intent.
- config.py: memoize _load_yaml_config_file keyed on (path, st_mtime_ns, st_size) under
a lock; cache the RAW parsed dict and re-run cheap _expand_env_vars per call (cache stays
correct vs os.environ and read-modify-save callers get a fresh structure; on-disk edits
picked up via mtime/size). Code byte-faithful from PR head f868f10a1ad7. 9 own tests pass.
Co-authored-by: nesquena-hermes <agent@nesquena-hermes>
rodboev. Default-off; OFF path byte-identical to master (cache key 'u' + _renderUserFencedBlocks
unchanged when window._renderUserMarkdown falsy). ON routes user text through the same renderMd
SAFE_TAGS/_isSafeUrl sanitizer as assistant msgs. Nathan override-approved (likes the idea, opt-in).
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
Converged both bounce items: enrich-existing-only (continue on webui_list-is-None,
no new providers -> routing side-effect gone, routes.py no longer touched) + respects
per-provider ID prefix convention (@nous:). Co-authored-by: kaishi00 <kaishi00@users.noreply.github.com>
Codex ship-gate (SILENT): both disk loaders dropped `aliases` (the save path
never persisted them), so a disk-cache hit — strict cold-path OR the new stale
fallback — served /api/models with empty aliases, silently breaking `/model
<alias>` slash-command resolution (static/commands.js resolves slash aliases
only from /api/models.aliases). Add _model_aliases_from_config() helper and
reconstruct aliases from current config in BOTH loaders (fixes the pre-existing
strict-loader gap too). Adds a regression test.
Co-authored-by: starship-s <starship-s@users.noreply.github.com>
Applies Opus ship-gate findings: (1) the autouse fixture now points
_get_models_cache_path at an isolated temp file so the over-budget static-fallback
test can't read a real ~/.../models_cache.json and become an order-dependent flake;
(2) _load_stale_models_cache_from_disk rejects a cross-schema cache (its groups/badge
shape may be incompatible with the current picker). Adds a cross-schema-rejection test.
Co-authored-by: starship-s <starship-s@users.noreply.github.com>
Rod Boev